1 INFO-VAX	Thu, 11 Jul 2002	Volume 2002 : Issue 378       Contents:# Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study # Re: 2002 Worldwide HP OpenVMS Study % Re: Alpha Shadowing Phase II question % Re: Alpha Shadowing Phase II question 1 Re: Another mention of VMS is an HP press release ) Re: Are there any humans in COMPAQ/HP US? 8 Re: Bios Password reset for DEC3000/400 and/or VS4000/90' Re: C and X11 programming, finding info ' Re: C and X11 programming, finding info ' Re: C and X11 programming, finding info  Re: Convert conundrum. Help? Re: Convert conundrum. Help? Re: Convert conundrum. Help? Re: Convert conundrum. Help? Re: Convert conundrum. Help?) Re: CPU time for batch job 4-MAY-1859 ???  Re: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down RE: DS10 shutting down Re: DS10 shutting down Re: DS10 shutting down Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion RE: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: Looking for your opinion Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh... Re: McKinley Cometh...+ Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + RE: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts + RE: McKinley tops SpecFP AND SpecInt charts + Re: McKinley tops SpecFP AND SpecInt charts 
 Mousewheel+ Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + RE: Only 20% drop in VMS systems (was: wow) + RE: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) + Re: Only 20% drop in VMS systems (was: wow) * OpenSSL and certificates concept questions. Re: OpenSSL and certificates concept questions. Re: OpenSSL and certificates concept questions. Re: OpenSSL and certificates concept questions OpenVMS Ambassadors  Re: OpenVMS Ambassadors  OpenVMS Polls are back!  Re: PURGE   version set to ;1  Re: PURGE   version set to ;1  Re: PURGE   version set to ;1  Re: Quick Poll - two sessions  Re: Quorum discussion/questions  Re: RECALL suggestion  Re: RECALL suggestion  Re: RECALL suggestion  Re: RECALL suggestion  Re: RECALL suggestion  Re: RRD42 Caddy   Re: SET FILE/TRUNCATE equivalent Re: Setting Term Re: Setting Term Re: SMTP 8bit hack not working Re: Stupid storage question 8 Re: System Disk Initialization Parameter recommendations8 Re: System Disk Initialization Parameter recommendations Re: TIMA Re: TIMA Re: Used Alpha's and Vax's Re: Used Alpha's and Vax's Re: Used Alpha's and Vax's/ Re: Using GNU C on OpenVMS FAQ (Looking for it)  Re: VS 3100/38: Boot message Re: VS 3100/38: Boot message+ What does UCX 6.2 SMTP intermittently fail? / Re: What does UCX 6.2 SMTP intermittently fail?  Re: Where to put startup stuff Re: Where to put startup stuff Re: Where to put startup stuff Which thread am I in ? Re: Who Knows Asterix ?  Re: Who Knows Asterix ?   F ----------------------------------------------------------------------  # Date: Wed, 10 Jul 2002 21:50:18 GMT # From: "John Smith" <a@nonymous.com> , Subject: Re: 2002 Worldwide HP OpenVMS StudyG Message-ID: <u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com>   7 "Kenneth Farmer" <kfarmer@openvms.org> wrote in message 6 news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com...H > A Rich Marcello email sent to selected OpenVMS customers anouncing the "2002 ; > Worldwide HP OpenVMS Study" and asking for participation.  > ; > http://www.openvms.org/stories.php?story=02/07/10/1060371  >   I Does this study mean that if enough customers don't plan to transition to F Itanic, and will be moving to IBM or Sun instead, that they'd consider reviving Alpha?    ------------------------------  # Date: Wed, 10 Jul 2002 21:58:58 GMT 1 From: "Terry C. Shannon" <terryshannon@attbi.com> , Subject: Re: 2002 Worldwide HP OpenVMS Study. Message-ID: <C62X8.460428$352.73697@sccrnsc02>  . "John Smith" <a@nonymous.com> wrote in messageA news:u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com...  > 9 > "Kenneth Farmer" <kfarmer@openvms.org> wrote in message 8 > news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com...J > > A Rich Marcello email sent to selected OpenVMS customers anouncing the > "2002 = > > Worldwide HP OpenVMS Study" and asking for participation.  > > = > > http://www.openvms.org/stories.php?story=02/07/10/1060371  > >  > K > Does this study mean that if enough customers don't plan to transition to H > Itanic, and will be moving to IBM or Sun instead, that they'd consider > reviving Alpha?  > J That's a question only HPQ is properly equipped to answer, but I'd say theJ odds are slim to none that the post-EV7 Alpha effort will be reinitialized" 13 months after it was terminated.   ------------------------------  # Date: Thu, 11 Jul 2002 00:52:08 GMT # From: "John Smith" <a@nonymous.com> , Subject: Re: 2002 Worldwide HP OpenVMS StudyG Message-ID: <YE4X8.7319$6DW1.3447@news01.bloor.is.net.cable.rogers.com>   < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:C62X8.460428$352.73697@sccrnsc02... > 0 > "John Smith" <a@nonymous.com> wrote in messageC > news:u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com...  > > ; > > "Kenneth Farmer" <kfarmer@openvms.org> wrote in message : > > news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com...L > > > A Rich Marcello email sent to selected OpenVMS customers anouncing the	 > > "2002 ? > > > Worldwide HP OpenVMS Study" and asking for participation.  > > > ? > > > http://www.openvms.org/stories.php?story=02/07/10/1060371  > > >  > > J > > Does this study mean that if enough customers don't plan to transition toJ > > Itanic, and will be moving to IBM or Sun instead, that they'd consider > > reviving Alpha?  > > L > That's a question only HPQ is properly equipped to answer, but I'd say theL > odds are slim to none that the post-EV7 Alpha effort will be reinitialized$ > 13 months after it was terminated.     If HP really listens.........    ------------------------------  % Date: Wed, 10 Jul 2002 21:55:44 -0400 2 From: rdeininger@mindspring.com (Robert Deininger), Subject: Re: 2002 Worldwide HP OpenVMS StudyJ Message-ID: <rdeininger-1007022155440001@1cust58.tnt3.nashua.nh.da.uu.net>  G In article <u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com>, $ "John Smith" <a@nonymous.com> wrote:  8 >"Kenneth Farmer" <kfarmer@openvms.org> wrote in message7 >news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com... I >> A Rich Marcello email sent to selected OpenVMS customers anouncing the  >"2002< >> Worldwide HP OpenVMS Study" and asking for participation. >>< >> http://www.openvms.org/stories.php?story=02/07/10/1060371 >> > J >Does this study mean that if enough customers don't plan to transition toG >Itanic, and will be moving to IBM or Sun instead, that they'd consider  >reviving Alpha?  G If Compaq couldn't afford to continue Alpha development, what makes you I think HP could afford to restart it?  Have microprocessor designers taken J a pay cut?  Is IBM charging less to fab chips?  What's the new factor thatD radically changes the economics of Alpha development and production?   ------------------------------  # Date: Thu, 11 Jul 2002 02:15:31 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> , Subject: Re: 2002 Worldwide HP OpenVMS Study' Message-ID: <3D2CEE88.C8C142AA@fsi.net>    "Terry C. Shannon" wrote:  > 0 > "John Smith" <a@nonymous.com> wrote in messageC > news:u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com...  > > ; > > "Kenneth Farmer" <kfarmer@openvms.org> wrote in message : > > news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com...L > > > A Rich Marcello email sent to selected OpenVMS customers anouncing the	 > > "2002 ? > > > Worldwide HP OpenVMS Study" and asking for participation.  > > > ? > > > http://www.openvms.org/stories.php?story=02/07/10/1060371  > > >  > > M > > Does this study mean that if enough customers don't plan to transition to J > > Itanic, and will be moving to IBM or Sun instead, that they'd consider > > reviving Alpha?  > > L > That's a question only HPQ is properly equipped to answer, but I'd say theL > odds are slim to none that the post-EV7 Alpha effort will be reinitialized$ > 13 months after it was terminated.  E In driver's ed. many years ago, they emphasized safety in the name of G "defensive driving" and stated it this way: you may be right, but don't  DEAD right!   H Not sure why that comes to mind just now, but let's face it: there are aD lot of folks who are dead, out of business or whatever, but at leastF they didn't have admit a mistake and back-pdeal to get back on track -E they gave up and died, went bankrupt, folded the company or whatever, * all the time knowing they were 100% right.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Wed, 10 Jul 2002 22:34:44 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> , Subject: Re: 2002 Worldwide HP OpenVMS Study, Message-ID: <3D2CEEA0.539CF480@videotron.ca>   Robert Deininger wrote: I > If Compaq couldn't afford to continue Alpha development, what makes you K > think HP could afford to restart it?  Have microprocessor designers taken L > a pay cut?  Is IBM charging less to fab chips?  What's the new factor thatF > radically changes the economics of Alpha development and production?  N Since Compaq did not release audited and honest numbers to show that Alpha wasN really losing money, it is wrong to blindly believe the Compaq propaganda that/ claims they could not afford to continue Alpha.   J Compaq and HP decided to help Intel at any/all costs and blindly jump intoI IA64's bandwagon. They then found any and all possible excuses to justify  their decisions.   ------------------------------  # Date: Thu, 11 Jul 2002 02:42:14 GMT 1 From: "Terry C. Shannon" <terryshannon@attbi.com> , Subject: Re: 2002 Worldwide HP OpenVMS Study. Message-ID: <ag6X8.145261$Uu2.33296@sccrnsc03>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3D2CEE88.C8C142AA@fsi.net...  > "Terry C. Shannon" wrote:  > > 2 > > "John Smith" <a@nonymous.com> wrote in messageE > > news:u_1X8.12195$wLk.7632@news01.bloor.is.net.cable.rogers.com...  > > > = > > > "Kenneth Farmer" <kfarmer@openvms.org> wrote in message < > > > news:p2_W8.4431$Sb3.137375@twister.southeast.rr.com...J > > > > A Rich Marcello email sent to selected OpenVMS customers anouncing the  > > > "2002 A > > > > Worldwide HP OpenVMS Study" and asking for participation.  > > > > A > > > > http://www.openvms.org/stories.php?story=02/07/10/1060371  > > > >  > > > L > > > Does this study mean that if enough customers don't plan to transition toL > > > Itanic, and will be moving to IBM or Sun instead, that they'd consider > > > reviving Alpha?  > > > J > > That's a question only HPQ is properly equipped to answer, but I'd say the @ > > odds are slim to none that the post-EV7 Alpha effort will be
 reinitialized & > > 13 months after it was terminated. > G > In driver's ed. many years ago, they emphasized safety in the name of I > "defensive driving" and stated it this way: you may be right, but don't 
 > DEAD right!  > J > Not sure why that comes to mind just now, but let's face it: there are aF > lot of folks who are dead, out of business or whatever, but at leastH > they didn't have admit a mistake and back-pdeal to get back on track -G > they gave up and died, went bankrupt, folded the company or whatever, , > all the time knowing they were 100% right. >   G Yep. But here's a question: having cast all the EV8 developers into the B not-so-tender embrace of L'Intella, and having lost over a year ofI development, is there any way in which HPQ could re-start the effort now?    ------------------------------  % Date: Wed, 10 Jul 2002 23:08:32 +0100 ) From: Witchy <news@sruasonidyranib.co.uk> . Subject: Re: Alpha Shadowing Phase II question8 Message-ID: <m0cpiuojtmejp56mvr7o51qci09m0mef4g@4ax.com>  8 On 9 Jul 02 13:15:32 PST, mckinneyj@cpva.saic.com wrote:  - >In article <3D2B4283.265BC8A1@videotron.ca>, 1 > JF Mezei <jfmezei.spamnot@videotron.ca> writes:  >> Witchy wrote:C >>> What I'm wondering is is there a way of creating the shadowsets 2 >>> without having to go thru a full shadow copy ? >>  M >> There was a discussion of this a few weeks ago. You may reduce the time to M >> bring the shadow set up to speed by using a /PHYSICAL backup, but then you Q >> need to have the source disk totally off-line with nobody using it (eg: system  >> down time). >>  Q >> If you can afford such system downtime, then the normal full shadow copy won't # >> really be that different to you.  > L >No need to have exclusive access to the source disk; just insure that afterP >the /physical copy is made that you intialize (or use some otherwise convenientL >technique to destroy the SCB) the /physical copy disk so that the shadowingO >software can determine which disk is to be the master. The initialization will M >only rewrite the reserved files, one of which is BITMAP.SYS, which I believe M >contains the SCB (storage control block containing description of the shadow L >set). When the initialized /physical copy is targeted by the shadow copy itL >will find that nearly all of the disk is already the same as the master and( >skip the copy portion of the operation.  D I didn't have to initialise the target drive, all I did was dismountB the 2nd member of the shadow set, mount it foreign with /ov=shadowE then do a backup/physical. When I remounted the drive into the shadowT  set it worked a treat....cheers!   ------------------------------   Date: 11 Jul 02 01:06:07 +0200) From: p_sture@elias.decus.ch (Paul Sture)r. Subject: Re: Alpha Shadowing Phase II question) Message-ID: <wPpTsQcaE86t@elias.decus.ch>M  d In article <m0cpiuojtmejp56mvr7o51qci09m0mef4g@4ax.com>, Witchy <news@sruasonidyranib.co.uk> writes:: > On 9 Jul 02 13:15:32 PST, mckinneyj@cpva.saic.com wrote: > . >>In article <3D2B4283.265BC8A1@videotron.ca>,2 >> JF Mezei <jfmezei.spamnot@videotron.ca> writes: >>> Witchy wrote:oD >>>> What I'm wondering is is there a way of creating the shadowsets3 >>>> without having to go thru a full shadow copy ?n >>> N >>> There was a discussion of this a few weeks ago. You may reduce the time toN >>> bring the shadow set up to speed by using a /PHYSICAL backup, but then youR >>> need to have the source disk totally off-line with nobody using it (eg: system >>> down time).n >>> R >>> If you can afford such system downtime, then the normal full shadow copy won't$ >>> really be that different to you. >>M >>No need to have exclusive access to the source disk; just insure that after Q >>the /physical copy is made that you intialize (or use some otherwise convenientoM >>technique to destroy the SCB) the /physical copy disk so that the shadowingiP >>software can determine which disk is to be the master. The initialization willN >>only rewrite the reserved files, one of which is BITMAP.SYS, which I believeN >>contains the SCB (storage control block containing description of the shadowM >>set). When the initialized /physical copy is targeted by the shadow copy itoM >>will find that nearly all of the disk is already the same as the master andk) >>skip the copy portion of the operation.o > F > I didn't have to initialise the target drive, all I did was dismountD > the 2nd member of the shadow set, mount it foreign with /ov=shadowG > then do a backup/physical. When I remounted the drive into the shadowC" > set it worked a treat....cheers!  ) Good tip - that destroys the SCB as well.   K OTOH, if you are running V7.3, have a  HELP INIT/SHADOW says the following:i  
 INITIALIZE  	   /SHADOWy  C         /SHADOW=(device_name_1, device_name_2, device_name_3) labely  F      Initializes multiple members of a future shadow set. InitializingF      multiple members in this way eliminates the requirement of a full-      copy when you later create a shadow set.C  B      The INITIALIZE command with the /SHADOW and /ERASE qualifiers'      performs the following operations:c  E      o  Formats up to six devices with one command, so that any threeiF         can be subsequently mounted together as members of a new host-         based shadow set  %      o  Writes a label on each volumeS  E     o  Deletes all information from the devices except for the systemyB         files containing identical file structure information. All.         former contents of the disks are lost.  @      You can then mount up to three of the devices that you haveB      initialized in this way as members of a new host-based shadow	      set.   E      Compaq strongly recommends that you use the /ERASE qualifier. ByDD      using the /ERASE qualifier, no merge operation is required when6      you create the shadow set with the MOUNT command.  B      All the devices must have identical, nonzero values for Total@      Blocks. You can view the Total Blocks value by entering theA      SHOW DEVICE/FULL command. If a device has never been mountedtD      or initialized on this system, the SHOW DEVICE/FULL command forE      the device will not display a value for Total Blocks. To correcta?      this condition, either mount and then dismount the device,tB      or initialize the device. The Total Blocks value will then be#      displayed by SHOW DEVICE/FULL.D  B      Note that the INITIALIZE/SHADOW command should not be used toC      initialize a disk to be added to an existing shadow set, as no       benefit is gained.   @      Additional information can be found in Volume Shadowing for
      OpenVMS.a    w    e __
 Paul Sture Switzerlands   ------------------------------  % Date: Wed, 10 Jul 2002 14:32:49 -0400o- From: JF Mezei <jfmezei.spamnot@videotron.ca> : Subject: Re: Another mention of VMS is an HP press release, Message-ID: <3D2C7DCC.999BDE08@videotron.ca>   Ken Robinson wrote:n > I > See <http://biz.yahoo.com/bw/020710/100128_1.html> entitled "HP ExpandssG > Industry's Broadest Range of Products and Solutions for Oracle9i Realc > Application Clusters"p  M What I noticed is the prominence of Alpha ES45 in that press release, as well J as the mention of Tru64. To me, that was far more significant than the oneN time mention of VMS, especially when you consider that Tru64 is more dead than VMS is at this point in time.s  J What confused me however is the later part of the press release that talksG about the same "clustered" database engine running on Linux and WindowsmL "clusters". If Oracle 9i runs on crap like Windows, what is the advantage of running Oracle 9i on Tru64 ?  N What would have been interesting to see is a mention of the features available# on Tru64  but not on Windows/Linux.A   ------------------------------  # Date: Wed, 10 Jul 2002 19:18:15 GMT ( From: "konabear" <maurert@ameritech.net>2 Subject: Re: Are there any humans in COMPAQ/HP US?@ Message-ID: <XL%W8.12125$t%5.5750354@newssvr28.news.prodigy.com>  L While I understand your frustration, I would not recommend using this non HP/ forum for what is clearly an official HP issue.a  K That said I pulled the case and looked at the notes.  It appears that thereeK was some kind of contact with the customer contact today.  (fifteen minutes-K ago?)  Also looks as if an ETA of delivery of parts was given.  I'd suggestmK that you check with the contact in the USA and see if this is true.  If notkK then have your person on the ground in the USA call 800-354-9000, referencen: the Case ID above and ask to speak to the Manager on Duty.   Todd Maurer.  6 "Andy Proctor" <aproctor@hotmail.com> wrote in message/ news:1026313754.743639@ananke.eclipse.net.uk...e5 > Not strictly a VMS query but this is a last resort./ >wJ > If there is anyone from any service centre in the US reading this please can L > you help, or if there is anyone from COMPAQ anywhere in the US, please can > you help?i >eF > A customer has a DS10 workstation under warranty and no-one from the ServiceaK > side in the US is actioning a service call. The monitor is defective. Thet" > ticket number is NA020701003976. >PJ > This is now almost 2 weeks old and the warranty terms state the response isL > NBD. My colleagues in our US office are unable to get a human on the phone > to obtain an update..why?  >nJ > COMPAQ look like they are breaching their own t/c of purchase here.....I' > would like to get this serviced asap.  >hH > I have COMPAQ UK helping me now also, as they are friendly helpful andA > efficient, but have no control over what the US operation does.  > * > Please help someone...anyone....{buler!] >/ > Andy Proctor > Customer Support Manager > Spirent Communications > Aspen Wayn > Devon, UKn > Tel: +44-1803-546315 > andy.proctor@spirentcom.comd > http://www.spirentcom.comi >g >p   ------------------------------   Date: 10 Jul 2002 23:51:06 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)A Subject: Re: Bios Password reset for DEC3000/400 and/or VS4000/90r* Message-ID: <agih9a$ntl$1@web1.cup.hp.com>  O In article <afvq56$al5$1@bob.news.rcn.net>, "me" <wicklinedd@erols.com> writes:cE :I have set up to VMS workstation on the above machine.  I everything4I :installed and the BIOS password is not set right now.  But I don't know n@ :the old password in order to change and enable it.  Anybody????  ?   I would generally discourage the use of the console password.u  H   On the DEC 3000 model 400 series, there is no password set by default.  J   To enable, you need to pick a sixteen digit hexadecimal value and enter I   it identically at each of the two SET PSWD prompts and you must SETENV eH   SECURE ON, and you need to move a jumper.  (Getting at the jumper willJ   involve access to the motherboard -- with the power supply to the left, I   the jumper is nearest the rightmost TURBOchannel slot.  It is labeled. eG   With the jumper on left and center posts, the password is disabled.   :   With the jumper on the center and right posts, enabled.)  7   Again, I would discourage using the console password.   G   Check the console help for the VAXstation 4000 model 60, the commandstF   should be similar.  (SET PSE 1 or 0, and SET PSWD, or some such.  ToH   get into the privileged command set once the password is enabled, use F   the console command LOGIN.)  AFAIK, also sixteen hexadecimal digits.'   Again, no password is set by default.R  @   But again, I would discourage the use of the console password.  E   If you want to learn how to reset the password and clear out all of.E   the settings, I think that's in the hardware service documentation.c    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 10 Jul 2002 16:47:13 -0400d; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>M0 Subject: Re: C and X11 programming, finding info" Message-ID: <3d2c9d6f@news.si.com>  4 >I don't have the book in front of me, but if memory >serves, the& >said it was "font.dir" or "fonts.dir" >r= >Anyway, where can I find such a list, if there is one, on mys >ever-so-wonderh- >alpha OpenVMS v7.2-1, and v7.3 workstations?v  - $ dir decw$xterminal_font:*directory*,*alias*-   will find them for you.u -- 2A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.com.= 3290 Patterson Ave. SE, MS      Addresses modified to preventm< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Wed, 10 Jul 2002 18:14:58 -04003; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>l0 Subject: Re: C and X11 programming, finding info$ Message-ID: <3d2cb200$1@news.si.com>  H >There is a logical name DECW$FONT in the server table that is a list of >directories to find fonts in.   Really?    $ sho log/table=* decw$fontt9 %SHOW-S-NOTRAN, no translation for logical name DECW$FONTt   Which version of Motif?  -- gA Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.comb= 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Wed, 10 Jul 2002 19:49:12 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>y0 Subject: Re: C and X11 programming, finding info+ Message-ID: <3D2CC7DE.84C4EEC@videotron.ca>    Brian Tillman wrote: > $ sho log/table=* decw$fontm; > %SHOW-S-NOTRAN, no translation for logical name DECW$FONT   L Same here. But I am on VAX, so I have learned to accept that even though theG version shows 7.2, in fact, it is probably closed to 6.1 due to all thet missing features./  / (same with FONT/DIR which doesn't exist on VAX)/   ------------------------------  % Date: Wed, 10 Jul 2002 14:37:18 -04001- From: JF Mezei <jfmezei.spamnot@videotron.ca>e% Subject: Re: Convert conundrum. Help?:, Message-ID: <3D2C7ED9.CAF83706@videotron.ca>   Bill Clark wrote:dK > (00:41)$  CONVERT/STAT/FDL=PD_FDL:PDP_L_CCSUM.FDL/nosort/fast/work=3    -uK >               DISK$SMSUSER0:[CONVERT]PDICCSUM.OLD                       - @ >                 DISK$SMSPRDDAT01:[SMSPRD.PDS.DATA]PDICCSUM.DAT( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memorya  L You need to increase your process (in UAF/Authorize) Page file quote PGFLQUOE to give the process greater access to virtual memory.  Note that thisnL reesource is limited by the size of your page/swap files on the system (SHOW= MEMORY/FILES will give you an idea of how much is available).s   ------------------------------  % Date: Wed, 10 Jul 2002 15:10:11 -0400s From: norm.raphael@metso.com% Subject: Re: Convert conundrum. Help?-? Message-ID: <OFF68B91C9.22E2986D-ON85256BF2.0068C5D2@metso.com>h   Yes, but...aF (Were we not just speaking of "Technical Journals" in another thread?)  F 1)  I always puzzle over a "-W-" error generated by a "-F-" error in a
 subsystem.K If LIB is fatal, how can SORT be only a warning (something died, but if youo leave 8 the area quickly enough, maybe you won't be dead, too?))  E 2)  Just what was the consequence of "insufficient virtual memory" in H this case (Is it safe to ignore, because the sort only "W" arned, and if finished?)?3  0 3)  Oh, and with what Status did CONVERT finish?  K There are more questions here than answers, regardless of QUOTAS and SYSGEN2 parameters.    -Norm         A JF Mezei <jfmezei.spamnot@videotron.ca> on 07/10/2002 02:37:18 PMw  9 Please respond to JF Mezei <jfmezei.spamnot@videotron.ca>    To:    Info-VAX@Mvb.Saic.Com cc:a( Subject:    Re: Convert conundrum. Help?     Bill Clark wrote: K > (00:41)$  CONVERT/STAT/FDL=PD_FDL:PDP_L_CCSUM.FDL/nosort/fast/work=3    -gK >               DISK$SMSUSER0:[CONVERT]PDICCSUM.OLD                       -e@ >                 DISK$SMSPRDDAT01:[SMSPRD.PDS.DATA]PDICCSUM.DAT( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memorye  D You need to increase your process (in UAF/Authorize) Page file quote PGFLQUOiE to give the process greater access to virtual memory.  Note that this F reesource is limited by the size of your page/swap files on the system (SHOWa= MEMORY/FILES will give you an idea of how much is available).    ------------------------------  % Date: Wed, 10 Jul 2002 21:50:18 +0200t9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>c% Subject: Re: Convert conundrum. Help?o' Message-ID: <3D2C8FFA.3914A8BE@aaa.com>t  = I'd say that SORT got an FATAL error from LIB that SORT had a = workaround for, so SORT just issued a WARNING. If SORT hadn't > had any workaround, I'd expect a FATAL from SORT also and that the whole process had failed.-   Jan-Erik Sderholm.e       norm.raphael@metso.com wrote:r > 
 > Yes, but...uH > (Were we not just speaking of "Technical Journals" in another thread?) > H > 1)  I always puzzle over a "-W-" error generated by a "-F-" error in a > subsystem.M > If LIB is fatal, how can SORT be only a warning (something died, but if youz > leavee: > the area quickly enough, maybe you won't be dead, too?)) >m   ------------------------------  % Date: Wed, 10 Jul 2002 18:46:53 -0400g- From: JF Mezei <jfmezei.spamnot@videotron.ca>e% Subject: Re: Convert conundrum. Help?g, Message-ID: <3D2CB948.59784A16@videotron.ca>   norm.raphael@metso.com wrote:/H > 1)  I always puzzle over a "-W-" error generated by a "-F-" error in a > subsystem.M > If LIB is fatal, how can SORT be only a warning (something died, but if yout > leavem: > the area quickly enough, maybe you won't be dead, too?))   How about "y  F Hi ! My name is SORT, I tried to get more memory to make the sort moreK efficient, but I couldn't, so I reverted to another, less efficient method,e which requires less memory"t   or  M Hi ! My name is CONVERT, my friend SORT told me a few times that LIB couldn'tnJ give him more memory, so I decided to give up on SORT and proceed with theM CONVERT as if it were a /NOSORT. The output file may be "correct", but it mayCL not be the most efficient since records may have been inserted out of order.    - (The above two message are mere speculation).d   ------------------------------  % Date: Wed, 10 Jul 2002 22:21:47 -0700h" From: Koloth <koloth@telocity.com>% Subject: Re: Convert conundrum. Help?i, Message-ID: <3D2D15EA.CF3E83AF@telocity.com>  ) Make sure WSEXTENT is less than PGFLQUOTA    Bill Clark wrote:h   > greetings @ >     while 'CONVERT'ing a file i get 'SORT' error messages; theD >     'CONVERT' seems to work, the file is fine when all is said andF >     done ('ANALYZE/RMS' returns no errors. Details below. Any ideas?) >         (File Size:  3,458,000 blocks.)iF >         ( Sortwork0, sortwork1... etc disks have ample room (or so i > thought))c > ............................
 > (00:41)$K > (00:41)$  CONVERT/STAT/FDL=PD_FDL:PDP_L_CCSUM.FDL/nosort/fast/work=3    -tK >               DISK$SMSUSER0:[CONVERT]PDICCSUM.OLD                       -s@ >                 DISK$SMSPRDDAT01:[SMSPRD.PDS.DATA]PDICCSUM.DAT( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memory ( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memorye( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memory>( > %SORT-W-SYSERROR, system service error/ > -LIB-F-INSVIRMEM, insufficient virtual memoryt >     CONVERT Statistics& > Number of Files Processed:         1J > Total Records Processed:     6196149    Buffered I/O Count:          143J > Total Exception Records:           0    Direct I/O Count:         634646J > Total Valid Records:         6196149    Page Faults:               24371J > Elapsed Time:          0 00:54:43.02    CPU Time:          0 00:14:03.93
 > (01:36)$ > (01:36)$ !$ > (01:36)$  END_CONVERT_PDICCS_FILE: > ................ >o) >     thanks in advance for any insights.o > bclark at  lrgh dot orgw   ------------------------------  % Date: Wed, 10 Jul 2002 13:59:19 -0700p0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>2 Subject: Re: CPU time for batch job 4-MAY-1859 ???, Message-ID: <3D2C3DB6.4B620B0C@Mvb.Saic.Com>   Theo Jakobus wrote:e >  > Hi!k > T > We use an Alpha2100 with 2 CPUs for our production data base, which is Rdb V7.0-1. > \ > I take part in searching for extraterrestrial intelligence "SETI" by running 2 batch jobs. > My actual status:ie > http://setiathome.ssl.berkeley.edu/fcgi-bin/fcgi?email=theo.jakobus%40iaf.fhg.de&cmd=user_stats_new  > % > I'm really amazed by the following:a >  > 026AXP$ SHOW SYSTEM/BATCHTQ > OpenVMS V7.1-1H2  on node IAF026   9-JUL-2002 11:29:30.26  Uptime  340 01:04:21_I >                                                                     ^^^e[ > Yeah, on this server maintenance is done once a year only. Next appointment is August 10.r > P >    Pid    Process Name    State  Pri      I/O       CPU       Page flts  PagesQ > 2120011A SETI@home 54%   CUR  1   1 57563516 4-MAY-1859 04:     87961   2372  BpQ > 2120011B SETI@home 88%   COM      1 57460489 3-MAY-1859 19:     91649   2340  Bi> >                                               ^^^^^^^^^^^^^^> >                                               ?????????????? > 1 > The field CPU shows the CPU time used normally.  > ; > Did the extraterrestrial intelligence change the numbers?l > 1 > 3-MAY-1859: Was the batch started at this date?t > ] > 4-MAY-1859: Both batch jobs are started at the same time, why does the 2nd CPU waste a day?i >   G This is caused by the fact that the data location storing the amount of H cpu time the process has used has overflowed from a positive number to a< very large negative number.  The time display routine is nowC interpreting this as a delta time, rather than an absolute time.  AtE delta time with a large negative value will print out a date close toY- the zero-date used by VMS (17-NOV-1858 IIRC).t  G The probable reason for the difference between the two cpus is that the E primary cpu does work that the secondary cpus never do on your systemtH (handling interrupts, for example).  So secondary cpus will usually have! more idle time available to them.d  
 Mark Berrymane Mark.Berryman@Mvb.Saic.Com   ------------------------------  # Date: Wed, 10 Jul 2002 18:58:22 GMT/5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>s Subject: Re: DS10 shutting downl2 Message-ID: <it%W8.23$as6.313401@news.cpqcorp.net>  J Is this booted from a serial port or the graphics port.  If it's graphics,G put it on a serial port to see if something is printed.  It really doesl@ sound like an environmental shutdown.  Thermal, fans, airflow...      : Kevin Handy wrote in message <3D2C6A63.3020703@srv.net>... >Fred Kleinsorge wrote:nE >> Over temp is the most likely cause.  You need to run the error logn analyzerJ >> tool on the error log to find out the machine check that occured.  CallI >> field service, they can do this for you, and find out the failing FRU.n >o5 >Nothing seems to be logged at the time of the crash.t> >DIAG/SINCE=YESTERDAY only shows a few entries occuring during@ >boot 'cold start, ...', all with a 'time since last boot' underC >one minute. There are some later tape mount/dismount messages, andtD >some timestamps, but these are expected.  Nothing shown at the time >of the power down.  >e= >I think the command at the '>>>' prompt I was looking for ise@ >'show power', and plan on looking at it next time it goes down. >l >> >>; >> Kevin Handy wrote in message <3D2C557B.10904@srv.net>...c >>B >>>I have a customer with an Alpha DS10 that keeps powering itself> >>>down (at least once a day).  Nothing on the screen (DS10 is4 >>>powered down). I know that if a fan stops, it can7 >>>do this, but the fans on this box all seem to be OK.. >>> @ >>>Is there anything besides fans that can cause one of these to@ >>>power itself off, and is there any place I can look that will? >>>tell me why it has done this, like a show command at the >>> 
 >>>prompt? >>> > >>>System seems to be clean (no dust bunnies inside the case),8 >>>it's not in a very hot room, it's plugged into a UPS,7 >>>UPS seems to be Ok, Windows PC's at the site are not  >>>crashing. >A   ------------------------------  % Date: Wed, 10 Jul 2002 16:11:41 -0400y- From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: DS10 shutting down , Message-ID: <3D2C94F2.EBB826A3@videotron.ca>   Kevin Handy wrote:6 > Nothing seems to be logged at the time of the crash.  E Then you must reinstall the AE-35 antenna and wait for it to fail :-)a  M Have you considered plugging a printer to the serial port to log all messagesbD ? (eg: plug in a VT and use a serial console mode, and set the VT toL auto-print mode to record everything. You might then catch some message thatF occurs at the time the machine stops. (make sure VT and printer are on different circuits).  K Alternatively, you could use a wintel box to record to disk the output from  the serial line/console.   ------------------------------   Date: 10 JUL 2002 20:19:07 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>e Subject: Re: DS10 shutting down.2 Message-ID: <10JUL02.20190792@feda01.fed.ornl.gov>  I Well, it if *looks* like an environmental shutdown but the fans are fine, 1 then perhaps it's the sensor itself that's flaky.    Dave  M In a previous article, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote:tL > Is this booted from a serial port or the graphics port.  If it's graphics,I > put it on a serial port to see if something is printed.  It really doesnB > sound like an environmental shutdown.  Thermal, fans, airflow... >    >    >  o< > Kevin Handy wrote in message <3D2C6A63.3020703@srv.net>... > >Fred Kleinsorge wrote:oG > >> Over temp is the most likely cause.  You need to run the error log6
 > analyzerL > >> tool on the error log to find out the machine check that occured.  CallK > >> field service, they can do this for you, and find out the failing FRU.> > >n7 > >Nothing seems to be logged at the time of the crash.d@ > >DIAG/SINCE=YESTERDAY only shows a few entries occuring duringB > >boot 'cold start, ...', all with a 'time since last boot' underE > >one minute. There are some later tape mount/dismount messages, andeF > >some timestamps, but these are expected.  Nothing shown at the time > >of the power down.u > > ? > >I think the command at the '>>>' prompt I was looking for isaB > >'show power', and plan on looking at it next time it goes down. > >n > >> > >>= > >> Kevin Handy wrote in message <3D2C557B.10904@srv.net>...  > >>D > >>>I have a customer with an Alpha DS10 that keeps powering itself@ > >>>down (at least once a day).  Nothing on the screen (DS10 is6 > >>>powered down). I know that if a fan stops, it can9 > >>>do this, but the fans on this box all seem to be OK.l > >>> B > >>>Is there anything besides fans that can cause one of these toB > >>>power itself off, and is there any place I can look that willA > >>>tell me why it has done this, like a show command at the >>>t > >>>prompt? > >>>t@ > >>>System seems to be clean (no dust bunnies inside the case),: > >>>it's not in a very hot room, it's plugged into a UPS,9 > >>>UPS seems to be Ok, Windows PC's at the site are notI > >>>crashing. > >e >  x >  m --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOVdH Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Wed, 10 Jul 2002 15:12:30 -0600s From: Kevin Handy <kth@srv.net>a Subject: Re: DS10 shutting down-& Message-ID: <3D2CA33E.5070006@srv.net>   Kevin Handy wrote: > Fred Kleinsorge wrote: > F >> Over temp is the most likely cause.  You need to run the error log  >> analyzernJ >> tool on the error log to find out the machine check that occured.  CallI >> field service, they can do this for you, and find out the failing FRU.  >  > 6 > Nothing seems to be logged at the time of the crash.? > DIAG/SINCE=YESTERDAY only shows a few entries occuring duringoA > boot 'cold start, ...', all with a 'time since last boot' undernD > one minute. There are some later tape mount/dismount messages, andE > some timestamps, but these are expected.  Nothing shown at the times > of the power down. > > > I think the command at the '>>>' prompt I was looking for isA > 'show power', and plan on looking at it next time it goes down.h  @ Ok, the system powered down and I had them read off the settings= from a 'show power' command shortly after the crash. This wasi= read to me over the phone, so it should have looked somethinge like:i   	>>> show powern   	Power Supply: good  	...           goodt 	CPU Fan:      goodi 	Temperature:  goodr  , 	Current ambient temperature is 41 degrees C' 	System shutdown is set to 60 degrees C4  + 	0 Enviornmental events are logged in nvramn   	>>>    < I had them watch the ambiant temp for a little while, but it only went from 40 to 41.   >> >>; >> Kevin Handy wrote in message <3D2C557B.10904@srv.net>...t >>C >>> I have a customer with an Alpha DS10 that keeps powering itself/? >>> down (at least once a day).  Nothing on the screen (DS10 ism5 >>> powered down). I know that if a fan stops, it cani8 >>> do this, but the fans on this box all seem to be OK. >>>tA >>> Is there anything besides fans that can cause one of these to0A >>> power itself off, and is there any place I can look that willa@ >>> tell me why it has done this, like a show command at the >>> >>> prompt?' >>>r? >>> System seems to be clean (no dust bunnies inside the case),e9 >>> it's not in a very hot room, it's plugged into a UPS,.8 >>> UPS seems to be Ok, Windows PC's at the site are not
 >>> crashing.  >> >    ------------------------------  % Date: Wed, 10 Jul 2002 15:15:44 -0600a From: Kevin Handy <kth@srv.net>N Subject: Re: DS10 shutting down & Message-ID: <3D2CA400.1020403@srv.net>   Bob Koehler wrote:H > In article <3D2C557B.10904@srv.net>, Kevin Handy <kth@srv.net> writes: > A >>I have a customer with an Alpha DS10 that keeps powering itself = >>down (at least once a day).  Nothing on the screen (DS10 is 3 >>powered down). I know that if a fan stops, it canz6 >>do this, but the fans on this box all seem to be OK. >> >  > G >    I know you say it looks clean, but we learned the hard way to keepdE >    the airflow detector swithces on our 11/780s clean or they would3G >    think the fan is not running.  The rest of the 780 did look clean.  >   > I think that the fans are monitered through their power cables< now. I don'e see anything that looks like a air-flow monitor) anywhere. Please correct me if I'm wrong.    ------------------------------    Date: 10 Jul 2002 15:35:59 -0700& From: jordan@ccs4vms.com (Rich Jordan) Subject: Re: DS10 shutting downn= Message-ID: <cc5619f2.0207101435.324b6102@posting.google.com>h  K Kevin Handy <kth@srv.net> wrote in message news:<3D2C557B.10904@srv.net>... A > I have a customer with an Alpha DS10 that keeps powering itself = > down (at least once a day).  Nothing on the screen (DS10 isn3 > powered down). I know that if a fan stops, it can66 > do this, but the fans on this box all seem to be OK. >   A I don't have a DS10 or manual handy.  Is this something where thep@ remote management console accessible from serial port 1 would be; useful?  If I'm remembering correctly there are 'com1_mode'gD environment variables that can allow or block access to the RMC, andD that if it is allowed, entering <escape><escape>rmc gets you the RMCA prompt, and 'quit' or 'exit' takes you back to console. (<escape>rE above is the escape key, or ctrl-[).  You probably want your terminald at 9600,N,8,1 for this.r  F I think there's online help, but I remember using 'STATUS' to see what. the RMC thought about the state of the system.  
 Good luck.   ------------------------------  % Date: Wed, 10 Jul 2002 19:18:39 -0400.- From: JF Mezei <jfmezei.spamnot@videotron.ca>a Subject: Re: DS10 shutting downv, Message-ID: <3D2CC0B8.D35EEDD8@videotron.ca>   Kevin Handy wrote:@ > I think that the fans are monitered through their power cables> > now. I don'e see anything that looks like a air-flow monitor+ > anywhere. Please correct me if I'm wrong.D  L Not familiar with that hardware, but perhaps optical sensors are what detect if the fan is turning or not ?   ------------------------------  % Date: Wed, 10 Jul 2002 20:39:11 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com>e Subject: RE: DS10 shutting downtT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4023D92A7@kaoexc01.americas.cpqcorp.net>   Kevin,  D Past general experience with these types of issues is that the powerC supply is flaky.  Usually nothing logged on the system errorlog andrF console message would just indicate a message which looks like someone( kicked a breaker at the rear of the cab.  C If I had to provide a WAG, it would be to replace the power supply.R   :-)    Regardsd  
 Kerry Main Senior ConsultantN Hewlett-Packard Canada! Consulting & Integration Servicesf Voice: 613-592-4660t Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----) From: Kevin Handy [mailto:kth@srv.net]=20  Sent: July 10, 2002 11:41 AM To: Info-VAX@Mvb.Saic.Comh Subject: DS10 shutting downz    H I have a customer with an Alpha DS10 that keeps powering itself down (atH least once a day).  Nothing on the screen (DS10 is powered down). I knowF that if a fan stops, it can do this, but the fans on this box all seem	 to be OK.-  C Is there anything besides fans that can cause one of these to powerFF itself off, and is there any place I can look that will tell me why it5 has done this, like a show command at the >>> prompt?   G System seems to be clean (no dust bunnies inside the case), it's not innE a very hot room, it's plugged into a UPS, UPS seems to be Ok, WindowsS" PC's at the site are not crashing.   ------------------------------  % Date: Wed, 10 Jul 2002 21:50:22 -0400d2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: DS10 shutting downtJ Message-ID: <rdeininger-1007022150220001@1cust58.tnt3.nashua.nh.da.uu.net>  D In article <it%W8.23$as6.313401@news.cpqcorp.net>, "Fred Kleinsorge"$ <kleinsorge@star.zko.dec.com> wrote:  K >Is this booted from a serial port or the graphics port.  If it's graphics,nH >put it on a serial port to see if something is printed.  It really doesA >sound like an environmental shutdown.  Thermal, fans, airflow...a >a  E Yes, but an environmental shutdown should be proceded by a warning, IlC think.  That would show up in the error logs.  Unless the sensor is2G flakey, and jumps from "good" to "fatal" with no intervening "warning".c  F Dunno if the DS10 has any NVR where these conditions might be logged.  Newer systems do.t   Is the firmware up to date?o   What version of VMS?  J I don't know of any specific issues around environmentals on the DS10, but/ then I don't pay much attention to that system.d   ------------------------------  % Date: Thu, 11 Jul 2002 00:55:39 -0400?( From: David Froble <davef@tsoft-inc.com> Subject: Re: DS10 shutting downr, Message-ID: <3D2D0FCB.7090604@tsoft-inc.com>  N I'm going to go with this one also.  I had one AlphaStation 200 that would do N the same thing.  Upon looking into the power supply, one of the contacts from N the plug wasn't soldered to whatever.  There was contact, and thus power, but D sometimes the contact would be broken, and the system would re-boot.   Dave     Main, Kerry wrote:   > Kevin, > F > Past general experience with these types of issues is that the powerE > supply is flaky.  Usually nothing logged on the system errorlog and H > console message would just indicate a message which looks like someone* > kicked a breaker at the rear of the cab. > E > If I had to provide a WAG, it would be to replace the power supply.: >  > :-)  > 	 > Regards  >  > Kerry Main > Senior ConsultantC > Hewlett-Packard Canada# > Consulting & Integration Servicesi > Voice: 613-592-4660  > Fax   : 613-591-4477 > Email: Kerry.Main@hp.com >  >  > -----Original Message-----) > From: Kevin Handy [mailto:kth@srv.net] e > Sent: July 10, 2002 11:41 AM > To: Info-VAX@Mvb.Saic.Com$ > Subject: DS10 shutting down. >  > J > I have a customer with an Alpha DS10 that keeps powering itself down (atJ > least once a day).  Nothing on the screen (DS10 is powered down). I knowH > that if a fan stops, it can do this, but the fans on this box all seem > to be OK.c > E > Is there anything besides fans that can cause one of these to powerJH > itself off, and is there any place I can look that will tell me why it7 > has done this, like a show command at the >>> prompt?p > I > System seems to be clean (no dust bunnies inside the case), it's not inbG > a very hot room, it's plugged into a UPS, UPS seems to be Ok, Windowsc$ > PC's at the site are not crashing. >  >    ------------------------------  % Date: Wed, 10 Jul 2002 14:18:43 -0400S- From: JF Mezei <jfmezei.spamnot@videotron.ca> % Subject: Re: Looking for your opinion:, Message-ID: <3D2C7A7F.FB06D8B6@videotron.ca>   system@SendSpamHere.ORG wrote:J > I have a color printer and it takes postscript.  Conversions from PDF to > PS just don't cut it.   H Care to explain ?  Since PDF is essentially pre-executed Postscript, theL output should be identical. Perhaps you are using a poor PDF reader software to print the documents.T  I "pre-executed" -> the PDF writer executes the postscript and records only  drawing commands.   
 for instance:     5 { (hello) show } repeat   gets "translated" to: A (hello) show (hello) show (hello) show (hello) show (hello) show e    M Now, where PDF also modifies the postscript is in images, if you specify that K images should be downsampled for on-line viewing (72dpi) or kept "whole" attO their original resolution. (same with colour conversions from CMYK to RGB etc).h   ------------------------------  % Date: Wed, 10 Jul 2002 20:09:30 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>m% Subject: Re: Looking for your opinionT' Message-ID: <3D2C785A.44CE51F0@aaa.com>i  5 Well, if you have a number of alternatives on line tot) download, which is the "primary" anyway ?l   Jan-Erik Sderholm.e   PS.h> I'd prefer PDF, but it isn't realy rellevant what I prefer, as< long as PDF will be one of the downloadable formats. And I'm' 100% sure that PDF can't be left out...  DS.      JF Mezei wrote:h >  > re: format for distribution. > O > I guess it depends on how the stuff is actually written. If the source of theaM > journal is done in a piece of software whose "export to HTML" really sucks,1F > then HTML should not be considered as primary distribution material.   ------------------------------  % Date: Wed, 10 Jul 2002 14:41:12 -04001- From: JF Mezei <jfmezei.spamnot@videotron.ca>i% Subject: Re: Looking for your opiniong, Message-ID: <3D2C7FC3.7BD63895@videotron.ca>   WILLIAM WEBB wrote: 9 > Bookreader by definition equals preaching to the choir.l > @ > And I think we all agree that preaching to the choir isn't the$ > most effective way to proselytize.  K But is a technical journal not designed specifically to educate the choir ?lF We're not talking about marketing materials here, we are talking aboutJ technical information designed to help the converted, not to help convert.   ------------------------------  # Date: Wed, 10 Jul 2002 18:41:45 GMTA From: system@SendSpamHere.ORGf% Subject: Re: Looking for your opinion 0 Message-ID: <00A10BA8.607CBF36@SendSpamHere.ORG>  \ In article <3D2C7A7F.FB06D8B6@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >system@SendSpamHere.ORG wrote:nK >> I have a color printer and it takes postscript.  Conversions from PDF tom >> PS just don't cut it. >hI >Care to explain ?  Since PDF is essentially pre-executed Postscript, the>M >output should be identical. Perhaps you are using a poor PDF reader softwaret >to print the documents. >kJ >"pre-executed" -> the PDF writer executes the postscript and records only >drawing commands. >w >for instance: >a > 5 { (hello) show } repeats >u >gets "translated" to:B >(hello) show (hello) show (hello) show (hello) show (hello) show  >u >oN >Now, where PDF also modifies the postscript is in images, if you specify thatL >images should be downsampled for on-line viewing (72dpi) or kept "whole" atP >their original resolution. (same with colour conversions from CMYK to RGB etc).    K Well, Adobe's Java crap bearly runs and looped for hours before I killed it'L when I tried to print a simple PDF.  The ASI PDF viewer is probably the bestL VMS viewer and renders a pretty good PS output.  Ghostview has problems withM outputting images that don't look like a picture from the Sunday comic pages.oL Some of the PDF viewers will not produce color PS.  Whatever fires up on theM PeeCee if a PDF is displayed does not produce color on my LNCO2.  Output (PS)-J from all the above looks quite different on the same printer.  There goes ) your "output should be identical" theory.i   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMn            e5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Wed, 10 Jul 2002 15:30:10 -0400d- From: JF Mezei <jfmezei.spamnot@videotron.ca>f% Subject: Re: Looking for your opinion-, Message-ID: <3D2C8B3A.4B57D589@videotron.ca>   Sue Skonetski wrote:L > Just so you know and Warren will correct me if I am wrong.  I believe that> > if we do a PDF output we can also translate the PDF to html,   PDF to HTML would not be easy.  L But it really depends what tools are at your disposal to create the materialM in the first place and from there, you can define the process that create theo final distributed output.   L It is more likely that the publishing software you use (Pagemaker, Quark, orI even Word) has the ability to generate both PDF and HTML from the source.m  M However, bear in mind that most software have a much easier time creating PDF L (and implicitely postscript) because it represents accurately the layout andO printing intentions, whereas creating good HTML will require work on your part.>  N For instance, you want to make sure that a certain typestyle gets "translated"L to something good in HTML. Software will have default tables to convert fromM their "printer ready" format to HTML, but those can be quite ugly and require : much customization to match the style of your publication.    E If this goes ahead, what I would suggest you do is to build a journalsK prototype using old material and then publish it to us so we can comment onrN the various formats you produced. From that feedback, you can then adjust your" software to produce better output.  J For instance, if you use PageMaker or Quark, you can set certain styles toH automatically generate an entry in the bookmarks so that one can quicklyL access parts of the PDF output. But if you don't know that exists, you won'tL be looking for it and the PDF document will just be one long string of pages without an index.w  M Similarly, depending on how your software generates HTML, we could comment onlL what we'd like to see changed in the produced HTML (font zizes, table widthsN etc etc) and you can then go back to your software and tewak the HTML settings  to generate more palatable HTML.  N If you do this during the development stage, by the time you produce the firstK official one, you will have much better output and can then proceed to just K embed new material into each issue, knowing that all the pre-set styles andr9 settings will produce a consistent and acceptable output.a   ------------------------------  # Date: Wed, 10 Jul 2002 19:46:08 GMTn( From: Don Sykes <annonymous@pacbell.net>% Subject: Re: Looking for your opinion-+ Message-ID: <3D2C8F7E.DF72EA02@pacbell.net>0   John Smith wrote:d >  > Sue, > L > An excellent idea, the original of which should never have been allowed to > die, btw.r > K > Maybe as a start, think about taking some of the older articles from dayseN > gone by, have them reviewed for technical accuracy in light of new features,M > bug fixes, new controllers, etc... and publish those as OpenVMS 'Classics'.  > G > Perhaps these older articles could be organized into Intro & AdvancediC > volumes ... ideal for distribution into .edu environments when HPuK > **donates** Alphas running OpenVMS to build up the next generation of VMSe
 > specifiers.e >  Hear, hear.l   -- p   Have VMS. Will Travel. Wire Paladin (@alphase.com)t
 San Franciscoe   ------------------------------  # Date: Wed, 10 Jul 2002 20:13:05 GMTo( From: Don Sykes <annonymous@pacbell.net>% Subject: Re: Looking for your opinionn+ Message-ID: <3D2C95CF.E476E15F@pacbell.net>y   Sue Skonetski wrote: > F > This is just some research that I am doing and I am looking for yourJ > opinion.  And while some of you may say that I am preaching to the choirM > again, I prefer to think of it as talking to family.  This is research onlyt& > and is not a commitment from anyone. > M > One of the things that I noticed when I was the editor of  OpenVMS times is L > that it was very difficult to convey in depth technical information in 300I > words or less.  If you have been around for awhile you may remember DTJo2 > (Digital Technical Journal) which was very good. > 3 > What do think about an OpenVMS Technical Journal?a >  Excellent idea.p  4 > Questions and Answers (what I have learned so far) >  > Q.    How often? > A.    Probably twice a year  I'd suggest quarterly.   >  > Q.    What kind of formattD > A.     Electronic only, probably PDF so it prints somewhat decent.< Basic HTML is my choice, since any browser can deal with it.  e > ! > Q.    Who could submit articleseJ > A.    I have not thought this through yet.  My first reaction is anyone,0 > there would be an acceptance and review cycle.M Agree, but who would do the review? I hope not "certain people" from COV that I just live to find holes in any statement or idea. The reviewers should be>O helpful in editing for correctness, rather than just rejecting. No one wants to B take their valuable time to write an article only to be dumped on.K I guess I'm saying the reviewers, whoever they are, should be courteous andDP supportive in their comments, so as to encourage people to submit again, even if their article is rejected.   >  > Q.    What would be the focuscM > A.    OpenVMS and related technologies on a technical level, combination ofa > advanced and intro level.b  N I'd say mid to advanced level, since there is a wealth of info already on lineM for intro level users. I'd also like special emphasis on using VMS in today'scA internet world - e.g. as a web client/server, use with Java, etc.u   > ( > Any questions or comments are welcome. >  > Warm Regards,f > Sue      --     Have VMS. Will Travel. Wire Paladin (@alphase.com)h
 San Francisco    ------------------------------  # Date: Wed, 10 Jul 2002 20:36:53 GMT-1 From: "Terry C. Shannon" <terryshannon@attbi.com>-% Subject: Re: Looking for your opinionr. Message-ID: <FV0X8.142284$Uu2.32129@sccrnsc03>  5 "Don Sykes" <annonymous@pacbell.net> wrote in message % news:3D2C95CF.E476E15F@pacbell.net...a > Sue Skonetski wrote: <snip> >  > > # > > Q.    Who could submit articlesaL > > A.    I have not thought this through yet.  My first reaction is anyone,2 > > there would be an acceptance and review cycle.J > Agree, but who would do the review? I hope not "certain people" from COV thatK > just live to find holes in any statement or idea. The reviewers should benH > helpful in editing for correctness, rather than just rejecting. No one wants toD > take their valuable time to write an article only to be dumped on.I > I guess I'm saying the reviewers, whoever they are, should be courteous> andeJ > supportive in their comments, so as to encourage people to submit again, even ifb > their article is rejected.  I Well stated. I've done my share of article acquisition and reviewing (andhK prior to that time I submitted plenty of swill) and IMHO the last thing one L wants to do is discourage the author of a rejected manuscript from giving itI another shot. Clearly those whose idea of having a life involves punching@G holes in every assertion that wanders by are not the sort of people who30 ought to be placed in an evaluation/review role.   ------------------------------  % Date: Wed, 10 Jul 2002 16:55:12 -0400s; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>l% Subject: Re: Looking for your opiniona" Message-ID: <3d2c9f4f@news.si.com>  2 >What do think about an OpenVMS Technical Journal?   Great idea!    >Q.    What kind of formatC >A.     Electronic only, probably PDF so it prints somewhat decent.y  $ How do you print PDF files on a VAX? -- wA Brian Tillman                   Internet: tillman_brian at si.comaA Smiths Aerospace                          tillman at swdev.si.come= 3290 Patterson Ave. SE, MS      Addresses modified to prevento< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Wed, 10 Jul 2002 17:02:28 -0400e* From: WILLIAM WEBB <WWEBB1@email.usps.gov>% Subject: RE: Looking for your opinionN- Message-ID: <0033000071975959000002L092*@MHS>i  9 =0AHymnals can be used to teach people to sing who aren'ti members of the choir (yet).-  9 A technical journal can have external as well as internal] impact.t  4 So it is with a technical journal, the resurgence of1 which can be seen as another small indicator thati4 HP may indeed take VMS more seriously than did prior
 stakeholders.y  4 I remain hopeful that bigger indicators are to come.   WWWebb   -----Original Message-----/ From: Info-VAX-Request@Mvb.Saic.Com at INTERNETs& Sent: Wednesday, July 10, 2002 2:31 PMB To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET% Subject: RE: Looking for your opinione     WILLIAM WEBB wrote:d9 > Bookreader by definition equals preaching to the choir.u >e@ > And I think we all agree that preaching to the choir isn't the$ > most effective way to proselytize.  H But is a technical journal not designed specifically to educate the cho= ir ?F We're not talking about marketing materials here, we are talking aboutH technical information designed to help the converted, not to help conve= rt.=   ------------------------------  % Date: Thu, 11 Jul 2002 00:01:53 +0200@9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>d% Subject: Re: Looking for your opiniona' Message-ID: <3D2CAED1.9D681D40@aaa.com>'   I'd print them on my printer...o   :-)e   Jan-Erik Sderholm.e   Brian Tillman wrote: > >Q.    What kind of formatE > >A.     Electronic only, probably PDF so it prints somewhat decent.o > & > How do you print PDF files on a VAX? > --   ------------------------------  # Date: Wed, 10 Jul 2002 22:26:59 GMTu# From: "John Smith" <a@nonymous.com>t% Subject: Re: Looking for your opinion1G Message-ID: <Tw2X8.7022$6DW1.2654@news01.bloor.is.net.cable.rogers.com>t  5 "Nic Clews" <sendspamhere@127.0.0.1> wrote in message # news:3D2C5172.65BBC4ED@127.0.0.1...  > John Smith wrote:  > >e@ > > "Terry C. Shannon" <terryshannon@attbi.com> wrote in message, > > news:M_XW8.471536$cQ3.40016@sccrnsc01... > > >2 > > >3# > > > > 15. Introduction to Vax/Vmss2 > > > > by Terry C. Shannon (Paperback - May 1985)* > > > > Out of Print--Limited Availability > > >n > > > J > > > "Limited Availability" is an understatement. I have exactly one copy of > > thisK > > > 17-year-old tome. Which is probably just as well as the book is basedd on > > > VAX/VMS 4.0s > >a" > > Time do do a 2nd edition then. > & > John, when are you starting work on: >lD > > Probably what is needed most for VMS newbies is a book entitled, "Everythingc? > > You Know Is Wrong - A Guide to OpenVMS for Unix and Windows.J > > Programmers/Users/Managers" (c) 2002, John Smith. All Rights Reserved. :-)t >d > The title is excellent.     
 Thank you.  L I've often thought about doing something like this but, unfortunately, givenL HP's lack of marketing of OpenVMS and Compaq's lack thereof before that,  it! would be a fairly limited seller.t  J What I am working on is a business best-seller, "Reverse Alchemy - TurningF Gold into Lead - The true story of OpenVMS and ineptitude in CorporateF America". I expect it will run to several volumes and be turned into aK series of action-adventure movies, much like the Star Wars or Indiana JonesgH films. For actors I'm thinking of getting Mike Myers to reprise his "Dr.J Evil", of Austin Powers fame in portraying Capellas; Roger Moore to play aK caricature role as Palmer;  Dick Cheney in a Machiavellian-like role (at an K undisclosed location) as Ben Rosen; and either Margaret Thatcher or Joan ofsG Arc as Carly. Of course there will be plenty of sycophants in the cast.    ------------------------------  % Date: Wed, 10 Jul 2002 18:40:16 -0400r- From: JF Mezei <jfmezei.spamnot@videotron.ca> % Subject: Re: Looking for your opinion , Message-ID: <3D2CB7BB.5980C859@videotron.ca>  M Of course, what I'd really want to see is the journal distributed in a native- VMS format:-  
 		WPS PLUS    ' :-) :-) :-) :-) :-) :-) :-) :-) :-) ;-)   L Actually, how about DDIF format ? You can type it on a character cell screenL and get text only, or you can display it on a workstation (VIEW ) , print it etc etc ....   ------------------------------  # Date: Wed, 10 Jul 2002 22:29:00 GMTs# From: "John Smith" <a@nonymous.com> % Subject: Re: Looking for your opinionPG Message-ID: <My2X8.7028$6DW1.2465@news01.bloor.is.net.cable.rogers.com>P  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D2C7FC3.7BD63895@videotron.ca... > WILLIAM WEBB wrote:r; > > Bookreader by definition equals preaching to the choir.e > > B > > And I think we all agree that preaching to the choir isn't the& > > most effective way to proselytize. >aK > But is a technical journal not designed specifically to educate the choir  ?dH > We're not talking about marketing materials here, we are talking aboutL > technical information designed to help the converted, not to help convert.    & Why not kill two birds with one stone?   ------------------------------  # Date: Wed, 10 Jul 2002 22:43:32 GMTa# From: "John Smith" <a@nonymous.com> % Subject: Re: Looking for your opinion0G Message-ID: <oM2X8.7048$6DW1.4733@news01.bloor.is.net.cable.rogers.com>n  F "Brian Tillman" <tillman_brian@notnoone.notnohow.com> wrote in message news:3d2c9f4f@news.si.com...4 > >What do think about an OpenVMS Technical Journal? >f
 > Great idea!l >o > >Q.    What kind of formatE > >A.     Electronic only, probably PDF so it prints somewhat decent.t >r& > How do you print PDF files on a VAX?     FTP it to a PC or Mac.   ------------------------------  # Date: Wed, 10 Jul 2002 22:45:34 GMTe1 From: "Terry C. Shannon" <terryshannon@attbi.com> % Subject: Re: Looking for your opiniont. Message-ID: <iO2X8.143365$Uu2.32439@sccrnsc03>  . "John Smith" <a@nonymous.com> wrote in messageA news:My2X8.7028$6DW1.2465@news01.bloor.is.net.cable.rogers.com...f >y< > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3D2C7FC3.7BD63895@videotron.ca... > > WILLIAM WEBB wrote:o= > > > Bookreader by definition equals preaching to the choir.t > > >rD > > > And I think we all agree that preaching to the choir isn't the( > > > most effective way to proselytize. > >hG > > But is a technical journal not designed specifically to educate thei choirt > ?hJ > > We're not talking about marketing materials here, we are talking aboutE > > technical information designed to help the converted, not to helpr convert. >n >a( > Why not kill two birds with one stone? >n  K Good thing to do if you can figure out how. Those unfamiliar with VMS mightFK not be able to slog through all the technobabble. If you don't know a printeF symbiont from a QIO, detailed delvings into such arcane subjects might) render the publication a bit off-putting.t   ------------------------------  # Date: Wed, 10 Jul 2002 22:36:30 GMTu# From: "John Smith" <a@nonymous.com>b% Subject: Re: Looking for your opinionoG Message-ID: <OF2X8.7036$6DW1.5667@news01.bloor.is.net.cable.rogers.com>t  5 "Dave Gudewicz" <dgudewicz@core.com> wrote in messaged2 news:3d2c486f$0$1433$1dc6e903@news.corecomm.net...! > > Q.    What would be the focussL > > A.    OpenVMS and related technologies on a technical level, combination > of > > advanced and intro level.e >hL >     All levels make sense to me.  Leave one out and you leave that crop of > readers out.  And I can F >     echo Ken's Windows in H.S.--> U**X in college comment.  I have a summer > intern and that's E >     exactly what's going on with him.  I'm sure there are others ini similar $ > situations.  This needs to change.I >     Perhaps this publication winding up in school libraries and CS labsd! > would make sense.  Nice PR too.     L Get every Computer Science prof's e-mail address at every campus in at leastJ the English-speaking world and send copies out via L-Soft's list-server toD get their attention. You have ten's of thousands of Computer Science/ academic staff who have NEVER heard of OpenVMS.f   ------------------------------  # Date: Wed, 10 Jul 2002 23:58:39 GMTlL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")% Subject: Re: Looking for your opinionW8 Message-ID: <00A10BBB.80517DBB@SSRL04.SLAC.STANFORD.EDU>  J In article <01KJXRWTGEQ296Y2PR@sysdev.deutsche-boerse.com>, Phillip Helbig, <HELBPHI@sysdev.deutsche-boerse.com> writes:   >:J >There is a frequent poster here who works in VMS engineering and is very I >knowledgeable about VMS and recently discussed formats for the FAQ here. E >Why not the same approach: use a tool (DOCUMENT?) which can produce hG >multiple formats, so you have one source and many formats without the  G >extra work.  Put them all on a WWW site and/or FTP server; people can eC >get those formats they need and redistribute them as appropriate, i >including printing them.   N I totally support this.  (DEC Document is the appropriate tool; it can produceN PostScript, HTML, something approximating plain text (LN03-printable, anyway),N and Bookreader (!) documents.  I think you need a second pass to make PDF from the PostScript.)   -- Alane  O =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056gM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210rO ===============================================================================n   ------------------------------  # Date: Thu, 11 Jul 2002 00:54:29 GMTw# From: "John Smith" <a@nonymous.com>r% Subject: Re: Looking for your opiniontG Message-ID: <9H4X8.7333$6DW1.4971@news01.bloor.is.net.cable.rogers.com>c  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:iO2X8.143365$Uu2.32439@sccrnsc03... >u0 > "John Smith" <a@nonymous.com> wrote in messageC > news:My2X8.7028$6DW1.2465@news01.bloor.is.net.cable.rogers.com...q > >y> > > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message* > > news:3D2C7FC3.7BD63895@videotron.ca... > > > WILLIAM WEBB wrote: ? > > > > Bookreader by definition equals preaching to the choir.o > > > >rF > > > > And I think we all agree that preaching to the choir isn't the* > > > > most effective way to proselytize. > > >sI > > > But is a technical journal not designed specifically to educate thet > choirg > > ? L > > > We're not talking about marketing materials here, we are talking aboutG > > > technical information designed to help the converted, not to helpa
 > convert. > >e > >j* > > Why not kill two birds with one stone? > >  >tG > Good thing to do if you can figure out how. Those unfamiliar with VMSp mightuG > not be able to slog through all the technobabble. If you don't know a0 print H > symbiont from a QIO, detailed delvings into such arcane subjects might+ > render the publication a bit off-putting.l  = Maybe the first issue should be only a glossary in that case.,   ------------------------------  # Date: Thu, 11 Jul 2002 01:57:22 GMT 1 From: "Terry C. Shannon" <terryshannon@attbi.com>e% Subject: Re: Looking for your opinionh. Message-ID: <5C5X8.144939$Uu2.32982@sccrnsc03>  . "John Smith" <a@nonymous.com> wrote in messageA news:9H4X8.7333$6DW1.4971@news01.bloor.is.net.cable.rogers.com...s >l> > "Terry C. Shannon" <terryshannon@attbi.com> wrote in message  I > > Good thing to do if you can figure out how. Those unfamiliar with VMSl > mightiI > > not be able to slog through all the technobabble. If you don't know ar > print J > > symbiont from a QIO, detailed delvings into such arcane subjects might- > > render the publication a bit off-putting.e >1? > Maybe the first issue should be only a glossary in that case.> >u  K Well, back when I wrote for Digital Review, we divided stuff into technicaltA articles (in-depth product reviews, articles surveying a specific I technology, eg databases, I/O controllers, OS features, whatever; and thea? like) and feature articles which were a bit more palatable to a 0 less-technical audience. The mix seemed to work.   ------------------------------  % Date: Wed, 10 Jul 2002 22:33:29 -0400c2 From: rdeininger@mindspring.com (Robert Deininger)% Subject: Re: Looking for your opinioneJ Message-ID: <rdeininger-1007022233290001@1cust58.tnt3.nashua.nh.da.uu.net>  E In article <XPXW8.651$6DW1.553@news01.bloor.is.net.cable.rogers.com>, $ "John Smith" <a@nonymous.com> wrote:    K >My Alpha doesn't fit on my seatback table at 35,000 feet, so Bookreader iswM >out. Besides when I want to open the window to cool the thing off, the cabina+ >attendants and pilots give me funny looks.t  H You obviously have the wrong airplane.  Maybe you should talk to Grumman# or someone similar about upgrading.s   ------------------------------  % Date: Wed, 10 Jul 2002 22:42:04 -0400e2 From: rdeininger@mindspring.com (Robert Deininger)% Subject: Re: Looking for your opinionlJ Message-ID: <rdeininger-1007022242040001@1cust58.tnt3.nashua.nh.da.uu.net>  G In article <OF2X8.7036$6DW1.5667@news01.bloor.is.net.cable.rogers.com>,a$ "John Smith" <a@nonymous.com> wrote:    M >Get every Computer Science prof's e-mail address at every campus in at least K >the English-speaking world and send copies out via L-Soft's list-server to E >get their attention. You have ten's of thousands of Computer Scienceu0 >academic staff who have NEVER heard of OpenVMS.   Wouldn't that be spam?   ------------------------------   Date: 10 Jul 2002 18:04:45 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh...i0 Message-ID: <aghsvt$a7m$1@pegasus.csx.cam.ac.uk>  , In article <3D2C7373.2E83A221@videotron.ca>,/ JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:a >Nick Maclaren wrote:iD >> As I said, nothing has changed in that respect.  HP cannot afford@ >> it, because the IA-64 architecture is a very high-cost one toE >> develop and develop for.  Not just the hardware, but the software.. >iI >But IA64 has been developped. It is almost a real chip by now. 	*IF* the-M >architecture was properly done, then incremental improvements shouldn't costnK >that much.  Also, if half the chip resides in the compilers, then HP still8O >ends up paying a big part of the chip through its own proprietary compiler(s).  >sK >In the end, a HP proprietary IA64 will be no different from Sun's SPARC or 
 >IBM's Power.o  B I think that you are still underestimating the ongoing cost caused@ by the IA-64's complexity.  I have located some pretty foul bugs? in pretty foul systems in my time - sometimes where the vendors"A had given up - and the thought of the IA-64 makes me gibber.  Anda@ I am one of the last few people in Cambridge (not to say the UK)> with recent experience of tracking down the sort of bug that I am referring to.  @ The cost of complexity does not go away even after the operatingB systems and compilers are 'tested' - it continues to cause trouble@ to people trying to locate bugs in 'working' applications.  Note> that the bug can well be a simple one - the complexity is what@ can hide it.  And the IA-64 architecture is the most complicated' I have ever seen, by a factor of three.n  @ Intel can pay ISV's to put the extra effort in.  Could HP do the? same, on its own?  If not, we would get into the situation that A some important applications were ported late, badly or not at allr> to HP's unique IA-64 systems.  Customers would then follow the- applications, as they have done for 40 years.-  N >Seems to me that, performance being equal, the 64 bit chip that will win will. >be the one that can capture a greater volume.  @ I hope that no single chip wins.  Intel were intending the IA-64@ to wipe out all other general-purpose designs; this would be badA for computing and, in the long term, bad for Intel.  Witness whatp5 happened to IBM.  An AMD monopoly would be no better.u  L >Consider the theoretical possibility of Apple, now having fresh source codeO >for its OS, would revamp it to run 64 bits on the current Power_x chips.  ThatsM >would surely beat IA64's volumes as long as IA64 is considered a high-margina >product restricted to servers.p  @ Not really.  Even in terms of number of systems, there are a lot5 more medium- and high-end servers than Apple systems.,     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679w   ------------------------------  # Date: Wed, 10 Jul 2002 17:46:31 GMTe* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh...tA Message-ID: <Xp_W8.123357$vq.6309550@bin6.nnrp.aus1.giganews.com>i  3 "Ken Green" <Ken.Green@kgcc.co.uk> wrote in message $ news:3D2C01B6.FD9AA4F0@kgcc.co.uk...   ...r  L > Your right Bill, the L2 is in the same package, but not on the same, chip,G > altough the tags are on the CPU chip, along with 2 cores, and the twoM= > L1 caches (the data caches are half the size of the PA8700)E  F Well, being in the same package does help - and also helps explain whyK adding additional PA processors to a system is rather expensive.  Apologiese+ for not being more familiar with the beast.t   > F > I would have thought that cost was a major factor in deciding not to followJ > a similar approach in the McKinley, I guess HP can still afford more forF > CPUs in their mid to upper range systems than Intel expects the IA64I > market place to want to pay. I might be wrong, maybe we'll see Makos in ? > the the rp24XX boxes which start with a list price of <$1000.5  I I wouldn't hold my breath waiting, given that even the runt-of-the-littere$ processor chips cost more than that.  K However, it may be a matter of decreased marginal benefit.  It sounds as ifiI the PA chips didn't have room for much on-chip cache (perhaps they do nowaE but there's never seemed to be sufficient gain to modify the originalsK architecture) so went the MCM route.  3 MB on the chip is at least starting2K to get respectable, and the 6 MB Madison will offer next year even more so,yI especially given that it's most likely noticeably lower-latency than PA'soH cache:  with that much fast cache on the chip, the marginal utility of a( larger one off-chip may be fairly small.   >w > > H > > The better question is exactly why HP would believe that EPIC (if it indeedL > > did revert to being an HP-proprietary product) would be a useful step upK > > from PA-RISC which, despite relative neglect of late, has just released  anG > > 875 MHz part that should get about 700 SPECint2K performance if thei increaseJ > > is close to linear and next year around Madison time will get a shrink toL > > 130 nm (keeping up with Madison in that regard) *plus* dual cores on theL > > chip (yielding close to double the server performance per die).  There's noH > > evidence that EPIC has left a great deal of performance on the table justJ > > waiting to be realized in future iterations (though the Alpha guys canH > > probably find some if anyone can, or throw it in the trash and start over),G > > so whatever degree of effort it would take to keep improving Itanici couldt> > > likely be equally effective in improving good old PA-RISC. > G > EPIC is PA3, it was designed as the replacement for PA-Risc, sure thei worldlI > has moved on a long way since 1990 when they started, but the PA8700 isT notgL > significantly different in terms of what it does from the PA8000. It's not like > withK > Alpha where over the same time frame things like OOO have been added, PA2u > had them already.t  H Understood.  However, since EPIC has come nowhere near to fulfilling theH "three times as fast as Alpha!" promises made 'way back then (i.e., it'sD here, and it's competitive, but no more than that), and since PA hasI continued to slog ahead to the point where the 875 MHz 8700s shouldn't be I significantly slower at SPECint than McKinley (with the shrunk, dual-corelL 8800s on the way next year to out-perform Madison in such server-style use),H and since the only paths forward I've seen anyone suggest for EPIC (SMT,J multi-core dice) apply equally well to the PA architecture, and since EPICH seems to be at the bottom of the barrel when it comes to performance perH Watt, I still maintain that it would seem difficult to justify funnelingL further development dollars into EPIC instead of into PA (and/or Alpha) *if*G Itanic should turn out to be only an HP-proprietary product rather thanoB become the 'industry standard' its supporters keep hoping it will.   >- > >- > > H > > Oh, wait:  that *was* the conclusion they reportedly came to a while ago,3 > > but couldn't make fly at the corporate level...8 > >$ >1L > Well the management have well a truely nailed their flag to the IA64 mast.G > Carley has already said that the whole farm has been bet on IA64, sheI5 > even went as far as buying Compaq to help prove it.  >aK > Technical people are easier to convince, generally if you can show enougheJ > evidence you can win arguements, they don't like being proved wrong, but > they can be. >d1 > Most corperate levels... won't be proved wrong.e   Amen.a   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 15:58:16 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>r Subject: Re: McKinley Cometh...t, Message-ID: <3D2C91CE.BC76705B@videotron.ca>   Sander Vesik wrote:0  2S > But HP does not need to wait for intel to decide that ia64 won't make an industrynR > standard - it can reach that conclusion on its own and act accordingly. At least! > in theory it should be able to.n > +++ Out of cheese error +++e    L No. Carly bet her job, her carreer and her reputation on this. Remember thatM prior to the merger, her competence was in question. Remember that during theoN merger pregnancy, most HP staff were against her project. And now that she hasI given birth to the "new" HP, there is a lot of work ahead to prevent Wall B Street from declaring that buying Compaq was a big costly mistake.  K It will be quite a while before Carly has enough tenure/power to admit that2L there were some strategic mistakes such as fofucisng entirely on an unproven chip.   - Compare Carly's decisions with those of IBM. o  N IBM developped the PowerPC and evolved it into Power_x and only well after theL architecture had proven itself did IBM start to deploy it onto its maintreamJ products, and only about 10 years alter is IBM confident enough to plan toQ migrate MVS from the 360 architecture to Power (and even that will take a while).r  I Meanwhile, HP committed and forced Compaq to commit to a totally unproven J architecture with different, also untested, paradigm well before that chip2 becomes available. That is a heck of a big gamble.   ------------------------------  # Date: Wed, 10 Jul 2002 19:52:46 GMTo( From: "konabear" <maurert@ameritech.net> Subject: Re: McKinley Cometh...o@ Message-ID: <ig0X8.12127$t%5.5753316@newssvr28.news.prodigy.com>  H IMHO the TEL part of Wintel needs to get to 64 bits.  MS is doubling itsG memory requirements every major release.  It's already in the 128-256MBaJ range.  That's only a couple of notches below 4 GB limit of 32 bits.  I'veE already seen motherboards saying they support 4 GB of main memory butoI because I/O device buffers consume space that the system will only reporte. 3.2-3.8 GB of memory.  Sounds like a VAX 7000!  I If MS is going to let PCs get into video editting in a major way, they'reeG going to need GBs of memory for the editors and thus more than 32 bits.e  J The fact that Windows needs to go there means there will be a major market. for 64 bit chips.  Will IA-64 be the solution?   Todd< "Terry C. Shannon" <terryshannon@attbi.com> wrote in message9 news:CFBW8.337959$6m5.348722@rwcrnsc51.ops.asp.att.net...  >a7 > "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messageg, > news:agenat$gsj$1@pegasus.csx.cam.ac.uk... > >i; > > In article <tRAW8.13228$K_4.329460@twister1.libero.it>,h. > > "Alberto" <uapalbertobu@libero.it> writes: > > |>K > > |> > Well, McKinley may be announced, but there is little evidence that C > > |> > HP are backing up their fine words with buttered parsnips.d > > |> >F > > |> > HP's "buy online" for the USA has only the zx2000 (plus its 6 MercedI > > |> > configurations) and says "There is no stock currently available"S for  > > |> > all of them.- > > |> >I > > |> > Neither the zx2000 nor the zx6000 are in HP's UK list yet.  And,( yes,- > > |> > we are a serious potential customer!L > > |> >H > > |> > If anyone manages to buy one using a normal mechanism, I shouldI > > |> > appreciate hearing what, for delivery when and in which country.g > > |> > > |> You have too hurry :-). > >: > > Perhaps :-)  > >q5 > > |> Itanic performs very well now, but 2 years arer< > > |> necessary to penetrate a full of difficulties market. > > D > > Considering that we have been getting an earful of how wonderfulE > > the imminent IA-64 systems will be for over 5 years now, I am notsD > > sympathetic.  More seriously, this is IA-64's last chance; if it7 > > isn't established within 12-18 months, it will die.s > >t >h& > Sounds like a reasonable assessment. >h >    ------------------------------   Date: 10 Jul 2002 19:58:40 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh...p0 Message-ID: <agi3lg$f0k$1@pegasus.csx.cam.ac.uk>  , In article <3D2C91CE.BC76705B@videotron.ca>,/ JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:= >Sander Vesik wrote: > T >> But HP does not need to wait for intel to decide that ia64 won't make an industryS >> standard - it can reach that conclusion on its own and act accordingly. At leastP" >> in theory it should be able to. >OM >No. Carly bet her job, her carreer and her reputation on this. Remember that=N >prior to the merger, her competence was in question. Remember that during theO >merger pregnancy, most HP staff were against her project. And now that she hasRJ >given birth to the "new" HP, there is a lot of work ahead to prevent WallC >Street from declaring that buying Compaq was a big costly mistake.c   That is certainly true.i  L >It will be quite a while before Carly has enough tenure/power to admit thatM >there were some strategic mistakes such as fofucisng entirely on an unproven  >chip.    E From her behaviour, she has a similar character to Margaret Thatcher.7B If so, she will never admit that she was wrong, not if her refusal" to do so means the collapse of HP.  . >Compare Carly's decisions with those of IBM.  >aO >IBM developped the PowerPC and evolved it into Power_x and only well after thesM >architecture had proven itself did IBM start to deploy it onto its maintream K >products, and only about 10 years alter is IBM confident enough to plan toiR >migrate MVS from the 360 architecture to Power (and even that will take a while).  @ I am sorry, but I was very close to IBM at the time, and that is@ fantasy.  IBM's faults were that of a large bureaucratic, poorlyC communicating organisation, and the PowerPC/POWER shambles was even B more impressive than HP's and Compaq's IA-64 behaviour has been to? date - though not as much as Intel's.  I could go into horrible $ detail, but it is grossly off-group.  J >Meanwhile, HP committed and forced Compaq to commit to a totally unprovenK >architecture with different, also untested, paradigm well before that chipT3 >becomes available. That is a heck of a big gamble.R  D No, that is NOT so.  Until fairly recently, HP was hedging its bets,B and was developing PA-RISC as actively as ever, just more quietly.= And HP did NOT force Compaq into the migration - Capellas hadT> already taken that decision, announced it, disbanded the Alpha> development and sold off some of the IPR and people before the merger with HP started.o  - Please give discredit where discredit is due.f     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679t   ------------------------------  % Date: Wed, 10 Jul 2002 21:53:22 +0100u& From: Ken Green <Ken.Green@kgcc.co.uk> Subject: Re: McKinley Cometh...g* Message-ID: <3D2C9EC2.A439891D@kgcc.co.uk>   konabear wrote:t  J > IMHO the TEL part of Wintel needs to get to 64 bits.  MS is doubling itsI > memory requirements every major release.  It's already in the 128-256MB-L > range.  That's only a couple of notches below 4 GB limit of 32 bits.  I'veG > already seen motherboards saying they support 4 GB of main memory butoK > because I/O device buffers consume space that the system will only report>0 > 3.2-3.8 GB of memory.  Sounds like a VAX 7000! >tK > If MS is going to let PCs get into video editting in a major way, they'resI > going to need GBs of memory for the editors and thus more than 32 bits.e  B Currently video editing uses less RAM than still picture editting.F The quantities of data involved with video are just so big that tryingE to do it in RAM is impractical. For DV 2GB is 9mins of video. So it'stG all done from disk. Besides theres no real benift in having many frames2@ in memory at one time. For stills though 100MBs are nothing, and= you often want several stages of manipulation in RAM at once,e+ otherwise the undo button takes to long :-)u   >> > L > The fact that Windows needs to go there means there will be a major market0 > for 64 bit chips.  Will IA-64 be the solution? >u > Todd   ------------------------------  % Date: Wed, 10 Jul 2002 21:57:01 +0100n& From: Ken Green <Ken.Green@kgcc.co.uk> Subject: Re: McKinley Cometh...n* Message-ID: <3D2C9F9D.BAD6430D@kgcc.co.uk>   Nick Maclaren wrote:  . > In article <3D2C91CE.BC76705B@videotron.ca>,1 > JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:a > >Sander Vesik wrote: > > V > >> But HP does not need to wait for intel to decide that ia64 won't make an industryU > >> standard - it can reach that conclusion on its own and act accordingly. At leastF$ > >> in theory it should be able to. > >eO > >No. Carly bet her job, her carreer and her reputation on this. Remember thatPP > >prior to the merger, her competence was in question. Remember that during theQ > >merger pregnancy, most HP staff were against her project. And now that she has6L > >given birth to the "new" HP, there is a lot of work ahead to prevent WallE > >Street from declaring that buying Compaq was a big costly mistake.t >  > That is certainly true.i >eN > >It will be quite a while before Carly has enough tenure/power to admit thatO > >there were some strategic mistakes such as fofucisng entirely on an unprovenw > >chip. >fG > From her behaviour, she has a similar character to Margaret Thatcher.uD > If so, she will never admit that she was wrong, not if her refusal$ > to do so means the collapse of HP. >e/ > >Compare Carly's decisions with those of IBM.  > >rQ > >IBM developped the PowerPC and evolved it into Power_x and only well after the O > >architecture had proven itself did IBM start to deploy it onto its maintreamrM > >products, and only about 10 years alter is IBM confident enough to plan todT > >migrate MVS from the 360 architecture to Power (and even that will take a while). >eB > I am sorry, but I was very close to IBM at the time, and that isB > fantasy.  IBM's faults were that of a large bureaucratic, poorlyE > communicating organisation, and the PowerPC/POWER shambles was evenlD > more impressive than HP's and Compaq's IA-64 behaviour has been toA > date - though not as much as Intel's.  I could go into horribleI& > detail, but it is grossly off-group. > L > >Meanwhile, HP committed and forced Compaq to commit to a totally unprovenM > >architecture with different, also untested, paradigm well before that chipa5 > >becomes available. That is a heck of a big gamble.e >iF > No, that is NOT so.  Until fairly recently, HP was hedging its bets,D > and was developing PA-RISC as actively as ever, just more quietly.  D The origenal roadmap was altered because I1 was so late, the PA8600,- PA8800 & PA8900 were not on the origenal map.r   >d? > And HP did NOT force Compaq into the migration - Capellas hadl@ > already taken that decision, announced it, disbanded the Alpha@ > development and sold off some of the IPR and people before the > merger with HP started.w >x/ > Please give discredit where discredit is due.  >l
 > Regards, > Nick Maclaren,, > University of Cambridge Computing Service,@ > New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. > Email:  nmm1@cam.ac.uk1 > Tel.:  +44 1223 334761    Fax:  +44 1223 334679M   ------------------------------  # Date: Wed, 10 Jul 2002 21:11:11 GMTe* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh...4A Message-ID: <Pp1X8.69661$Im2.2981540@bin2.nnrp.aus1.giganews.com>-  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message * news:agi3lg$f0k$1@pegasus.csx.cam.ac.uk.... > In article <3D2C91CE.BC76705B@videotron.ca>,1 > JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:e   ...n  L > >Meanwhile, HP committed and forced Compaq to commit to a totally unprovenH > >architecture with different, also untested, paradigm well before that chip5 > >becomes available. That is a heck of a big gamble.2 >rF > No, that is NOT so.  Until fairly recently, HP was hedging its bets,D > and was developing PA-RISC as actively as ever, just more quietly.? > And HP did NOT force Compaq into the migration - Capellas hade@ > already taken that decision, announced it, disbanded the Alpha@ > development and sold off some of the IPR and people before the > merger with HP started.   F While we don't really know (though can perhaps make reasonable guessesJ about) what effect the planned merger had on Alpha's future, Carly & CurlyL have clearly stated that merger talks were *well* under way long before JuneF 25, 2001.  My own suspicion is that if Curly had not been able to lookG forward to an imminent merger to save his job (at least temporarily) heoL might have been considerably less eager to throw away such an important partI of Compaq's profit stream - and also would have had far less incentive toh> avoid competing with Itanic (and HP) at least at the high-end.   - bill   ------------------------------   Date: 10 Jul 2002 21:14:42 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh...d0 Message-ID: <agi842$i98$1@pegasus.csx.cam.ac.uk>  * In article <3D2C9F9D.BAD6430D@kgcc.co.uk>,( Ken Green  <Ken.Green@kgcc.co.uk> wrote: >Nick Maclaren wrote:i >tG >> No, that is NOT so.  Until fairly recently, HP was hedging its bets, E >> and was developing PA-RISC as actively as ever, just more quietly.  > E >The origenal roadmap was altered because I1 was so late, the PA8600, . >PA8800 & PA8900 were not on the origenal map.  : No, they weren't on the MAP, but they assuredly were under= development!  Perhaps only in draft (i.e. on paper etc.), but @ you don't deliver a complex chip within 2 years of starting from scratch.  A Back in 1995, many people said that Intel had shafted HP, becauseu? Intel could survive a seriously late Merced, but HP had stoppede> PA-RISC development.  I said "Maybe but, if I were HP's CEO, I? would be developing PA-RISC quietly in case IA-64 is delayed or  flops."l  @ Well - lo and behold! - HP announced the previously denied 8500,E to ship a mere 6 months after announcement.  And, shortly thereafter, C HP announced a complete roadmap, stretching on to the 8900.  I feel A that I and the previous HP CEO (whatever his name was) would haveS got on, if we had ever met :-)     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679s   ------------------------------   Date: 10 Jul 2002 21:17:44 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh... 0 Message-ID: <agi89o$ian$1@pegasus.csx.cam.ac.uk>  A In article <Pp1X8.69661$Im2.2981540@bin2.nnrp.aus1.giganews.com>, ) Bill Todd <billtodd@metrocast.net> wrote:r6 >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message+ >news:agi3lg$f0k$1@pegasus.csx.cam.ac.uk...t/ >> In article <3D2C91CE.BC76705B@videotron.ca>,e2 >> JF Mezei  <jfmezei.spamnot@videotron.ca> wrote: >eM >> >Meanwhile, HP committed and forced Compaq to commit to a totally unprovenfI >> >architecture with different, also untested, paradigm well before thatl >chip 6 >> >becomes available. That is a heck of a big gamble. >>G >> No, that is NOT so.  Until fairly recently, HP was hedging its bets,aE >> and was developing PA-RISC as actively as ever, just more quietly.n@ >> And HP did NOT force Compaq into the migration - Capellas hadA >> already taken that decision, announced it, disbanded the Alpha A >> development and sold off some of the IPR and people before the  >> merger with HP started. > G >While we don't really know (though can perhaps make reasonable guesses K >about) what effect the planned merger had on Alpha's future, Carly & Curly M >have clearly stated that merger talks were *well* under way long before JuneiG >25, 2001.  My own suspicion is that if Curly had not been able to lookuH >forward to an imminent merger to save his job (at least temporarily) heM >might have been considerably less eager to throw away such an important part J >of Compaq's profit stream - and also would have had far less incentive to? >avoid competing with Itanic (and HP) at least at the high-end.-  B Yes, that is true, but I have good reason to believe that CapellasC has decided to scrap Alpha by the end of 2000, and have some reason.@ to believbe that the negotiations with Intel were well under way in early 2001.     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------  # Date: Wed, 10 Jul 2002 21:33:54 GMTt1 From: "Terry C. Shannon" <terryshannon@attbi.com>h Subject: Re: McKinley Cometh...m. Message-ID: <6L1X8.142711$Uu2.31512@sccrnsc03>  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messagei* news:agi89o$ian$1@pegasus.csx.cam.ac.uk... <snip> > D > Yes, that is true, but I have good reason to believe that CapellasE > has decided to scrap Alpha by the end of 2000, and have some reasonrB > to believbe that the negotiations with Intel were well under way > in early 2001.  / Completely consistent with what I've heard. ;-}    ------------------------------  # Date: Wed, 10 Jul 2002 22:04:28 GMTw* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh... A Message-ID: <Mb2X8.26543$Bt1.1547454@bin5.nnrp.aus1.giganews.com>e  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messagee* news:agi89o$ian$1@pegasus.csx.cam.ac.uk...C > In article <Pp1X8.69661$Im2.2981540@bin2.nnrp.aus1.giganews.com>,m+ > Bill Todd <billtodd@metrocast.net> wrote: 8 > >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message- > >news:agi3lg$f0k$1@pegasus.csx.cam.ac.uk...l1 > >> In article <3D2C91CE.BC76705B@videotron.ca>,14 > >> JF Mezei  <jfmezei.spamnot@videotron.ca> wrote: > >cF > >> >Meanwhile, HP committed and forced Compaq to commit to a totally unprovenK > >> >architecture with different, also untested, paradigm well before thatt > >chip 8 > >> >becomes available. That is a heck of a big gamble. > >>I > >> No, that is NOT so.  Until fairly recently, HP was hedging its bets,lG > >> and was developing PA-RISC as actively as ever, just more quietly.BB > >> And HP did NOT force Compaq into the migration - Capellas hadC > >> already taken that decision, announced it, disbanded the AlphaXC > >> development and sold off some of the IPR and people before ther > >> merger with HP started. > >fI > >While we don't really know (though can perhaps make reasonable guesseswG > >about) what effect the planned merger had on Alpha's future, Carly &  CurlyAJ > >have clearly stated that merger talks were *well* under way long before JuneI > >25, 2001.  My own suspicion is that if Curly had not been able to look J > >forward to an imminent merger to save his job (at least temporarily) heJ > >might have been considerably less eager to throw away such an important partL > >of Compaq's profit stream - and also would have had far less incentive toA > >avoid competing with Itanic (and HP) at least at the high-end.  > D > Yes, that is true, but I have good reason to believe that CapellasE > has decided to scrap Alpha by the end of 2000, and have some reasoniB > to believbe that the negotiations with Intel were well under way > in early 2001.  L Indeed.  But if you examine your statement you'll see that it clearly statesL that he had "announced it, disbanded the Alpha development and sold off someL of the IPR and people before the merger with HP started" - and that's what I
 responded to.h   - bill   ------------------------------   Date: 10 Jul 2002 22:08:25 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh... 0 Message-ID: <agib8p$kdk$1@pegasus.csx.cam.ac.uk>  A In article <Mb2X8.26543$Bt1.1547454@bin5.nnrp.aus1.giganews.com>, ) Bill Todd <billtodd@metrocast.net> wrote: 6 >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message+ >news:agi89o$ian$1@pegasus.csx.cam.ac.uk...  >>E >> Yes, that is true, but I have good reason to believe that Capellas'F >> has decided to scrap Alpha by the end of 2000, and have some reasonC >> to believbe that the negotiations with Intel were well under way  >> in early 2001.  >5M >Indeed.  But if you examine your statement you'll see that it clearly stateshM >that he had "announced it, disbanded the Alpha development and sold off someBM >of the IPR and people before the merger with HP started" - and that's what Ie >responded to.  5 All right, 'started' is a word with many meanings :-)h  @ I certainly don't believe that HP instructed Compaq to scrap theA Alpha, but there is a considerable grey area involving nudges andh
 winks ....     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679i   ------------------------------  # Date: Wed, 10 Jul 2002 22:09:03 GMTr2 From: "Stephen Fuld" <s.fuld@PleaseRemove.att.net> Subject: Re: McKinley Cometh...aG Message-ID: <3g2X8.91907$UT.6070534@bgtnsc05-news.ops.worldnet.att.net>   : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D2C91CE.BC76705B@videotron.ca... >oL > IBM developped the PowerPC and evolved it into Power_x and only well after the D > architecture had proven itself did IBM start to deploy it onto its	 maintream L > products, and only about 10 years alter is IBM confident enough to plan toK > migrate MVS from the 360 architecture to Power (and even that will take at while).   G Has IBM announced a decision to migrate MVS to POWER?  I guess I misseddG that.  Given the huge amount of legacy software, a fair amount of it in E assembler, much dating back decades, this seems like a huge decision.e   --  - Stephen Fuldt,    e-mail address disguised to prevent spam.   ------------------------------   Date: 10 Jul 2002 22:12:12 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: McKinley Cometh...h0 Message-ID: <agibfs$kfm$1@pegasus.csx.cam.ac.uk>  G In article <3g2X8.91907$UT.6070534@bgtnsc05-news.ops.worldnet.att.net>,m1 Stephen Fuld <s.fuld@PleaseRemove.att.net> wrote:  > ; >"JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in messagee' >news:3D2C91CE.BC76705B@videotron.ca...  >>M >> IBM developped the PowerPC and evolved it into Power_x and only well afterm >theE >> architecture had proven itself did IBM start to deploy it onto itsf
 >maintreamM >> products, and only about 10 years alter is IBM confident enough to plan touL >> migrate MVS from the 360 architecture to Power (and even that will take a >while). >aH >Has IBM announced a decision to migrate MVS to POWER?  I guess I missedH >that.  Given the huge amount of legacy software, a fair amount of it inF >assembler, much dating back decades, this seems like a huge decision.  C I have heard it, but not seen definite confirmation.  I can witnessoC that it has been seriously considered from about 1990, and a decade 7 is about what it takes IBM to take a major decision :-)-     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679e   ------------------------------  # Date: Wed, 10 Jul 2002 22:13:50 GMTy1 From: "Terry C. Shannon" <terryshannon@attbi.com>  Subject: Re: McKinley Cometh...o. Message-ID: <yk2X8.143068$Uu2.32410@sccrnsc03>  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messager* news:agib8p$kdk$1@pegasus.csx.cam.ac.uk...C > In article <Mb2X8.26543$Bt1.1547454@bin5.nnrp.aus1.giganews.com>,.+ > Bill Todd <billtodd@metrocast.net> wrote: 8 > >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message- > >news:agi89o$ian$1@pegasus.csx.cam.ac.uk...r > >>G > >> Yes, that is true, but I have good reason to believe that CapellaslH > >> has decided to scrap Alpha by the end of 2000, and have some reasonE > >> to believbe that the negotiations with Intel were well under way  > >> in early 2001.d > >oH > >Indeed.  But if you examine your statement you'll see that it clearly statesJ > >that he had "announced it, disbanded the Alpha development and sold off someH > >of the IPR and people before the merger with HP started" - and that's what I > >responded to. > 7 > All right, 'started' is a word with many meanings :-)l > B > I certainly don't believe that HP instructed Compaq to scrap theC > Alpha, but there is a considerable grey area involving nudges and2 > winks .... >r  H Yep, a lot of what went on will likely remain opaque to us mere mortals.L Still, I believe Alpha's fate was sealed well before any serious acquisition negotiations got underway.   ------------------------------  % Date: Wed, 10 Jul 2002 18:31:00 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: McKinley Cometh...', Message-ID: <3D2CB590.465787A0@videotron.ca>  U Merced was terribly slow, but it was more or less a beta version of the architecture.   L Can one consider McKinley to be "production" quality and judge IA64 based on
 McKinley ?  K If McKinley is now production quality, isn't it fair to state that from nowiM on, IA64 would progress are roughly the same rate as competing chips ? If so,aM its relative position in the pack wouldn't change much over time. It may leadv; for a few months, and then be overtaken by another etc etc.i  N What I don't quite understand is that EPIC should simplify a chip design sinceL it offloads lots of the logic to the compilers, right ? If that is the case,M is it fair to state that an EPIC's chip improvements will come mostly throughh8 clock increases with far less done through chip design ?  N For instance, with a philosophy of having much of the performance logic in theE chip, the Alpha engineers were ahle to add fancy stuff such as branch L prediction and pre-fetching/compilation of both instruction streams after anN IF statement. So these improvements were added to the clock increases to yield an even more performing chip.   N But just how much fancy stuff can be added to IA64 before it is no longer EPICI ? And when IA64 is imporved to the point that it is a RISC chip, won't itiN require that compilers that had been built to do the work that IA64 doesn't doK be modified to become more conventional compilers that allow the chip to dot& much of the performance enhancements ?   ------------------------------  # Date: Wed, 10 Jul 2002 22:21:29 GMTo* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh...i@ Message-ID: <Jr2X8.18484$iX5.979787@bin3.nnrp.aus1.giganews.com>  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messagev* news:agibfs$kfm$1@pegasus.csx.cam.ac.uk...I > In article <3g2X8.91907$UT.6070534@bgtnsc05-news.ops.worldnet.att.net>,k3 > Stephen Fuld <s.fuld@PleaseRemove.att.net> wrote:f   ...e  J > >Has IBM announced a decision to migrate MVS to POWER?  I guess I missedJ > >that.  Given the huge amount of legacy software, a fair amount of it inH > >assembler, much dating back decades, this seems like a huge decision. > E > I have heard it, but not seen definite confirmation.  I can witnessdE > that it has been seriously considered from about 1990, and a decade 9 > is about what it takes IBM to take a major decision :-)r  G I recently encountered a reference to this being targeted for POWER6 (IlI think - *could* have been POWER5), but whether in any really official IBM, context I don't recall.b   - bill   ------------------------------  + Date: Wed, 10 Jul 2002 22:33:35 +0000 (UTC)1A From: "Rupert Pigott" <dark.try-eating-this.b00ng@btinternet.com>e Subject: Re: McKinley Cometh... / Message-ID: <agicnu$es7$1@paris.btinternet.com>c  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message2* news:agi842$i98$1@pegasus.csx.cam.ac.uk... [SNIP]B > Well - lo and behold! - HP announced the previously denied 8500,G > to ship a mere 6 months after announcement.  And, shortly thereafter,ME > HP announced a complete roadmap, stretching on to the 8900.  I feelPC > that I and the previous HP CEO (whatever his name was) would havei  > got on, if we had ever met :-)  C I suspect that he also got news from the "Horses Mouth" quite earlye on too.i   Cheers,  Rupert   ------------------------------  % Date: Wed, 10 Jul 2002 18:38:24 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>m Subject: Re: McKinley Cometh...r, Message-ID: <3D2CB74B.4AD6095E@videotron.ca>   Nick Maclaren wrote:B > can hide it.  And the IA-64 architecture is the most complicated) > I have ever seen, by a factor of three.e  L In what way would the architecture be 3x more complex ? Shouldn't it be muchI simpler since the chip doesn't have to bother with out of order executionr logic etc etc ?   L What "gizmos" does the chip have ?  Does it rely a lot on pipelining ?  does4 it have branch prediction ? out of order execution ?  G Or does it rely on the compiler to funnel each instruction in the rightp execution stream ?  N In other words, what are the differences between EPIC and modern RISC chips in) terms of how instructions are processed ?a  M Is IA64 more complex because it is EPIC, or because it was made by Intel withn" lots of compromises to please HP ?  5 Also, what is different between Merced and McKinley ?c   ------------------------------  # Date: Wed, 10 Jul 2002 22:45:52 GMTD* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh...cA Message-ID: <AO2X8.129931$vq.6535766@bin6.nnrp.aus1.giganews.com>o  3 "konabear" <maurert@ameritech.net> wrote in message,: news:ig0X8.12127$t%5.5753316@newssvr28.news.prodigy.com...J > IMHO the TEL part of Wintel needs to get to 64 bits.  MS is doubling itsI > memory requirements every major release.  It's already in the 128-256MBiF > range.  That's only a couple of notches below 4 GB limit of 32 bits.  F To be a bit more precise, it's a couple of notches below 512 MB - 1 GBL (given the 'doubling' context you chose):  call it 4 - 5 notches below 4 GB.     I'veG > already seen motherboards saying they support 4 GB of main memory buteK > because I/O device buffers consume space that the system will only report 0 > 3.2-3.8 GB of memory.  Sounds like a VAX 7000! >SK > If MS is going to let PCs get into video editting in a major way, they'rePI > going to need GBs of memory for the editors and thus more than 32 bits.n  H You seem to have missed the ability that's existed for at least the lastJ half-decade or so for IA32 processors (and applications thereon) to use upG to 64 GB of physical memory.  While accessing that much physical memoryMC within an application takes a modest amount of additional code, foreL something as specialized as a video editor that's hardly a major impediment.   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 19:05:41 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>. Subject: Re: McKinley Cometh...M, Message-ID: <3D2CBDAF.C0490C36@videotron.ca>   Nick Maclaren wrote:B > fantasy.  IBM's faults were that of a large bureaucratic, poorlyE > communicating organisation, and the PowerPC/POWER shambles was evensD > more impressive than HP's and Compaq's IA-64 behaviour has been to' > date - though not as much as Intel's.d  L Since IBM and POWER are now one of the few options left for those forced off> VMS, having more insight into POWER's story would be welcomed.    ? > And HP did NOT force Compaq into the migration - Capellas hade@ > already taken that decision, announced it, disbanded the Alpha@ > development and sold off some of the IPR and people before the > merger with HP started.d  M The couple are on record to having had merger and product roadmap discussionsOH well before the official announcemnet of Alpha's murder. While there areG strong indications that Capellas had decided long before not to bet hiso[ business on Alpha, it is likely that Carly had an impact on the timing of the announcement.k  N Lets assume that Capellas decided at the time Windows was killed on Alpha thatJ Alpha would be ditched as soon as IA64 was capable of taking over. At thatG time, how come he didn't kill off the EV8 team right way and decided toiL continue to keep and fund them ?  Handing the EV8 team to Intel sooner would8 have helped Intel with the introduction of its new chip.  N So, why did Compaq keep the EV8 team funded and in place even though they knewL that Alpha would be killed ? They probably didn't trust Intel to deliver andM wanted to keep their options opened. Remember that Alpha was starting to takevS off in the high performance area, especially with the successes of celera genomics.,  N So, if Compaq had decided to keep Alpha alive and not take a big leap of faithL until Intel had a real chip available, what happened to make Compaq make theM announcement last year ? Some say first boot of IA64. I say strategic productnN planning with knowledge of what would happen once the merger with HP happened.  L In essence, Curly was made the scapegoat, announcing the bad news , breakingL commitments to customers etc etc, and Carly would then take over the companyH without having to announce any bad news and just stating that they wouldN honour the previously announced roadmap. This way Carly doesn't have any blood
 on her hands.t   ------------------------------  % Date: Wed, 10 Jul 2002 19:13:26 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>e Subject: Re: McKinley Cometh...n, Message-ID: <3D2CBF80.B81CED33@videotron.ca>   Ken Green wrote:D > Currently video editing uses less RAM than still picture editting.  H That is debatable. If you are compositing multiple images into one movieH stream, you start to consume lots of memory. And the the special effectsM applied on a movie are close to if not the same to those phostoshop provides..  M But if you are talking about computer generated images from 3d models, if youeE want production quality, you will need heaps of memory and CPU speed.b   ------------------------------  % Date: Wed, 10 Jul 2002 21:46:48 +0100t& From: Ken Green <Ken.Green@kgcc.co.uk> Subject: Re: McKinley Cometh...i* Message-ID: <3D2C9D37.D2EDA67E@kgcc.co.uk>   Bill Todd wrote:  5 > "Ken Green" <Ken.Green@kgcc.co.uk> wrote in messageo& > news:3D2C01B6.FD9AA4F0@kgcc.co.uk... >t > ...n >vN > > Your right Bill, the L2 is in the same package, but not on the same, chip,I > > altough the tags are on the CPU chip, along with 2 cores, and the two ? > > L1 caches (the data caches are half the size of the PA8700)t >eH > Well, being in the same package does help - and also helps explain whyM > adding additional PA processors to a system is rather expensive.  Apologiesa- > for not being more familiar with the beast.n >     OK,  a quick guide to PA caches.  : Most PA processors only have L1 cache, since PA's pipelineA stores results in the clock cycle after execute, this cache has ap  load to use latency of 2 cycles.  : Upto the PA8200 this cache was external, most PA8200 based> systems had 2MB for data & 2MB for instructions. Beyond 240MHz+ they moved to onchip cache with the PA8500.   @ The PA8500 has 1MB for data and 512K for instruction. The PA8600$ also has the same 1.5MB of L1 cache.  C PA8700 (&PA8700+ not sure of the difference apart from clock speed)g" increases both cache sizes by 50%.  @ The next generation PA8800 is the first main stream PA processorA to bother with L2 cache. Each core has a 768K data and 768K instr 9 cache, they share a 32MB L2 cache with a 40cycle latency.a   >o > >eH > > I would have thought that cost was a major factor in deciding not to > followL > > a similar approach in the McKinley, I guess HP can still afford more forH > > CPUs in their mid to upper range systems than Intel expects the IA64K > > market place to want to pay. I might be wrong, maybe we'll see Makos intA > > the the rp24XX boxes which start with a list price of <$1000.  >rK > I wouldn't hold my breath waiting, given that even the runt-of-the-litter-& > processor chips cost more than that.  K The rp2430 uses a PA8700 in a sub $1000 box, it's not at max speed (650Mhz)e but it's a PA8700.   >e >uM > However, it may be a matter of decreased marginal benefit.  It sounds as if.6 > the PA chips didn't have room for much on-chip cache  K Well going beyond 2.25MB of cache on a chip kinda of makes it quite big :-)c   > (perhaps they do nowG > but there's never seemed to be sufficient gain to modify the originalsM > architecture) so went the MCM route.  3 MB on the chip is at least startingeM > to get respectable, and the 6 MB Madison will offer next year even more so, K > especially given that it's most likely noticeably lower-latency than PA'sh > cache:   The papers talk aboutn$ L1 being 2x 16K with 1 cycle latancy! L2 being 256K with 5cycle latencyr" L3 being 3MB with 12cycles latency  F Sorry I don't know the I2s pipeline, so I'm not sure whether these are! real load to use latency figures.   B > with that much fast cache on the chip, the marginal utility of a* > larger one off-chip may be fairly small. >i  B I believe theres quite a difference between the Power4 SPEC scores? when they're configured with 64MB of L3 cache instead of 128MB.H   >l > >  > > > J > > > The better question is exactly why HP would believe that EPIC (if it > indeedN > > > did revert to being an HP-proprietary product) would be a useful step upM > > > from PA-RISC which, despite relative neglect of late, has just released9 > anI > > > 875 MHz part that should get about 700 SPECint2K performance if theo
 > increaseL > > > is close to linear and next year around Madison time will get a shrink > toN > > > 130 nm (keeping up with Madison in that regard) *plus* dual cores on theN > > > chip (yielding close to double the server performance per die).  There's > noJ > > > evidence that EPIC has left a great deal of performance on the table > justL > > > waiting to be realized in future iterations (though the Alpha guys canJ > > > probably find some if anyone can, or throw it in the trash and start > over),I > > > so whatever degree of effort it would take to keep improving Itanicc > could,@ > > > likely be equally effective in improving good old PA-RISC. > >dI > > EPIC is PA3, it was designed as the replacement for PA-Risc, sure the  > world0K > > has moved on a long way since 1990 when they started, but the PA8700 ise > nothN > > significantly different in terms of what it does from the PA8000. It's not > like > > withM > > Alpha where over the same time frame things like OOO have been added, PA2o > > had them already.j >zJ > Understood.  However, since EPIC has come nowhere near to fulfilling the> > "three times as fast as Alpha!" promises made 'way back then   True  
 > (i.e., it'seF > here, and it's competitive, but no more than that), and since PA hasK > continued to slog ahead to the point where the 875 MHz 8700s shouldn't bed/ > significantly slower at SPECint than McKinley6  F assuming perfect scaling it should score 701 SPECint 2000 & 677 for FP+ so it's still quite some way addrift of I2.?  L For Mako (PA8800) their estimating SPECint 900 & 1000 for FP, compiler gainsK could well take I2 past 900 SPECint by the time Mako ships, the FP score isc way down on I2.h   > (with the shrunk, dual-corenN > 8800s on the way next year to out-perform Madison in such server-style use),J > and since the only paths forward I've seen anyone suggest for EPIC (SMT,L > multi-core dice) apply equally well to the PA architecture, and since EPICJ > seems to be at the bottom of the barrel when it comes to performance per > Watt,e  L I've no idea what the wattage of a PA8700 is, but it's a similar size to theI I2, and all runs at 875Mhz, so I wouldn't expect it to be much lower thana the I2.i  D > I still maintain that it would seem difficult to justify funnelingN > further development dollars into EPIC instead of into PA (and/or Alpha) *if*I > Itanic should turn out to be only an HP-proprietary product rather than5D > become the 'industry standard' its supporters keep hoping it will.  > I'm sure HP expects it to remain/become an "industry standard"   >n >: > >o > > >p > > >cJ > > > Oh, wait:  that *was* the conclusion they reportedly came to a while > ago,5 > > > but couldn't make fly at the corporate level...d > > >0 > > N > > Well the management have well a truely nailed their flag to the IA64 mast.I > > Carley has already said that the whole farm has been bet on IA64, shet7 > > even went as far as buying Compaq to help prove it.d > >oM > > Technical people are easier to convince, generally if you can show enough L > > evidence you can win arguements, they don't like being proved wrong, but > > they can be. > >t3 > > Most corperate levels... won't be proved wrong.t >l > Amen.e >v   :-)H   >t > - bill   Cheers   Ken    ------------------------------  # Date: Wed, 10 Jul 2002 21:58:55 GMTi* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh...rA Message-ID: <z62X8.127805$vq.6494696@bin6.nnrp.aus1.giganews.com>y  3 "Ken Green" <Ken.Green@kgcc.co.uk> wrote in message-$ news:3D2C9D37.D2EDA67E@kgcc.co.uk... > Bill Todd wrote: > 7 > > "Ken Green" <Ken.Green@kgcc.co.uk> wrote in messageu( > > news:3D2C01B6.FD9AA4F0@kgcc.co.uk...   ...B  J > > > I would have thought that cost was a major factor in deciding not to
 > > followJ > > > a similar approach in the McKinley, I guess HP can still afford more foryJ > > > CPUs in their mid to upper range systems than Intel expects the IA64J > > > market place to want to pay. I might be wrong, maybe we'll see Makos inC > > > the the rp24XX boxes which start with a list price of <$1000.e > > : > > I wouldn't hold my breath waiting, given that even the runt-of-the-litter( > > processor chips cost more than that. >iD > The rp2430 uses a PA8700 in a sub $1000 box, it's not at max speed (650Mhz) > but it's a PA8700.  L Apologies again for being PA-knowledge-impaired:  I thought 'Mako' above wasJ a nick-name for McKinley.  Your comment does however make it clear that PAI already provides 64-bit competition for McKinley at a significantly lowerhJ price point, just as SPARC does - which takes considerable wind out of theK sails of those who claim Itanic will win on pure price because other 64-bituD manufacturers (conveniently excluding AMD) can't compete down there.   ...   D > > with that much fast cache on the chip, the marginal utility of a, > > larger one off-chip may be fairly small. > >  >lD > I believe theres quite a difference between the Power4 SPEC scoresA > when they're configured with 64MB of L3 cache instead of 128MB.t  D Ah, but that's comparing off-chip cache to off-chip cache - and at aG particularly interesting pair of granularities, since IIRC no SPEC test?J currently requires more than about 200 MB to run in (and quite a few wouldJ run completely in 128 MB).  POWER4 has about 1.5 MB of on-chip (L2) cache,J which is smaller than most if not all of the SPEC tests, so external cache is still a major win..  K I recognize that as long as cache is well over an order of magnitude fastercK than main memory then cache speed tends to be less critical than cache missuI rate.  But miss rate (at least for truly random data that greatly exceedsdG the cache size) goes down pretty slowly with size increases - at best auJ square-root relationship and ISTR it's even worse - so for at least *some*J values of size and speed a much faster on-chip cache can work as well as aJ slower off-chip one, and 6 MB is a sufficient fraction of 32 MB that whileD it might be *somewhat* slower overall the amount may not be all that significant.   ...r   > > since PA hasJ > > continued to slog ahead to the point where the 875 MHz 8700s shouldn't be1 > > significantly slower at SPECint than McKinleyy >fH > assuming perfect scaling it should score 701 SPECint 2000 & 677 for FP- > so it's still quite some way addrift of I2.   J I don't consider 13% to be a 'significant' difference, given all the otherJ variables associated with system performance and price/performance.  And IE did specifically say SPECint, since that's what's important (at leasti3 compared with SPECfp) to most of the server market.l   > ? > For Mako (PA8800) their estimating SPECint 900 & 1000 for FP,e  H Given that IIRC Mako is a process generation ahead of 8700, that doesn'tI sound like very much of a speed-up (though adjusting to a new process caneK take a bit of time).  And the difference in percentage gain between SPECintt and SPECfp is also intriguing.    compiler gainso= > could well take I2 past 900 SPECint by the time Mako ships,-  I Maybe.  But more to the point I thought Madison was supposed to arrive byMK about then:  while I don't expect the rosiest estimates for its performancegL to materialize, I do expect something like 1100 SPECint from it and possiblyB somewhat more (anything much over 1200 would surprise me, though).  I However, in server use Mako's dual cores (I assume they're still planned) ! should give it a definite leg up.p   ...c  F > > I still maintain that it would seem difficult to justify funnelingK > > further development dollars into EPIC instead of into PA (and/or Alpha)g *if*K > > Itanic should turn out to be only an HP-proprietary product rather thannF > > become the 'industry standard' its supporters keep hoping it will. > @ > I'm sure HP expects it to remain/become an "industry standard"  C I'm sure you're right.  But the context of this discussion was youroJ suggestion that HP would continue Itanic as a proprietary effort if it didI *not* become industry-standard, and that was what I was addressing above.M   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 19:47:20 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>j Subject: Re: McKinley Cometh...s, Message-ID: <3D2CC76F.543A6D4E@videotron.ca>   Nick Maclaren wrote:9 > >Has IBM announced a decision to migrate MVS to POWER? 8  I There was a post yesterday about an article which described the IBM PowereL strategies. And that included moving the 360 (z series) mainframes to Power. (I think it was Power-5).i   ------------------------------  # Date: Wed, 10 Jul 2002 23:56:48 GMTm4 From: David Mosberger-Tang <David.Mosberger@acm.org> Subject: Re: McKinley Cometh...B. Message-ID: <ug65zn16ip.fsf@panda.mostang.com>  K >>>>> On 10 Jul 2002 08:15:45 GMT, nmm1@cus.cam.ac.uk (Nick Maclaren) said:e  6   Nick> In article <ugy9cktx2f.fsf@panda.mostang.com>,>   Nick> David Mosberger-Tang  <David.Mosberger@acm.org> wrote:  F   >>  That would be news to me.  Note that the highest reported SPECfpF   >> number is precisely for Linux (which is a first in and of itself,   >> I  E   Nick> Is it just?  You mean the 1365?  That is MOST interesting, ifh   Nick> I have understood you.  ( A fair amount of detail can be found at:  H  http://www.hp.com/products1/itanium/infolibrary/pdfs/6004_wp_070302.pdf  D On page 8, it shows SPECfp for different platforms.  The 1356 SPECfpC number is for Linux with the Intel compiler on a zx1-based machine.a  ?   >> believe).  Also, PNL seems to be quite happy with Linux on-   >> Itanium 2:-   >> -W   >> http://www.emsl.pnl.gov:2080/capabs/mscf/?/capabs/mscf/hardware/results_hpcs2.html.  A   Nick> Hmm.  The abstract refers to getting 96% of peak on a keydE   Nick> kernel (a matrix multiply).  Sorry, but that is NOT relevant.oF   Nick> Firstly, matrix multiply is trivial to optimise.  Secondly, ifF   Nick> competently coded, it will be dominated by calls to hand-codedF   Nick> DGEMM or ZGEMM anyway.  I don't have time to look at the paper   Nick> now, but will later.  E It is relevant.  The claim was that lack of fp performance would maket= Linux unsuitable for fp-heavy number crunching.  PNL is doingrC precisely that.  And while hand-optimized routines certainly play acF role, PNL most likely would not have bought an Linux Itanium 2 cluster$ without a high-performance compiler.   	--david   ------------------------------    Date: 10 Jul 2002 17:00:56 -0700, From: David Gay <dgay@beech.CS.Berkeley.EDU> Subject: Re: McKinley Cometh...e3 Message-ID: <s71y9cjw2tj.fsf@beech.CS.Berkeley.EDU>h  / JF Mezei <jfmezei.spamnot@videotron.ca> writes: F > IBM developped the PowerPC and evolved it into Power_x and only wellJ > after the architecture had proven itself did IBM start to deploy it onto > its maintream products,   J Some order confusion. PowerPC was derived from Power. Power was used in ofJ IBM's workstation products of the time -- how are you defining "mainstreamH products" ?. I'm sure somebody here can fill in the details on the other Power_n versions.o   --  	 David Gayc dgay@acm.org   ------------------------------  + Date: Thu, 11 Jul 2002 01:21:39 +0000 (UTC)p7 From: dsiebert@excisethis.khamsin.net (Douglas Siebert)d Subject: Re: McKinley Cometh...o* Message-ID: <agimj3$kk$1@sword.avalon.net>  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:e  L >Consider the theoretical possibility of Apple, now having fresh source codeO >for its OS, would revamp it to run 64 bits on the current Power_x chips.  ThatiM >would surely beat IA64's volumes as long as IA64 is considered a high-margin  >product restricted to servers.i    F I don't think this is very theoretical at all.  It is just a matter ofH when the PowerPC G5 is ready.  Itanium seems likely to be only the thirdI highest marketshare for 64 bit machines by this time next year.  Assumingn it can beat Sparc :)   --H Douglas Siebert                          dsiebert@excisethis.khamsin.net  J A good friend will help you move, a true friend will help you move a body.   ------------------------------  # Date: Thu, 11 Jul 2002 01:18:42 GMT1# From: "John Smith" <a@nonymous.com>e Subject: Re: McKinley Cometh...:G Message-ID: <S15X8.7354$6DW1.3504@news01.bloor.is.net.cable.rogers.com>   3 "Ken Green" <Ken.Green@kgcc.co.uk> wrote in messageA$ news:3D2C01B6.FD9AA4F0@kgcc.co.uk... > L > Well the management have well a truely nailed their flag to the IA64 mast.G > Carley has already said that the whole farm has been bet on IA64, shea5 > even went as far as buying Compaq to help prove it.  >vK > Technical people are easier to convince, generally if you can show enoughlJ > evidence you can win arguements, they don't like being proved wrong, but > they can be. >-1 > Most corperate levels... won't be proved wrong.6  ) Even if it sinks the company to prove it.o   ------------------------------  # Date: Thu, 11 Jul 2002 01:23:03 GMTc# From: "John Smith" <a@nonymous.com>n Subject: Re: McKinley Cometh...oG Message-ID: <X55X8.7358$6DW1.2120@news01.bloor.is.net.cable.rogers.com>   5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message * news:agi842$i98$1@pegasus.csx.cam.ac.uk...  B > Well - lo and behold! - HP announced the previously denied 8500,G > to ship a mere 6 months after announcement.  And, shortly thereafter,sE > HP announced a complete roadmap, stretching on to the 8900.  I feel C > that I and the previous HP CEO (whatever his name was) would have   > got on, if we had ever met :-)    
 Lew Platt.   ------------------------------  # Date: Thu, 11 Jul 2002 01:25:23 GMTe# From: "John Smith" <a@nonymous.com>  Subject: Re: McKinley Cometh...aH Message-ID: <785X8.15591$wLk.13760@news01.bloor.is.net.cable.rogers.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D2CBDAF.C0490C36@videotron.ca... >vE > In essence, Curly was made the scapegoat, announcing the bad news ,  breakingF > commitments to customers etc etc, and Carly would then take over the companynJ > without having to announce any bad news and just stating that they wouldJ > honour the previously announced roadmap. This way Carly doesn't have any bloodb > on her hands.h     Out, out, damn spot.   ------------------------------  + Date: Thu, 11 Jul 2002 01:35:07 +0000 (UTC) 7 From: dsiebert@excisethis.khamsin.net (Douglas Siebert)f Subject: Re: McKinley Cometh...w* Message-ID: <agincb$p3$1@sword.avalon.net>  ( Ken Green <Ken.Green@kgcc.co.uk> writes:  M >For Mako (PA8800) their estimating SPECint 900 & 1000 for FP, compiler gainseL >could well take I2 past 900 SPECint by the time Mako ships, the FP score is >way down on I2.    D I'm quite surprised the FP is so low, given that my understanding isF that the PA-8800 will use the McKinley bus instead of the HP's ancientE Runway bus.  McKinley bus is 6.4GB/s, Runway was doubled to 1.9GB/s axE few years ago, but can only realize about 70% of its peak under idealmE conditions, so it is really more like 1.5GB/s.  I never got STREAM todF even exceed 1GB/s on even a PA-8600!  That'll really help SPECfp sing,B since PA has been lacking for some time in the bandwidth dependant' half of SPECfp due to the outdated bus.f  B Of course, HP would have reason to sandbag the Mako numbers a bit,C since they want everyone to go for McKinley.  It is easy convincing C the FP hungry segment of the market to do that, but their bread andyB butter is TPC-type stuff, which is why they've always gone for theE slower clock giant cache type of CPU.  Not the ideal for SPECint, but = works great for TPC.  That 32MB cache will really help there.l  
 <tinfoil hat> F One interesting thing is that I was told by a (now former) HP employeeF that PA-8700 was testing fine at 900MHz, and a few even exceeded 1GHz.E I was quite surprised it was released at only 750MHz.  Now the "+" isiH released, at just under the magic 900MHz number.  Was it a new stepping,C or was it being kept down until McKinley's performance numbers were , known to avoid hurting HP's migration story? </tinfoil hat>   --H Douglas Siebert                          dsiebert@excisethis.khamsin.net  J A good friend will help you move, a true friend will help you move a body.   ------------------------------  # Date: Thu, 11 Jul 2002 01:44:48 GMTu* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: McKinley Cometh... A Message-ID: <sp5X8.17234$iB1.1132349@bin4.nnrp.aus1.giganews.com>   = "Stephen Fuld" <s.fuld@PleaseRemove.att.net> wrote in messageoA news:3g2X8.91907$UT.6070534@bgtnsc05-news.ops.worldnet.att.net...a   ...   I > Has IBM announced a decision to migrate MVS to POWER?  I guess I missedrI > that.  Given the huge amount of legacy software, a fair amount of it innG > assembler, much dating back decades, this seems like a huge decision.t  J Here's an article that purports to have seen and heard such plans directly	 from IBM:e  ) http://news.com.com/2100-1001-919579.htmli  < That's the closest I've seen to any semi-official statement.   - bill   ------------------------------  # Date: Thu, 11 Jul 2002 02:19:46 GMT01 From: "David J. Dachtera" <djesys.nospam@fsi.net>s Subject: Re: McKinley Cometh...h' Message-ID: <3D2CEF8B.18BC60C8@fsi.net>    John Smith wrote:  > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3D2CBDAF.C0490C36@videotron.ca... > >oG > > In essence, Curly was made the scapegoat, announcing the bad news , 
 > breakingH > > commitments to customers etc etc, and Carly would then take over the	 > company L > > without having to announce any bad news and just stating that they wouldL > > honour the previously announced roadmap. This way Carly doesn't have any > bloodn > > on her hands.n >  > Out, out, damn spot.   <music>  WHO LET THE DOGS OUT??!! </music>   -- 1 David J. Dachtera  dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/o   ------------------------------  # Date: Thu, 11 Jul 2002 02:29:02 GMT + From: Anne & Lynn Wheeler <lynn@garlic.com>  Subject: Re: McKinley Cometh...n) Message-ID: <u8z4jhuce.fsf@earthlink.net>n  . David Gay <dgay@beech.CS.Berkeley.EDU> writes:L > Some order confusion. PowerPC was derived from Power. Power was used in ofL > IBM's workstation products of the time -- how are you defining "mainstreamJ > products" ?. I'm sure somebody here can fill in the details on the other > Power_n versions.P  B 801 came out of research. then there was a big project called FortD Knox that would have replaced the numerous m'code engines around theF company (including those used in the low-end and mid-range 370s); thatD got killed. At about this time i believe you started to see some 801; chip engineers showing up working on risc projects at othera companies.    F Then 801 ROMP (Research OPD Micro Processor) project was started to beD a office products displaywriter follow-on with CPr as base operatingB system (and PL.8 language). I believe there was then some analysisC that while a ROMP-based displaywriter was cost effective given enuftF seats sharing the same machine ... that the least expensive ROMP-based> displaywriter was still more expensive that the most expensiveD acceptable displaywriter configuration. That spawned the morphing ofC ROMP/CPr into the PC/RT workstation using the company that had beenp= hired to do the PC/IX port .... doing the port to VMR machine B abstraction layer (which retained some amount of the original PL.8 technology & technicians).  F Then came RIOS/POWER (and RS/6000, as follow-on to PC/RT) ... and thenC somerset project (joint with motorola) for power/pc (aka 601) (also-B involving apple). Up until somerset/powerpc a basic premise of 801E chip designs had been non-cache coherent shared memory multiprocessorzE (actually no multiprocessing except for a special 4-way RSC aka "RIOSnA single chip" implementation that didn't support cache coherency).oA RS/6000 workstations continued on with both RIOS/POWER & POWER/PCuC chipsets (for some time you could tell power from power/pc based onB> whether they supported multiprocessor configurations or not).   D With the as/400 moving to a power/pc chipset ... some things sort of# came full circle back to Fort Knox.   E motorola bought out somerset in '98 ... and ibm came out with chipset4 that was rios/powerpc merge.   27 years of IBM risc:m% http://www.rootvg.net/column_risc.htm   F note the above leaves out the CPr & PL.8 work. it also leaves out fortA knox.  it also leaves out PC/RT using ROMP which was targed as an ? office product division (aka OPD) displaywriter.  also there is @ comment about aix/ps2 in the above. aix for the pc/rt was a at&t8 system v port by the same people that had done the pc/ix? implementation. AIX/PS2 (and its companion AIX/370) was a LocusYE implementation from UCLA. There was also a BSD port to pc/rt from ibm 4 called AOS (that was to the bare metal w/o any vrm).   random url with locus refs: 7 http://plasmid.dyndns.org:81/plasmidweb/joehopfield.htmw  / random past ROMP, somerset, fort knox postings:rX http://www.garlic.com/~lynn/98.html#26 Merced & compilers (was Re: Effect of speed ... )X http://www.garlic.com/~lynn/98.html#27 Merced & compilers (was Re: Effect of speed ... )1 http://www.garlic.com/~lynn/99.html#66 System/1 ?r@ http://www.garlic.com/~lynn/99.html#129 High Performance PowerPCG http://www.garlic.com/~lynn/99.html#136a checks (was S/390 on PowerPC?)cZ http://www.garlic.com/~lynn/2000.html#59 Multithreading underlies new development paradigmG http://www.garlic.com/~lynn/2000b.html#54 Multics dual-page-size scheme0q http://www.garlic.com/~lynn/2000d.html#60 "all-out" vs less aggressive designs (was: Re: 36 to 32 bit transition)hI http://www.garlic.com/~lynn/2001c.html#84 database (or b-tree) page sizessA http://www.garlic.com/~lynn/2001f.html#43 Golden Era of Compilersv= http://www.garlic.com/~lynn/2001g.html#23 IA64 Rocks My Worldpd http://www.garlic.com/~lynn/2001h.html#69 Very CISC Instuctions (Was: why the machine word size ...)> http://www.garlic.com/~lynn/2001i.html#28 Proper ISA lifespan?> http://www.garlic.com/~lynn/2001j.html#37 Proper ISA lifespan?U http://www.garlic.com/~lynn/2002c.html#40 using >=4GB of memory on a 32-bit processorhK http://www.garlic.com/~lynn/2002g.html#12 "Soul of a New Machine" Computer? K http://www.garlic.com/~lynn/2002g.html#14 "Soul of a New Machine" Computer?t? http://www.garlic.com/~lynn/2002g.html#17 Black magic in POWER5 K http://www.garlic.com/~lynn/2002g.html#39 "Soul of a New Machine" Computer? @ http://www.garlic.com/~lynn/2002g.html#70 Pipelining in the past< http://www.garlic.com/~lynn/2002h.html#19 PowerPC Mainframe?@ http://www.garlic.com/~lynn/2002h.html#63 Sizing the application       -- eH Anne & Lynn Wheeler   | lynn@garlic.com -  http://www.garlic.com/~lynn/    ------------------------------  + Date: Thu, 11 Jul 2002 05:09:47 +0000 (UTC) A From: "Rupert Pigott" <dark.try-eating-this.b00ng@btinternet.com>m Subject: Re: McKinley Cometh... / Message-ID: <agj3uq$mhh$1@paris.btinternet.com>&  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:yk2X8.143068$Uu2.32410@sccrnsc03... >u7 > "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message , > news:agib8p$kdk$1@pegasus.csx.cam.ac.uk...E > > In article <Mb2X8.26543$Bt1.1547454@bin5.nnrp.aus1.giganews.com>,i- > > Bill Todd <billtodd@metrocast.net> wrote:n: > > >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message/ > > >news:agi89o$ian$1@pegasus.csx.cam.ac.uk...u > > >>I > > >> Yes, that is true, but I have good reason to believe that CapellasaJ > > >> has decided to scrap Alpha by the end of 2000, and have some reasonG > > >> to believbe that the negotiations with Intel were well under way  > > >> in early 2001.c > > > J > > >Indeed.  But if you examine your statement you'll see that it clearly > statesL > > >that he had "announced it, disbanded the Alpha development and sold off > someJ > > >of the IPR and people before the merger with HP started" - and that's > what I > > >responded to. > >i9 > > All right, 'started' is a word with many meanings :-)r > >sD > > I certainly don't believe that HP instructed Compaq to scrap theE > > Alpha, but there is a considerable grey area involving nudges andt > > winks .... > >n >sJ > Yep, a lot of what went on will likely remain opaque to us mere mortals.B > Still, I believe Alpha's fate was sealed well before any serious acquisitionM > negotiations got underway.  5 My wild guess is that it was just the same process as 6 the slice & dice that went on leading up to the Compaq merger.n   Cheers,. Rupert   ------------------------------  % Date: Wed, 10 Jul 2002 14:24:00 -0400M- From: JF Mezei <jfmezei.spamnot@videotron.ca>84 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2C7BBC.CB05E2B7@videotron.ca>   "Terry C. Shannon" wrote:eK > Now that's a really great whine. Company puts together a training session-# > (pilot program) and people bitch.2 > 4 > Company puts together a roadshow and people bitch.  E Please define "company". I see a big difference between "Company" anda "Department".   K Folks are very aware that the VMS department is working wonders to do theireM best to promote VMS *despite* the "Company"  (Digital, Compaq, HP or whateverr	 is next).D  N I have absolutely no expectation that HP will promote VMS. I have expectationsL that HP will prevent the VMS department from promoting itself too much. As aK result, I have no expectations of seing easy access to VMS information fromM the Corporate HP web site.   ------------------------------  # Date: Wed, 10 Jul 2002 19:09:38 GMTh5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>44 Subject: Re: McKinley tops SpecFP AND SpecInt charts2 Message-ID: <SD%W8.25$HB6.728133@news.cpqcorp.net>  I No why don't *you* leave comp.os.vms to those of us who still care, afterNB all *I* still have something to do with VMS, and you are just someE disgruntled hack.  Your response is typical of an egomaniacal gas bagsJ leaking.  You want to go and try and impress people in comp.arch - have atG it - not that I know what your involvement in computer architecture is.e  I Nothing, and I mean *nothing* you have posted in a very long time has had C anything of substantive value to offer.  Just negative opinions andcJ rantings.  Shouting down and insulting anyone who might disagree.  Here weL have some evidence that at least some of your authoratative claptrap is just that.o     Bill Todd wrote in message ... >rA >"Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in messagei, >news:NpXW8.8$Vs6.331862@news.cpqcorp.net...K >> Oooh.  Boy am I offended comming from you.  Responding to "specifics" ofuI >> your speculations, bad mouthing, and rabble rousing is a waste of time- > L >The typical response of one who has nothing substantive to offer in return. >Fuck off, Fred. >e >- bill. >  >s >-   ------------------------------  % Date: Wed, 10 Jul 2002 15:40:39 -0400i- From: JF Mezei <jfmezei.spamnot@videotron.ca>r4 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2C8DAE.9935E02C@videotron.ca>   John Reagan wrote:D > Itanium support.  Some C and C++ front-end folks were also sold toG > Intel.  Front-end folks (like me) for other compilers (Pascal, COBOL,e) > BASIC, etc.) all remained at Compaq/HP.c  N What was the rationale behind donating only "some" of the C and C++ to Intel ?N How much is left at "Digital" for C and C++ ? Just maintenance or still enough% to develop the compilers a bit more ?p  I > 3) Given that Intel doesn't have a BLISS compiler :-), it made sense tonJ > enhance GEM to support an OpenVMS Itanium target (the GEM developers did' > this work before departing to Intel).n  M Now that they were shipped to Intel, is the GEM support for VMS IA64 "mature"sH with no more development, or will Intel continue to develop it for VMS ?  G Just how different is GEM for VMS versus GEM for another OS on the samei architecture ?  H > COBOL, Basic), we think this is adequate for the long-term.  For C, inJ > the short-term (to aid porting of OpenVMS and to provide the ultimate inG > source compatibility for the customers), we have a Compaq C front-endnG > hooked to the OpenVMS Itanium GEM.  In the longer term, we are moving  > towards using Intelr    K Once you have DEC-C ported to IA64, using the GEM-VMS that was also ported,hM and if you retain some of the DEC-C engineers at the "Digital" portion of the2K company, what would really be the advantages of dumping DEC-C and going foro	 Intel-C ?e  L Will Intel-C use the same GEM on VMS as DEC-C will/is using ? If that GEM isN "mature" for VMS, doesn't it mean that Intel-C will be as stagnant as DEC-C onK IA64 in terms of fancy machine code generation to make best use of the EPICA thing ?      Thanks for the explanations.   ------------------------------  % Date: Wed, 10 Jul 2002 15:50:19 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>e4 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2C8FF2.CA93D897@videotron.ca>   WILLIAM WEBB wrote:, > 3 > Did it occur to anyone that, when small steps aret2 > taken in a positive direction, complaining about3 > the size of the steps doesn't improve the chancese, > for additional movement in that direction?  7 Foole me one, shame on you. Fool me twice, shame on me.   J After a decade of similar techniques used to shut up the naysayers only toL revert back to the "Digital is killing VMS" philosophy a little while later,L nobody should be saying "HP has finally changed its philosophy" just because: the word "OpenVMS" was seen in a couple of press releases.  L 6 months from now, if there has been a consistent mention of VMS in HP pressC releases and if during that time, the exposure of VMS has increased 0 consistently, then yeah, kudos will be in order.  N But HP must understand that VMS customers have been lied to so many times thatN it takes time to rebuild trust and during that time, it is HP's responsability[ to be patient before customers come to believe this is a sincere and permanent improvement.e   ------------------------------  % Date: Wed, 10 Jul 2002 16:04:07 -0400n* From: WILLIAM WEBB <WWEBB1@email.usps.gov>4 Subject: RE: McKinley tops SpecFP AND SpecInt charts- Message-ID: <0033000071966660000002L002*@MHS>   4 =0AI can get sarcastic with the best of them, bill-- seppuku and france.u  3 If my statements sound like unqualified enthusiasm,o3 perhaps it's what happens when they are compared tob) the Eeyore-like pronouncements of others..  4 My objective in this forum has been to publicly laud3 positive actions taken by HP since the acquisition..  > God knows we've had enough "Imminent Death of VMS- Film at 11"/ threads to last a whole lot of microfortnights.   0 My objective in email to folks at HP has been to6 highlight past bad decisions and suggest remediation--( not to call individuals idiots or worse.  1 It has been my experience that you get taken mored0 seriously when you discuss both the positive and negatives of a given issue.n  3 I won't publish what anyone at HP has said to me ind5 reply emails-- except to say that I've not been givenB the secret game plans.  8 I felt that HP was completely baffled by the uproar that8 resulted from their initial pronouncements about VMS and9 the fact that folks were doing everything but diagrammingn: sentences and reading chicken entrails every time anything got said about VMS--  : so I sat down and wrote a brief history of actions wrt VMS; from the GQBob era forward, and discussed how many of theseg9 were perceived in a less-than-favorable light by a goodlyr portion of the user community.  ; Among the things I've discussed at one time or another are:r@ the (mis)use of VMS/NSK profits to prop up marginally profitable= WIntel sales; the perception that VMS was in maintenance model? and the urgent necessity of taking major steps to overcome thataA perception; the ever-present application availablity problem; and  the ever-present .edu problem.  ? Nothing we haven't discussed here in great detail; I just wrotee a condensed version.  ; I've also suggested that they mine the archives for all thee@ excellent suggestions that have been posted here by so many good folks.  ' And, again I say, they *are* listening.r  ! Large ships change course slowly.d  / Am I the only one who senses a sea change here?-  C A change that might just be a willingness to loosen the constraintsn, under which VMS has been placed in the past?  7 This is just opinion--  I think that they've decided to-< change course, but they haven't decided how far to turn yet.  C And, as I've said more than once, polite and reasoned feedback willg; be listened to and, in my experience, will receive a reply.    WWWebb   -----Original Message-----/ From: Info-VAX-Request@Mvb.Saic.Com at INTERNET0& Sent: Wednesday, July 10, 2002 1:31 PMB To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET4 Subject: RE: McKinley tops SpecFP AND SpecInt charts    7 "WILLIAM WEBB" <WWEBB1@email.usps.gov> wrote in message ' news:0033000071918656000002L062*@MHS...g  1 Did it occur to anyone that, when small steps arev0 taken in a positive direction, complaining about1 the size of the steps doesn't improve the chancest* for additional movement in that direction?    E Yes.  And it probably also occurred to most of those same people thatuH expressing unqualified enthusiasm rather than saying "That's a good sta= rt -H now here's what you *really* need to do" might be a good way to ensure = that+ those small steps were the last ones taken.@   - bill=a   ------------------------------  # Date: Wed, 10 Jul 2002 20:48:07 GMTn& From: John Reagan <john.reagan@hp.com>4 Subject: Re: McKinley tops SpecFP AND SpecInt charts% Message-ID: <3D2C9BEE.9060700@hp.com>    JF Mezei wrote:  > John Reagan wrote: > D >>Itanium support.  Some C and C++ front-end folks were also sold toG >>Intel.  Front-end folks (like me) for other compilers (Pascal, COBOL, ) >>BASIC, etc.) all remained at Compaq/HP.  >  > P > What was the rationale behind donating only "some" of the C and C++ to Intel ?P > How much is left at "Digital" for C and C++ ? Just maintenance or still enough' > to develop the compilers a bit more ?- >-  @ I wasn't involved in the decision making.  I can't speak to any H rationale. :-)  As to "how much" is left, I really can't say (and don't C really want to say).  We will do whatever is needed to get quality .= compilers in place with the features needed by the customers.e   > I >>3) Given that Intel doesn't have a BLISS compiler :-), it made sense to-J >>enhance GEM to support an OpenVMS Itanium target (the GEM developers did' >>this work before departing to Intel).e >  > O > Now that they were shipped to Intel, is the GEM support for VMS IA64 "mature"xJ > with no more development, or will Intel continue to develop it for VMS ? >u  @ We didn't sell GEM to Intel.  GEM is still a Compaq/HP piece of I software.  Intel doesn't have any vested interest in developing GEM.  As sI for "mature" or not, the remaining GEM developers are committed to doing  I whatever is needed to GEM to make sure it is adaquate for new chips like   Itanium 2 and beyond.   I > Just how different is GEM for VMS versus GEM for another OS on the sameh > architecture ?  H Some pieces identical.  Some pieces very different.  To give specifics, I lets compare GEM for OpenVMS Alpha and GEM for Tru64 Alpha...  Certainly iF the instruction set is the same as well as most of the optimizations. F However, calling sequences, object file format, debug info format, OS G support for exceptions and speculation, etc. all add up to significant o& differences between 2 Alpha platforms.   >  > H >>COBOL, Basic), we think this is adequate for the long-term.  For C, inJ >>the short-term (to aid porting of OpenVMS and to provide the ultimate inG >>source compatibility for the customers), we have a Compaq C front-endaG >>hooked to the OpenVMS Itanium GEM.  In the longer term, we are moving  >>towards using Intelo >  >  > M > Once you have DEC-C ported to IA64, using the GEM-VMS that was also ported,pO > and if you retain some of the DEC-C engineers at the "Digital" portion of the M > company, what would really be the advantages of dumping DEC-C and going foru > Intel-C ?o  E One possible advantage is that while the Itanium target for GEM will pI generate adequate code, the Intel Itanium code generator may out perform  H it in several areas.  Moving towards the Intel compiler would provide C B and C++ users with better code quality.  This area is still under  investigation.   > @ > Will Intel-C use the same GEM on VMS as DEC-C will/is using ?   I Goodness no.  The Intel C/C++ compiler uses Intel's code generator.  See o above.       -- n John Reaganl' Compaq Pascal/{A|I}MACRO Project Leadere Hewlett-Packard Companyd   ------------------------------  # Date: Wed, 10 Jul 2002 19:19:35 GMTn* From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: McKinley tops SpecFP AND SpecInt charts@ Message-ID: <bN%W8.12769$iB1.798055@bin4.nnrp.aus1.giganews.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D2B92F0.D8873850@videotron.ca... > Bill Todd wrote:J > > All the above ignores pricing issues (which work to Itanic's advantageJ > > against the usual RISC suspects, with the possible exception of SPARC, butw= > > work against Itanic in comparisons with IA32 and Hammer).  >s. > Does the $4000 Itanic outperform its peers ?  1 That's not as simple a question as it might seem.   I First of all, the context of the post to which you responded was not justnL this instant in time but the near- and not-so-near-term future.  McKinley isJ about to appear (despite its launch, general availability does not seem toE exist yet), 1.25 GHz Alphas are about to appear, enhanced-performanceo$ low-cost USIIIis are about to appearG  http://www.aceshardware.com/read.jsp?id=50000244 ), faster POWER4s aresK about to appear, faster IA32 platforms are *constantly* popping up, EV7 andoK Hammer will appear before year's end, ...  And that's just for *this* year./  L Second, despite their efforts to be consistent, benchmarks sometimes aren't.E For example, the good $/tpmC figures HP posted for McKinley come from . unspecified specially-discounted system pricesD  http://www.theregus.com/content/3/25530.html ) that may not reflect0 configurations customers will be likely to want.  J Third, Itanic performance may depend more heavily than usual on the systemL it's embedded in - e.g., HP is claiming its chip sets will offer a 15% - 20%' advantage over competing configurationsg0  http://www.theregus.com/content/3/25532.html ).  G Fourth, what's a 'peer' to Itanic?  It might not be too unreasonable toWG exclude IA32 platforms (though they will be vigorous competitors in theoJ low-end 'standard' server space that was originally supposed to create theK volume for Itanic that would move it toward 'commodity'-level pricing), buti Hammer definitely will qualify.t  J All that's really clear is that McKinley is *very* performance-competitiveJ (unlike its predecessor) but already trails Pentium 4 in SPECint and as ofJ October should slightly trail both Alpha and POWER4 in SPECint (since theyG both are scheduled for speed-bumps shortly) and likely POWER4 in SPECfpuA (with Alpha coming in a not-too-distant 3rd).  POWER4 retains itsaF significant dual-core advantage in server use (i.e., in a chip-to-chipJ server comparison it remains clearly ahead).  If EV7 appears before year'sH end, it should regain the lead fairly decisively in both SPEC benchmarksG unless Hammer bests it in SPECint (but the dual-core POWER4 will remainnJ ahead in a chip-to-chip server-performance comparison).  If USIIIi appearsI soon as scheduled, it should narrow the gap with McKinley but still trail 9 it, as should the recently-released 875 MHz PA-RISC 8700.d  G Beyond this year, most of the other architectures (with the conspicuousv? exception of Alpha) have additional enhancements (beyond simpleiI process-shrinks) planned whereas Itanic does not.  This seems unlikely to'# improve Itanic's relative position.    > I > If Intel publicly says the 3meg chip sells for $4000, what would be the J > ballpark prince that HP would actually pay for it ? (considering all the dealsI > it and Compaq did with Intel)l >iK > If volumes do not pick up for that enterprise-only chip, isn't it fair to ; > think that the prices for the chip aren't going to drop ?t >e1 > At that cost, is IA64 really that competitive ?t  H HP has little incentive to make Alphas or PAs competitive cost-wise withE Itanic, so in those two cases the answer may be 'yes'.  IBM might actwK similarly except that they seem to be going after Sun, and Sun's pricing is J very competitive indeed, so we'll just to keep an eye on POWER4.  I really2 don't know what SGI is doing with MIPS these days.    Seems to me that there is no-( > way it could compete against the 8086.  L Or Hammer.  Or, possibly, USIIIi if Sun is really serious about playing down there.   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 14:53:19 -0700"0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com>4 Subject: Re: McKinley tops SpecFP AND SpecInt charts+ Message-ID: <3D2C4A5F.F648C55@Mvb.Saic.Com>t   Nick Maclaren wrote: > . > In article <3D2B3BFA.BC5ADFD8@videotron.ca>,1 > JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:o > >Nick Maclaren wrote:hE > >> Then I suggest that you quote performance figures based on them,dD > >> and do not flame me when I point out that the figures quoted by@ > >> HP on Monday are not what most McKinley customers will get. > >eO > >Since IA64 is currently still poised to be mostly HP's chip, much like Alpha P > >was Digital's stating that most IA64 customers would be running HP's compiler > >might not be wrong. > A > HP's IA-64 announcement referred to 3 operating systems: HP-UX,3@ > Linux and some sort of Microsoft product.  It has at least twoA > others under development: VMS and NSK.  Microsoft has said that%> > it is developing .NET.  As far as I know, HP's compilers are< > available only for HP-UX, and that is not going to change. > @ > In particular, this is comp.os.vms, and VMS has been stated to? > be planned NOT to use HP's compiler.  I think that I am being.< > fairly reasonable in pointing out that performance figures@ > obtainable only with HP's compilers aren't what customers will > get. >   F Well, let's see.  Of the benchmarks done: (data taken from Intel's web page)n  ? Transaction Processing: work done at Intel, O.S. not specified.   ) SAP 2-tier SD: Windows Advanced Server LE    Security Performance: '    Coradiant Secure Transactions: Linux     SPECweb99_SSL: HP-UXp   Linpack: HP-UX  & System Memory Bandwidth: not specified  = Mechanical Computer Aided Engineering (Nastran): Windows .NETg   SPECint_base2000: HP-UX2   SPECfp_base2000: Linux  E It certainly doesn't look to me like only HP compilers were used.  In G fact, I haven't found anything yet that says what compilers were used. -F So what is the basis for your claim that these performance figures areC only avaiable with HP compilers (and, by extension, only on HP-UX)?-  
 Mark Berryman- Mark.Berryman@Mvb.Saic.Com   ------------------------------   Date: 10 Jul 2002 22:04:07 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren)4 Subject: Re: McKinley tops SpecFP AND SpecInt charts0 Message-ID: <agib0n$k4f$1@pegasus.csx.cam.ac.uk>  + In article <3D2C4A5F.F648C55@Mvb.Saic.Com>,e2 Mark Berryman  <Mark.Berryman@Mvb.Saic.Com> wrote: >Nick Maclaren wrote:, >> )B >> HP's IA-64 announcement referred to 3 operating systems: HP-UX,A >> Linux and some sort of Microsoft product.  It has at least twogB >> others under development: VMS and NSK.  Microsoft has said that? >> it is developing .NET.  As far as I know, HP's compilers area= >> available only for HP-UX, and that is not going to change.t >> pA >> In particular, this is comp.os.vms, and VMS has been stated top@ >> be planned NOT to use HP's compiler.  I think that I am being= >> fairly reasonable in pointing out that performance figuresrA >> obtainable only with HP's compilers aren't what customers willi >> get.  > G >Well, let's see.  Of the benchmarks done: (data taken from Intel's web, >page) >h >SPECint_base2000: HP-UX >e >SPECfp_base2000: Linuxr >eF >It certainly doesn't look to me like only HP compilers were used.  InH >fact, I haven't found anything yet that says what compilers were used. G >So what is the basis for your claim that these performance figures are D >only avaiable with HP compilers (and, by extension, only on HP-UX)?  @ We were talking about the Spec figures.  I have so far failed to@ find this page referred to, but two of you have said that SpecFPC was produced on Linux.  However, the figure that is most indicativeIB of CPU performance of general applications is SpecInt, and that is@ the one that is believed to be 20% lower with Intel's compilers.     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 3346790   ------------------------------  # Date: Wed, 10 Jul 2002 22:19:00 GMT * From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: McKinley tops SpecFP AND SpecInt charts@ Message-ID: <op2X8.18423$iX5.977067@bin3.nnrp.aus1.giganews.com>  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messagee* news:agib0n$k4f$1@pegasus.csx.cam.ac.uk...   ...d  B > We were talking about the Spec figures.  I have so far failed toB > find this page referred to, but two of you have said that SpecFPE > was produced on Linux.  However, the figure that is most indicativelD > of CPU performance of general applications is SpecInt, and that isB > the one that is believed to be 20% lower with Intel's compilers.  F AFAIK HP's is the only SPECint figure yet made public (though an IntelK figure of '760+' was leaked a few days ago) - and while there has been some L speculation about the relative merit of HP's compilers vs. others I have notD seen anything quantitative (if you have a source I'd be interested).  2 But HP themselves have asserted (last paragraph ofK http://www.theregus.com/content/3/25532.html ) that they believe their chipoJ sets give them a 15% - 20% performance advantage - so (among other things)J knowing whose chip sets were used for the many other performance figures I# clipped above might be interesting.u   - bill   ------------------------------  # Date: Wed, 10 Jul 2002 22:37:56 GMT0* From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsA Message-ID: <8H2X8.70743$Im2.3051998@bin2.nnrp.aus1.giganews.com>a  @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message, news:SD%W8.25$HB6.728133@news.cpqcorp.net...K > No why don't *you* leave comp.os.vms to those of us who still care, after D > all *I* still have something to do with VMS, and you are just someG > disgruntled hack.  Your response is typical of an egomaniacal gas bagfL > leaking.  You want to go and try and impress people in comp.arch - have atI > it - not that I know what your involvement in computer architecture is.t  I That appears to be consistent with your general level of knowledge, Fred.i  G If I didn't care, I wouldn't be here.  And if I didn't say things worthi listening to, no one would.s   >aK > Nothing, and I mean *nothing* you have posted in a very long time has hadl) > anything of substantive value to offer.   L How would you know?  You aren't willing to engage in the discussion it wouldG take to determine it (assuming it's not clear to you at first reading).   F We have different viewpoints, and are both angry (though for differentK reasons).  The other difference between us is that I substantiate my views,sJ but you just disparage them while refusing to present anything substantiveI in the way of rebuttal.  In other words, while you probably have areas inrE which you're technically competent, in these particular waters you'veeB presented nothing to suggest any competence whatsoever (rather the
 opposite).     Just negative opinions andC > rantings.  Shouting down and insulting anyone who might disagree.i  E No, just those who disparage without offering anything substantive ins return:  they deserve it.;  	   Here weeI > have some evidence that at least some of your authoratative claptrap iso just > that.e  K Really?  Since I've always been very careful in attributing the bases of myaJ extrapolations - usually to public statements made by Intel (in this case;H Compaq in many others) - I'd say you just haven't (as usual) been payingJ enough attention.  If HP's claim that their chip sets are 15% - 20% betterL than the competition applies to Intel's chip sets, then guess what?  Intel'sG SPECint performance will wind up comfortably within the range I've beeneH predicting for it, and be entirely consistent with their statements upon  which I based those predictions.  L Or not:  we'll see, if and when SPECint figures for Intel chip sets (and forH that matter HP's as well) appear at spec.org.  But that won't change theD reasonableness of the predictions I made based on the Intel-provided material available at the time.h  D So I'll say it yet again, Fred:  step up to the plate with something substantial, or fuck off.*   - bill   ------------------------------  # Date: Wed, 10 Jul 2002 20:58:48 GMT * From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsA Message-ID: <ce1X8.126071$vq.6447346@bin6.nnrp.aus1.giganews.com>   7 "WILLIAM WEBB" <WWEBB1@email.usps.gov> wrote in messagee' news:0033000071966660000002L002*@MHS...o  1 I can get sarcastic with the best of them, bill--" seppuku and france.t  3 If my statements sound like unqualified enthusiasm,t3 perhaps it's what happens when they are compared to ) the Eeyore-like pronouncements of others.m   ***RJ What your statement that I responded to sounded like was a suggestion thatE the insignificant (though unexpected) recognition HP has accorded VMSnL recently should be sufficient to silence criticism.  If you meant otherwise,  you failed make it at all clear. ***l  4 My objective in this forum has been to publicly laud3 positive actions taken by HP since the acquisition.t   ***oK And my objective since last June has been to highlight the continuing loadseD of manure Compaq and its cronies insist on spreading here.  Both are  reasonable and useful endeavors. ***-  > God knows we've had enough "Imminent Death of VMS- Film at 11"/ threads to last a whole lot of microfortnights.e   ***nJ Gee, I wonder why?  Such things usually have *some* reason for occurring -L and given that you've chosen to grossly exaggerate their nature, I'd say theI discussion that has actually occurred (concern that VMS development wouldeG languish, customers get increasingly milked, and use dwindle over time,eI rather than the instantaneous demise your comment suggests) was more thane
 justified. ***e  0 My objective in email to folks at HP has been to6 highlight past bad decisions and suggest remediation--( not to call individuals idiots or worse.   *** I That's nice, but the subject here is not email to HP but conversations indL c.o.v. - and both what's appropriate to discuss and how to discuss it differ  radically between the two media. ***v  1 It has been my experience that you get taken moree0 seriously when you discuss both the positive and negatives of a given issue.e   ***MK Unfortunately, that generally hasn't happened - and those on the 'positive'eK side are not one whit less responsible for this situation than those on thepK 'negative' side.  It's called polarization, and once it's occurred it tendss to be self-perpetuating.  L Since the instigating force behind this has been the long history of neglectH of VMS (and of course Alpha) that culminated in the Alphacide, and sinceI *no* moves have yet been made to compensate in any measurable way for thehE damage this has done, and since VMS's owners have yet to deviate fromoH treating VMS's customers like mushrooms (kept in the dark and fed shit -L *especially* in the garbage promulgated by Compaq and its minions after lastH June), I'd say it was up to the 'positive' side to move to the center ifF they want to break this deadlock:  if they do so, I and I suspect manyK others will be more than happy to meet them there and discuss the situation"L on its merits - though until those merits themselves change the conversationK remains likely to be uncomfortable for those who would prefer that everyonei> just forget about the past and be optimistic about the future. ***e  3 I won't publish what anyone at HP has said to me ina5 reply emails-- except to say that I've not been givend the secret game plans.  8 I felt that HP was completely baffled by the uproar that8 resulted from their initial pronouncements about VMS and9 the fact that folks were doing everything but diagrammingr: sentences and reading chicken entrails every time anything got said about VMS--  : so I sat down and wrote a brief history of actions wrt VMS; from the GQBob era forward, and discussed how many of these39 were perceived in a less-than-favorable light by a goodly: portion of the user community.  ; Among the things I've discussed at one time or another are:1@ the (mis)use of VMS/NSK profits to prop up marginally profitable= WIntel sales; the perception that VMS was in maintenance mode ? and the urgent necessity of taking major steps to overcome thattA perception; the ever-present application availablity problem; ands the ever-present .edu problem.  ? Nothing we haven't discussed here in great detail; I just wrote  a condensed version.  ; I've also suggested that they mine the archives for all thei@ excellent suggestions that have been posted here by so many good folks.   ***sK And all those things are great.  I wouldn't have had the optimism to expectMG such efforts to be worth making (since everyone at Marcello's level andsI below should already be well-acquainted with reality), and quite possiblyfK neither the patience to state them unconfrontationally in any event, but ifrB you did then that justifies your efforts - and more power to them. ***p  ' And, again I say, they *are* listening.n  ! Large ships change course slowly.e  / Am I the only one who senses a sea change here?d  C A change that might just be a willingness to loosen the constraints , under which VMS has been placed in the past?   ***KH Am I the only one who senses something very much like the purported 'sea< change' that was called the 'VMS renaissance' two years ago? ***g  7 This is just opinion--  I think that they've decided to < change course, but they haven't decided how far to turn yet.   ***bK And some people thought exactly the same thing two years ago.  Which is onegK reason many people (including some of that former group of optimists) won'tmI be convinced this time until they *see* change, and substantive change ate that.x ***   C And, as I've said more than once, polite and reasoned feedback willn; be listened to and, in my experience, will receive a reply.a   *** L That wasn't *ever* a problem at Marcello's level and below, and never made a bit of difference before.i  H While it certainly won't hurt (and might help) for anyone so inclined toJ provide such feedback, I suspect what most people here want is not a replyL but action.  Meanwhile, until such definitive action occurs, expect the lack of it to be highlighted here.e   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 20:29:49 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca>c4 Subject: Re: McKinley tops SpecFP AND SpecInt charts+ Message-ID: <3D2CD161.2847575@videotron.ca>    John Smith wrote: 8 > Now how about a statement from Carly to the effect of:L > "HP takes all its products seriously and whenever we find that we have not# > done so, we remedy the situation.o  E I agree entirely. Admitting past mistakes is the first step in givingeL credibility to HP's plans on VMS. And this means immediatly stopping the useI of expressions such as "plan of record" or "previously announced roadmap"tU which simply say that HP intends to continue the mismanagement of VMS done by Compaq.n  J As long as they continue to use those expressions, and as long as StallardJ continues to mention that HP will help those who want to migrate to HP-UX,5 then believing that things have changed will be hard.    ------------------------------  # Date: Thu, 11 Jul 2002 01:07:44 GMTa# From: "John Smith" <a@nonymous.com>i4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsG Message-ID: <AT4X8.7345$6DW1.6940@news01.bloor.is.net.cable.rogers.com>h  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:Ux3X8.475574$cQ3.39450@sccrnsc01... > 0 > "John Smith" <a@nonymous.com> wrote in messageC > news:p63X8.7112$6DW1.3439@news01.bloor.is.net.cable.rogers.com...w > >iJ > > And as to MVS, I'd be really surprised if Compaq/HP has anyone left on > staffiJ > > who knows how to spell MVS in a competitive sales situation, much less canGJ > > accurately speak about the strengths and weaknesses of it vs. OpenVMS. >uL > I think the VMS Ambassadors might take exception to that statement. That's > the good news. >yJ > Unfortunately there are only a couple of hundred Ambassadors. That's the > not-so-good news.0    K That's sort of like telling a person in a war zone: "I've got good news andsG bad news. The bad news is that you've got cancer, it's untreatable, and B you're going to die. The good news is that you've got cancer, it's& untreatable, and you're going to die."    It's always a matter of context.      J > Unfortunately there are only a couple of hundred Ambassadors. That's the > not-so-good news.   K No, that's the good news, but it won't last for long. Without marketing andmJ advertising to grow the VMS market and sales to *NEW* customers, why wouldF HP keep these people around? No offense to them, and I hope it doesn'tK happen, but when the 15,000 heads have rolled, I'd suggest you do a recount.I of the number of VMS Ambassadors left in HP. I'd be shocked if there were ! more than very low double-digits.o    J "Burn your bridges behind you and toe the corporate line if you want a job tomorrow." - Anon.   ------------------------------  # Date: Thu, 11 Jul 2002 01:16:08 GMTn# From: "John Smith" <a@nonymous.com>o4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsG Message-ID: <s%4X8.7353$6DW1.2368@news01.bloor.is.net.cable.rogers.com>   : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message% news:3D2CD161.2847575@videotron.ca...h > John Smith wrote:t: > > Now how about a statement from Carly to the effect of:J > > "HP takes all its products seriously and whenever we find that we have not % > > done so, we remedy the situation.  > G > I agree entirely. Admitting past mistakes is the first step in givingCJ > credibility to HP's plans on VMS. And this means immediatly stopping the use K > of expressions such as "plan of record" or "previously announced roadmap"rL > which simply say that HP intends to continue the mismanagement of VMS done
 by Compaq. >aL > As long as they continue to use those expressions, and as long as StallardL > continues to mention that HP will help those who want to migrate to HP-UX,7 > then believing that things have changed will be hard.y     Touch.h   ------------------------------  % Date: Wed, 10 Jul 2002 21:42:00 -0400B- From: JF Mezei <jfmezei.spamnot@videotron.ca>-4 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2CE247.DFA5745C@videotron.ca>   John Smith wrote: H > HP keep these people around? No offense to them, and I hope it doesn'tM > happen, but when the 15,000 heads have rolled, I'd suggest you do a recountCK > of the number of VMS Ambassadors left in HP. I'd be shocked if there were # > more than very low double-digits.n  M Are VMS ambassadors  "VMS ambassadors" in job, or do they have a real job ande( act as ambassadors in their spare time ?  N Isn't Kerry Main a consultant in the services division and does the ambassadorQ stuff as a second job ? If so, he wouldn't be let go because he is an ambassador.i  E How many ambassadors are sales critters ? Those might be in jeoperdy.    ------------------------------  # Date: Wed, 10 Jul 2002 23:07:01 GMTt# From: "John Smith" <a@nonymous.com>l4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsG Message-ID: <p63X8.7112$6DW1.3439@news01.bloor.is.net.cable.rogers.com>o  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message( news:mnYW8.140116$Uu2.32124@sccrnsc03... >  >kL > At the very least things are less stealthy in VMSland today than they were > three months ago.m  L I would agree with that, however (there's always a however, isn't there?) toI use an analogy,  F-117's are less stealthy than B-2's but you still can'tH see them on radar.    F > > They aren't serious about VMS. If they were, they'd be growing the market+ > > rather than preaching to the converted.d >dI > I think it depends on who the "they" are. Certainly Mark Gorham and hiseK > understudies are serious. As for the folks higher up in the food chain, I L > can't help but wonder why NSK is viewed as a growth opportunity and VMS isH > not. I can understand why HPQ would want to stress HP-UX, but methinks there2F > are prospects who have no interest in HP-UX. VMS might be the Better AnswerI > for them. In competitive situations when going against IBM and MVS, VMS0" > might be the Better Answer. Etc.  L There has always been a core cadre of people in charge of the VMS 'division'F that has held out high hope and stuck their necks on the line for VMS.J Unfortuantely, there thave been goofs (I'd use the term 'fool' except thatK sort of implies that they have considered their actions first) in charge offL executive managment who don't know a bit from a byte, a customer need from a> marketing target, and a coherent strategy from a pile of dung.  I NSK isn't viewed as a growth strategy except to say that it will increase K sales in lockstep with the general growth of the population, perhaps 3% perPF annum, as companies who use NSK increase their capacity. Meanwhile theJ general computing market will grow at perhps twice that rate once businessJ really turns around. And that's with marketing and sales effort. Since VMSL is not accorded that same marketing and sales effort, its base will decline.  L And as to MVS, I'd be really surprised if Compaq/HP has anyone left on staffJ who knows how to spell MVS in a competitive sales situation, much less canF accurately speak about the strengths and weaknesses of it vs. OpenVMS.     >hH > HPQ has yet to adequately articulate a rationale for its VMS strategy. >   J They just don't have a clue, any more than Comapq had a clue.  They shouldH be looking to General Motors and Ford for ideas. Both of these companiesI have products within their own lineups which overlap market segments, andsK may even have multiple models within a 'brand' which overlap. ie. large andmD small SUV's by Chevy, Cadillac, Buick, Hummer H2. They can choose toL advertise to one particular demographic group, or by changing the pitch theyK can aim a Hummer H2 at the same guy who could buy a Cadillac SUV or a ChevyuH SUV or a Buick SUV, or some other set of combinations, while at the sameK time presenting ads that favorably position GM products vs. Ford's or BMW'sw or Land Rover's SUV's.  : But HP is choosing not to advertise/market OpenVMS at all.  J Maybe George Bush in his push for good corporate governance should look atL the effects stupid corporate decision making on the part of suppliers has onJ the competitiveness and effectiveness of American industry. If I were him,K and thank god I'm not, I'd look at ComHpaq as my first case of 'destructive'$ synergy' plaguing corporate America.    L > >  Bet you'd feel a lot more comfortable financially if you had more "SK*'L > > newsletter customers. Most VMS customers would feel more secure in their > VMS @ > > investments if HP were serious about growing the VMS market. > >  > ! > Can't argue that point, either.t  K So, he said rhetorically, if these are so many points that can't be argued,yJ maybe a special free issue of "SK*" detailing the corporate malfeasance ofL HP towards its products and customers and the financial implications thereofJ should be sent to all the fund managers (shareholders) that own big chunksH of HP stock. Maybe that could force HP to be more public about just whatI they think their strategy is, and whether what they think is disconnectedtK from the reality in the field. Perhaps the fund managers could convince thev7 Board of Directors to change horses and strategies too.a   ------------------------------  # Date: Wed, 10 Jul 2002 23:32:14 GMTu1 From: "Terry C. Shannon" <terryshannon@attbi.com>s4 Subject: Re: McKinley tops SpecFP AND SpecInt charts. Message-ID: <2u3X8.143847$Uu2.31111@sccrnsc03>  . "John Smith" <a@nonymous.com> wrote in messageA news:p63X8.7112$6DW1.3439@news01.bloor.is.net.cable.rogers.com...t >I> > "Terry C. Shannon" <terryshannon@attbi.com> wrote in message > > J > > HPQ has yet to adequately articulate a rationale for its VMS strategy. > >p > L > They just don't have a clue, any more than Comapq had a clue.  They shouldJ > be looking to General Motors and Ford for ideas. Both of these companiesK > have products within their own lineups which overlap market segments, andsI > may even have multiple models within a 'brand' which overlap. ie. larget andaF > small SUV's by Chevy, Cadillac, Buick, Hummer H2. They can choose toI > advertise to one particular demographic group, or by changing the pitch2 theyG > can aim a Hummer H2 at the same guy who could buy a Cadillac SUV or a  ChevyeJ > SUV or a Buick SUV, or some other set of combinations, while at the sameG > time presenting ads that favorably position GM products vs. Ford's ore BMW's0 > or Land Rover's SUV's.  I Umm, yes... I have used the General Motors analogy for over a decade now.rK Seems pretty obvious to me (and to you) but the idea never got any tractionwE at DEC. Or CPQ. As for HPQ, well, it's a bit early to deliver a finalo: verdict. But that's just my opinion, and I could be wrong.   ------------------------------  # Date: Wed, 10 Jul 2002 23:30:20 GMT # From: "John Smith" <a@nonymous.com>e4 Subject: Re: McKinley tops SpecFP AND SpecInt chartsG Message-ID: <gs3X8.7197$6DW1.3309@news01.bloor.is.net.cable.rogers.com>e  5 "Bill Todd" <billtodd@metrocast.net> wrote in messages; news:ce1X8.126071$vq.6447346@bin6.nnrp.aus1.giganews.com...e >n9 > "WILLIAM WEBB" <WWEBB1@email.usps.gov> wrote in messagef) > news:0033000071966660000002L002*@MHS...r >i3 > I can get sarcastic with the best of them, bill--  > seppuku and france.  > 5 > If my statements sound like unqualified enthusiasm,l5 > perhaps it's what happens when they are compared ton+ > the Eeyore-like pronouncements of others.s >f > ***rL > What your statement that I responded to sounded like was a suggestion thatG > the insignificant (though unexpected) recognition HP has accorded VMSnC > recently should be sufficient to silence criticism.  If you meante
 otherwise," > you failed make it at all clear. > ***  >m6 > My objective in this forum has been to publicly laud5 > positive actions taken by HP since the acquisition.a >b > ***sG > And my objective since last June has been to highlight the continuing  loads F > of manure Compaq and its cronies insist on spreading here.  Both are" > reasonable and useful endeavors. > ***  >c@ > God knows we've had enough "Imminent Death of VMS- Film at 11"1 > threads to last a whole lot of microfortnights.k >m > ***eL > Gee, I wonder why?  Such things usually have *some* reason for occurring -J > and given that you've chosen to grossly exaggerate their nature, I'd say theeK > discussion that has actually occurred (concern that VMS development wouldhI > languish, customers get increasingly milked, and use dwindle over time,eK > rather than the instantaneous demise your comment suggests) was more than  > justified. > ***  >n2 > My objective in email to folks at HP has been to8 > highlight past bad decisions and suggest remediation--* > not to call individuals idiots or worse. >  > ***tK > That's nice, but the subject here is not email to HP but conversations insG > c.o.v. - and both what's appropriate to discuss and how to discuss iti differ" > radically between the two media. > ***  >e3 > It has been my experience that you get taken mored2 > seriously when you discuss both the positive and > negatives of a given issue.  >u > ***aB > Unfortunately, that generally hasn't happened - and those on the
 'positive'I > side are not one whit less responsible for this situation than those oni theeG > 'negative' side.  It's called polarization, and once it's occurred iti tendsf > to be self-perpetuating. >iF > Since the instigating force behind this has been the long history of neglecteJ > of VMS (and of course Alpha) that culminated in the Alphacide, and sinceK > *no* moves have yet been made to compensate in any measurable way for thedG > damage this has done, and since VMS's owners have yet to deviate fromeJ > treating VMS's customers like mushrooms (kept in the dark and fed shit -I > *especially* in the garbage promulgated by Compaq and its minions after  lastJ > June), I'd say it was up to the 'positive' side to move to the center ifH > they want to break this deadlock:  if they do so, I and I suspect manyC > others will be more than happy to meet them there and discuss them	 situationrA > on its merits - though until those merits themselves change thet conversationD > remains likely to be uncomfortable for those who would prefer that everyone@ > just forget about the past and be optimistic about the future. > ***p >a5 > I won't publish what anyone at HP has said to me inr7 > reply emails-- except to say that I've not been giveni > the secret game plans. > : > I felt that HP was completely baffled by the uproar that: > resulted from their initial pronouncements about VMS and; > the fact that folks were doing everything but diagrammingg< > sentences and reading chicken entrails every time anything > got said about VMS-- >e< > so I sat down and wrote a brief history of actions wrt VMS= > from the GQBob era forward, and discussed how many of thesea; > were perceived in a less-than-favorable light by a goodlyr  > portion of the user community. >f= > Among the things I've discussed at one time or another are:*B > the (mis)use of VMS/NSK profits to prop up marginally profitable? > WIntel sales; the perception that VMS was in maintenance modetA > and the urgent necessity of taking major steps to overcome thateC > perception; the ever-present application availablity problem; andg  > the ever-present .edu problem. >hA > Nothing we haven't discussed here in great detail; I just wrote  > a condensed version. >h= > I've also suggested that they mine the archives for all theeB > excellent suggestions that have been posted here by so many good > folks. >g > ***aF > And all those things are great.  I wouldn't have had the optimism to expectI > such efforts to be worth making (since everyone at Marcello's level anduK > below should already be well-acquainted with reality), and quite possiblyeJ > neither the patience to state them unconfrontationally in any event, but ifD > you did then that justifies your efforts - and more power to them. > ***w >n) > And, again I say, they *are* listening.  >h# > Large ships change course slowly.t >l1 > Am I the only one who senses a sea change here?, >sE > A change that might just be a willingness to loosen the constraintse. > under which VMS has been placed in the past? >a > ***iJ > Am I the only one who senses something very much like the purported 'sea> > change' that was called the 'VMS renaissance' two years ago? > ***i >t9 > This is just opinion--  I think that they've decided tor> > change course, but they haven't decided how far to turn yet. >e > ***oI > And some people thought exactly the same thing two years ago.  Which isr onesG > reason many people (including some of that former group of optimists)e won'tiK > be convinced this time until they *see* change, and substantive change ate > that.l > ***H > E > And, as I've said more than once, polite and reasoned feedback wills= > be listened to and, in my experience, will receive a reply.h >  > ***sL > That wasn't *ever* a problem at Marcello's level and below, and never made an > bit of difference before.s >xJ > While it certainly won't hurt (and might help) for anyone so inclined toL > provide such feedback, I suspect what most people here want is not a replyI > but action.  Meanwhile, until such definitive action occurs, expect the: lack > of it to be highlighted here.o >c  1 All well reasoned and thoughtful, by both Bill's.h    6 Now how about a statement from Carly to the effect of:J "HP takes all its products seriously and whenever we find that we have notL done so, we remedy the situation. OpenVMS is one of those situations. As allF those in the OpenVMS community know, OpenVMS has not been marketed andK promoted effectively for many years. That attitude stops today. Henceforth,oJ OpenVMS will be promoted, marketed, advertised, and suggested to customersH as often, and with the same enthusiasm as all our other operating systemG offerings. In the future if you find that this is not the case, call mysH office and let me know the details, and I personally assure you that the matter will be investigated.  G OpenVMS has reliably been running businesses around the world, teachinghI students, advancing science, and maintaining people's health and security D for over 25 years. HP intends that this will be the case for an everI increasing number of businesses, schools, and governments for the next 25rH years and beyond. OpenVMS is an important part of HP's investment in the/ future of our company. You have my word on it."n    G ps. Carly...leave the speech-writing money in small unmarked bills in as! mayonnaise jar on my front porch.u   ------------------------------  # Date: Wed, 10 Jul 2002 23:36:20 GMTu1 From: "Terry C. Shannon" <terryshannon@attbi.com> 4 Subject: Re: McKinley tops SpecFP AND SpecInt charts. Message-ID: <Ux3X8.475574$cQ3.39450@sccrnsc01>  . "John Smith" <a@nonymous.com> wrote in messageA news:p63X8.7112$6DW1.3439@news01.bloor.is.net.cable.rogers.com...- >iH > And as to MVS, I'd be really surprised if Compaq/HP has anyone left on staffjL > who knows how to spell MVS in a competitive sales situation, much less canH > accurately speak about the strengths and weaknesses of it vs. OpenVMS.  J I think the VMS Ambassadors might take exception to that statement. That's the good news.  H Unfortunately there are only a couple of hundred Ambassadors. That's the not-so-good news.8   ------------------------------  # Date: Thu, 11 Jul 2002 01:50:10 GMTn. From: peter@langstoeger.at (Peter LANGSTOEGER)4 Subject: Re: McKinley tops SpecFP AND SpecInt charts4 Message-ID: <mv5X8.212163$IR.3666316@news.chello.at>  ` In article <a8rniuo2q28ieipb6ga2661oj5cdm762rf@4ax.com>, Alan Greig <a.greig@virgin.net> writes:- >On Tue, 09 Jul 2002 18:28:53 -0400, JF Mezeit& ><jfmezei.spamnot@videotron.ca> wrote: > M >>On the other hand, once VMS run on Intel, *IF* Intel subsidizes advertisingtP >>the same way it does for wintel crap (put the awful Intel logo and tune and weN >>pay 75% of your ads), then it will be most interesting to see how HP decided. >>to market its non-core products such as VMS. > D >It may be telling that the current Intel "Itanium-2" ads running inE >the UK trade press list the "enterprise solutions" operating systems0 >being ported as:i >f >Windows Advanced Server >Windows Datacenter Server >Windows Enterprise Server >Caldera >HP-UX >Miracle (what's that?)c >Redhatn >SuSEw >TurboLinuxk > E >No mention of VMS (or NSK for that matter). Given that Intel want tonD >show as much support for the chip as possible it seems strange they >don't list either.V  F HP does now offer another 'list' in their ad video for their Itanic atE http://www.hp.com/products1/itanium/infolibrary/multimedia/index.htmlo   	HP/UX 	Linux 	Windows  H Also no mention of VMS and NSK. So, when HP does not want to mention it,* why do you expect INTEL to want to do it ?   --   Peter "EPLAN" LANGSTOEGERs% Network and OpenVMS system specialista E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------  # Date: Thu, 11 Jul 2002 02:24:53 GMT@1 From: "David J. Dachtera" <djesys.nospam@fsi.net>N4 Subject: Re: McKinley tops SpecFP AND SpecInt charts' Message-ID: <3D2CF0BC.84D8EB22@fsi.net>T   JF Mezei wrote:e >  > John Smith wrote:tJ > > HP keep these people around? No offense to them, and I hope it doesn'tO > > happen, but when the 15,000 heads have rolled, I'd suggest you do a recountaM > > of the number of VMS Ambassadors left in HP. I'd be shocked if there were?% > > more than very low double-digits.  > O > Are VMS ambassadors  "VMS ambassadors" in job, or do they have a real job andi* > act as ambassadors in their spare time ? > P > Isn't Kerry Main a consultant in the services division and does the ambassadorS > stuff as a second job ? If so, he wouldn't be let go because he is an ambassador.s > G > How many ambassadors are sales critters ? Those might be in jeoperdy.S  G VMS ambassadors tend be much like Rob Young: spout the company line and " damn the torpedoes and the nukes!   E Guess that's why they won't hire me as an ambassador: they know I'm atE straight shooter and when someone's taking a dump on my head, I don'tg? just pass it off as a rain shower. I call a spade a spade in no  uncertain terms.  F Then again, "the 'new' HP" seems more responsive to the customers. So,< perhaps someone like me could be an asset to them after all.  > Let's see now, do I have Mark Gorham's phone number somewhere?   -- o David J. Dachterat dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/r   ------------------------------  % Date: Wed, 10 Jul 2002 23:18:11 -0400.- From: JF Mezei <jfmezei.spamnot@videotron.ca>t4 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2CF8F3.A8AF3BC2@videotron.ca>   Peter LANGSTOEGER wrote:K  > Also no mention of VMS and NSK. So, when HP does not want to mention it,t, > why do you expect INTEL to want to do it ?  M Since VMS won't be available commercially on IA64 for at least 2 years, it is F perfectly reasonable for it not to be mentioned in a list of currently available products.,  M It would be OK for Intel to list other OS that are planned to go to IA64, ands# in that list, VMS should be listed.t   ------------------------------  % Date: Wed, 10 Jul 2002 23:46:34 -0400i' From: "Main, Kerry" <Kerry.Main@hp.com>a4 Subject: RE: McKinley tops SpecFP AND SpecInt chartsT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660821@kaoexc01.americas.cpqcorp.net>   Dave -  F >> VMS ambassadors tend be much like Rob Young: spout the company line) and damn the torpedoes and the nukes! <<<r  F You certainly would not say this if you attended an Ambassador meetingE ..and yes, I am am one of the 200+ Ambassadors (Consulting side). The2G Ambassadors are from many different divisions of HP (Compaq) pre-Sales,yH Customer Services, Consulting, Customer Support Center, Engineering, etc etc.=20l  C There were also about 20 new ones from around the globe at the lastn meeting in May.=20  E Putting 200 or so of experienced VMS types from around the globe thatMG represent Customers like 5 of the top 10 world stock exchanges, some oftG the biggest financial, manufacturing and banking companies on the globerG is a very interesting experience. There were also about 40 ISV partners & at the Partner evening. Great turnout.  H Fwiw - More than one speaker has been reduced to ashes when they did notF do their homework, did not understand the issues and/or tried to fluffB off some issue that the Ambassadors thought was important to their
 Customers.=20i  1 Sue does a good job of keeping us in line though.   G >>> So, perhaps someone like me could be an asset to them after all.<<<   ) Hey, I suspect you might fit right in :-)e  G As far as future job security goes, - I suspect many aspects of HP willyH get some paring down, and some area's will certainly be expanded, so who	 knows?=20S   :-)a   Regardso  
 Kerry Main Senior Consultants Hewlett-Packard Canada! Consulting & Integration Servicest Voice: 613-592-4660t Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----9 From: David J. Dachtera [mailto:djesys.nospam@fsi.net]=20y Sent: July 10, 2002 10:25 PM To: Info-VAX@Mvb.Saic.Comd4 Subject: Re: McKinley tops SpecFP AND SpecInt charts     JF Mezei wrote:d >=20 > John Smith wrote:rE > > HP keep these people around? No offense to them, and I hope it=20-I > > doesn't happen, but when the 15,000 heads have rolled, I'd suggest=20eH > > you do a recount of the number of VMS Ambassadors left in HP. I'd be  ; > > shocked if there were more than very low double-digits.t >=20J > Are VMS ambassadors  "VMS ambassadors" in job, or do they have a real=202 > job and act as ambassadors in their spare time ? >=20H > Isn't Kerry Main a consultant in the services division and does the=20D > ambassador stuff as a second job ? If so, he wouldn't be let go=20 > because he is an ambassador. >=20G > How many ambassadors are sales critters ? Those might be in jeoperdy.B  G VMS ambassadors tend be much like Rob Young: spout the company line andl$ damn the torpedoes and the nukes!=20  E Guess that's why they won't hire me as an ambassador: they know I'm amE straight shooter and when someone's taking a dump on my head, I don'td? just pass it off as a rain shower. I call a spade a spade in noi uncertain terms.  F Then again, "the 'new' HP" seems more responsive to the customers. So,< perhaps someone like me could be an asset to them after all.  > Let's see now, do I have Mark Gorham's phone number somewhere?   --=20s David J. Dachtera0 dba DJE Systems' http://www.djesys.com/  H Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   ------------------------------  % Date: Thu, 11 Jul 2002 00:20:17 -0400r( From: David Froble <davef@tsoft-inc.com>4 Subject: Re: McKinley tops SpecFP AND SpecInt charts, Message-ID: <3D2D0781.5080905@tsoft-inc.com>   JF Mezei wrote:e   > WILLIAM WEBB wrote:2 > 3 >>Did it occur to anyone that, when small steps are:2 >>taken in a positive direction, complaining about3 >>the size of the steps doesn't improve the chancese, >>for additional movement in that direction? >> > 9 > Foole me one, shame on you. Fool me twice, shame on me.i > L > After a decade of similar techniques used to shut up the naysayers only toN > revert back to the "Digital is killing VMS" philosophy a little while later,N > nobody should be saying "HP has finally changed its philosophy" just because< > the word "OpenVMS" was seen in a couple of press releases. > N > 6 months from now, if there has been a consistent mention of VMS in HP pressE > releases and if during that time, the exposure of VMS has increased 2 > consistently, then yeah, kudos will be in order. > P > But HP must understand that VMS customers have been lied to so many times thatP > it takes time to rebuild trust and during that time, it is HP's responsability] > to be patient before customers come to believe this is a sincere and permanent improvement.b > L I'm not arguing with the skepticism.  It has real reason to exist.  However:    O For anyone who has ever been a parent, what response to a positive action by a e child works better?n  @ 1) Praise for the positive action, and encouragement to do more.   or  J 2) Chastisement for not doing even better, and faint or no praise for the  positive action.   Dave   ------------------------------  # Date: Thu, 11 Jul 2002 04:53:03 GMTv& From: lewis@e.cox.net (Keith A. Lewis) Subject: Mousewheel 7 Message-ID: <Pa8X8.84693$DB.2418985@news1.east.cox.net>e  I I'm running Mozilla 1.1A on VMS 7.3 displayed on a Mac running Yellow Dogd- Linux.  Guess what -- the mousewheel works!!!   K Now if it would just work in TPU I'd be set.  Anybody know how to set it upjH to scroll?  When I turn the wheel it says "No user mouse button defined,' completing the default release action."   $ --Spud Demon		thundermaker$yahoo.com9 Posted from my own damn Alpha in my own damn living room t with my own damn software (tnr)d   ------------------------------  # Date: Wed, 10 Jul 2002 18:05:04 GMTi* From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: Only 20% drop in VMS systems (was: wow)A Message-ID: <kH_W8.65246$Im2.2853860@bin2.nnrp.aus1.giganews.com>g  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF402660813@kaoexc01.americas.cpqcorp.net. .. Bill,b  F >>> So, despite what you appear to believe, your explanation in no wayF refutes Andrew's statement (which was similar to one of mine):  an SMPA is the preferred (more efficient, better performing, simpler, andl; usually lower-cost) approach until it runs out of steam.<<<   F And I still contend that it depends on the application and its design,C what the business expectations are for performance and availabilitynB considerations (including single vs multi-site and keeping data inH sync), the IT shops desire to balance the load (maximize use of existing1 HW) and the level of skill of the staff involved.f   ***aA You're still missing the point.  I've in no way asserted that SMPeH *substitutes* for clustering in terms of availability, just that *withinK whatever availability constraints exist* (e.g., that might require at leasthH two clustered systems plus whatever additional systems one might need toL keep performance acceptable should one fail) larger individual cluster nodesK are usually preferable to greater numbers of smaller cluster nodes.  AndrewtL has been saying much the same thing, but tends to emphasize SMP-ness without( placing it in a cluster context as well.  J Of course, many corporate systems (especially those not visible externallyI on the Web) *can* tolerate brief outages as long as they're rare and thussJ may not need to be clustered at all (redundant storage - mirrored remotelyK if necessary - will protect their data) save when one runs out of head-roomoI in one's SMP.  The ability to cluster such large SMPs if it makes overall I sense as an alternative to fork-lift-upgrading them eliminates that falsee
 dichotomy.   - bill   ------------------------------    Date: 10 Jul 2002 11:22:21 -0700( From: bob@instantwhip.com (Bob Ceculski)4 Subject: Re: Only 20% drop in VMS systems (was: wow)= Message-ID: <d7791aa1.0207101022.253d8b11@posting.google.com>t   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<aghft6$eqh$1@new-usenet.uk.sun.com>...a > David J. Dachtera wrote: >  > > Bill Todd wrote: > > 
 > >>[snip]O > >>robust batch queues, since I'm not familiar with Solaris support or lack of.C > >>it for that or equivalent facilities (nor, I suspect, are you).  > >> > > H > > FWIW, Solaris is more or less "UN*X vanilla" here: at, atq, crontab,5 > > etc., but no batch queues as VMS folks know them.h > >  >  > D > But there is a wealth of 3rd party batch queue products, Control-M
 > etc etc. > 	 > Regards  > Andrew Harrisone  ; from one of your prior posts Andrew about VMS IP stacks ...   C Nice try but not the point. Who gives a rats arse if Multinet isn't D vunerable. Now days no customer expects to have to buy an additionalA layered 3rd party product to get IP support in an OS and very few C customers would be impressed with the idea that you need to do thisg* because the vendors own product is broken.   Regardss   Andrew Harrisons  * now I can slightly modify this to read ...  D Nice try but not the point. Who gives a rats arse if you can buy 3rdH party batch queues. Now days no customer expects to have to buy an add'lC layered 3rd party product to get batch queues in an OS and very fewtC customers would be impressed with the idea that you need to do this * because the vendors own product is broken.   chew on your own words matey!o   ------------------------------    Date: 10 Jul 2002 11:24:18 -0700( From: bob@instantwhip.com (Bob Ceculski)4 Subject: Re: Only 20% drop in VMS systems (was: wow)= Message-ID: <d7791aa1.0207101024.7fa81e5d@posting.google.com>i   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<aghft6$eqh$1@new-usenet.uk.sun.com>...e > David J. Dachtera wrote: >  > > Bill Todd wrote: > > 
 > >>[snip]O > >>robust batch queues, since I'm not familiar with Solaris support or lack ofaC > >>it for that or equivalent facilities (nor, I suspect, are you).o > >> > > H > > FWIW, Solaris is more or less "UN*X vanilla" here: at, atq, crontab,5 > > etc., but no batch queues as VMS folks know them.e > >  >  > D > But there is a wealth of 3rd party batch queue products, Control-M
 > etc etc. > 	 > Regardsf > Andrew Harrison   ; from one of your prior posts Andrew about VMS IP stacks ...t  C Nice try but not the point. Who gives a rats arse if Multinet isn't D vunerable. Now days no customer expects to have to buy an additionalA layered 3rd party product to get IP support in an OS and very fewtC customers would be impressed with the idea that you need to do thisn* because the vendors own product is broken.   Regardsn   Andrew Harrison   * now I can slightly modify this to read ...  D Nice try but not the point. Who gives a rats arse if you can buy 3rdH party batch queues. Now days no customer expects to have to buy an add'lC layered 3rd party product to get batch queues in an OS and very fewoC customers would be impressed with the idea that you need to do this * because the vendors own product is broken.   chew on your own words matey!e   ------------------------------    Date: 10 Jul 2002 11:24:18 -0700( From: bob@instantwhip.com (Bob Ceculski)4 Subject: Re: Only 20% drop in VMS systems (was: wow)= Message-ID: <d7791aa1.0207101024.50d1ade0@posting.google.com>.   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<aghft6$eqh$1@new-usenet.uk.sun.com>...1 > David J. Dachtera wrote: >  > > Bill Todd wrote: > > 
 > >>[snip]O > >>robust batch queues, since I'm not familiar with Solaris support or lack of5C > >>it for that or equivalent facilities (nor, I suspect, are you).s > >> > > H > > FWIW, Solaris is more or less "UN*X vanilla" here: at, atq, crontab,5 > > etc., but no batch queues as VMS folks know them.e > >  >  > D > But there is a wealth of 3rd party batch queue products, Control-M
 > etc etc. > 	 > Regardse > Andrew Harrisont  ; from one of your prior posts Andrew about VMS IP stacks ...d  C Nice try but not the point. Who gives a rats arse if Multinet isn'tnD vunerable. Now days no customer expects to have to buy an additionalA layered 3rd party product to get IP support in an OS and very few,C customers would be impressed with the idea that you need to do thisE* because the vendors own product is broken.   Regardsr   Andrew Harrison5  * now I can slightly modify this to read ...  D Nice try but not the point. Who gives a rats arse if you can buy 3rdH party batch queues. Now days no customer expects to have to buy an add'lC layered 3rd party product to get batch queues in an OS and very fewuC customers would be impressed with the idea that you need to do thiss* because the vendors own product is broken.   chew on your own words matey!    ------------------------------  % Date: Wed, 10 Jul 2002 18:42:40 +0100tU From: Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com>m4 Subject: Re: Only 20% drop in VMS systems (was: wow)0 Message-ID: <aghrmg$iid$1@new-usenet.uk.sun.com>   jlsue wrote:  G > On Tue, 09 Jul 2002 19:23:18 +0100, Andrew Harrison SUNUK Consultancyr6 > <andrew_nospam.harrison_remove_this@sun#.com> wrote: >  >  >  >>1. k3 >>You need to be specially trained to manage an OPSv4 >>	environment for starters, this might in itself be1 >>	a hint to you that managing an OPS cluster has 6 >>	additional skill requirements to a single instance. >> > B > I think you overstate the complexities involved.  Sure, there isE > additional training, so what?  Every time you get a new version youfA > need some bit of training for the changes/new features as well.e+ > And it's not as if there are no benefits.  >     + Come on you must understand the difference.i  / There is an Oracle 8i DBA course or a 9i course : and then once you have done that if you want to administerA a OPS or RAC cluster there is a 8i OPS course or a 9i RAC course.0  / Its an additional layer that you have to learn.p    B > It's a trade-off.  Many, many customers find the availabilty andD > scalability features of VMSclustering far outweigh this complexity" > that you keep spouting on about. >     A Get back to the point, Kerrys claim is that from a DBA standpointtB tuning and running a DBMS on a large SMP system is as difficult as0 tuning and running a DBMS on a cluster of nodes.  < This is BS and no DBA will support his claim as we all know.     >  >>2. a5 >>Tuning is very different, in addition to the normala4 >>	tuning issues of data placement etc you also need4 >>	to consider partitioning data, something that you5 >>	may do on an SMP system but for different reasons.o >> > F > What?  Well, none of this was true in the VMSclusters I ran that hadF > Oracle clustered on them.  Tuning consisted of merely making the SGAD > as large as we wanted to reduce real I/O to disk, and there was NO( > need whatsoever to partition the data. >     A Really, so what was your scalability, did the DBMS deliver 2x theo8 thoughput on two nodes when compared with one node ?????  = I have been responsible for OPS clusters on Solaris which are > based on OPS to provide very high reliability, not scalability; no tuning for it but then it wasn't the reason that OPS wase being used.t      F > Now, I admit that these Oracle clusters were some time ago, and thatA > Oracle OPS has many of VMScluster's features built-in, and thus.E > requires different attention.  However, from what I've seen from my/A > colleagues, it's not all that difficult, and the trade-offs and0+ > contraints are well-known and documented.p >     H Again what was the scalability. Oracle themselves are quite conservativeD about OPS scalability and much happier to talk about it as a HA toolB that can also deliver additional throughput if tuned or if you are lucky.     >  >>3. p4 >>There are financial penalties for using OPS, which5 >>	any administrator in his/hers right mind will need  >>	to consider.n >> > H > Sure, as well as the financial penalties of increasing the size of theE > single server when other apps on the same server grow.  I know this F > one from experience.  Oracle apps remained constant, other apps wereG > growing, requiring a server upgrade, which meant over $250K (USD) fory( > Oracle upgrades we really didn't need. >     8 Nope sorry. Learn your charging model before responding.  : Oracle charge on a per CPU basis, so my 32 CPU server will8 cost 32x the price of a single CPU system. Then they add; OPS on top. So is you have 2 x 16 CPU nodes you get charged 0 the standard 32 CPU charge plus the cost of OPS.  8 Many customers have global deals with Oracle that bypass; the per CPU cost, however none of the customers that I haver8 worked with who have this type of deal have included OPS< in this charging making the OPS costs relatively even higher than the norm.  7 The deal looks even worse when you realise that currentn9 large GS boxes deliver less throughput per CPU than Sun'sS9 and in a cluster will not unless we are talking somethinge> like TPC-C deliver linear or even remotely linear scalability.  5 The best case ratio which is if you don't have volumer9 deal and are buying per CPU puts a cluster being used for09 scalability at more than 2x the Oracle licensing costs tow7 get the same throughput. And that ignores the increased  TCO of the clustered option.       >>4. d7 >>Your storage requirements are entirely different, yout9 >>	now need a shared storage subsystem rather than singler >>	host attach.s >> > D > Not a problem.  It's completely transparent in my VMSclusters.  InF > fact, when these customers see how other servers & clusters work wrt > storage, they cringe.a >     ? You seem to miss the point again, the best and cheapest storagee@ for a datawarehouse is often JBOD drives, the read-ahead caching7 in a array just gets in the way. If you don't want a HAuA datawarehouse system and a lot of people don't then clustering to A get scalability for DW will increase your storage costs, you needmB dual pathing etc and reduce your choices. All drives/arrays withinB reason can be connected to a single system subsets of this can be D connected to clusters. Compaqs own clustered TPC-H number for ES45's@ shows this with a relatively complex switch/array infrastructure@ imposed by a need to have connectivity from all the nodes in the cluster to all the storage.       H > Support/Schmupport.  You'd never imagine how many times vendors try toH > pull that crap.  If problems occur with software, it's a simple matterG > to reproduce the problem with only one server running the app... that,D > shuts the vendor up as far as finger-pointing goes.  Rarely, and IF > mean RARELY (as in, never in 17 years) have I had a problem that was > cluster-specific.m >     B Akk, and you are customer facing, you would propose, configure andB install a platform to host an application that wasn't qualified or supported for that platform.  = Why would you do something so contrary to your customers besta interests ?   B I can only assume that your attitude isn't shared either offically7 or unofficially by your previous or current management.r    G > If the vendor's software works in a cluster at all (i.e., they didn't ; > pull some real bone-headed programming mistakes) then any E > cluster-related problems are not usually related to their software,,	 > per se.o >     A So even the risk of this happening would not stop you proposing aa@ cluster without having the vendors apps qualified for a cluster.  B Ouch again, interesting though it may explain the relative lack ofE interest that ISV's show in OpenVMS if thats the prevailing attitude.-  B Just the sort of think that would tee me off if I was an ISV, someC hardware vendor putting forward a "solution" to support my softwarepC that I didn't support and then blaming any issues on my "bone" headi programming.   RegardsB Andrew Harrisoni   ------------------------------  # Date: Wed, 10 Jul 2002 20:04:20 GMTi( From: "konabear" <maurert@ameritech.net>4 Subject: Re: Only 20% drop in VMS systems (was: wow)@ Message-ID: <8r0X8.12129$t%5.5754329@newssvr28.news.prodigy.com>  # "Andrew Harrison SUNUK Consultancy" > <andrew_nospam.harrison_remove_this@sun#.com> wrote in message* news:agbl8o$ffk$1@new-usenet.uk.sun.com... <snip>; > So I would doubt that many OpenVMS servers are being usedy6 > as big cross-platform consolidation platforms. In my. > experience its is normally a donor platform. <snip>  G Personally, I'd say you're in the wrong business if YOUR experience wasVH anything but.  YOUR experience is working where SUN succedes in making aL sale.  YOUR experience would not be expected to include sites were SUN fails to make a sale.t   Todd   ------------------------------  # Date: Tue, 09 Jul 2002 21:35:35 GMTd* From: "Mark Buda" <buda_NO@SPAM.yahoo.com>4 Subject: Re: Only 20% drop in VMS systems (was: wow)= Message-ID: <HGIW8.15849$071.3970148@news1.news.adelphia.net>o  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF402660805@kaoexc01.americas.cpqcorp.net. .. Andrew -  E >How do you proactively shut that single big SMP box down for planneddA maint or OS / HW upgrades / tuning / whatever with ZERO impact onv? application availability ie. no application restarting, no diska( fail-over, no batch jobs restarting etc.    ; EBay was a perfect example of how it is done - all at once.A   mark   ------------------------------  % Date: Wed, 10 Jul 2002 20:46:28 -0400n' From: "Main, Kerry" <Kerry.Main@hp.com>.4 Subject: RE: Only 20% drop in VMS systems (was: wow)T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4023D92A8@kaoexc01.americas.cpqcorp.net>   Bill,w  H <<< You're still missing the point. .... larger individual cluster nodesF are usually preferable to greater numbers of smaller cluster nodes.>>>  D Here is what I stated - "imho, the best approach is to cluster fewer? large SMP servers and only add additional servers as required."-   So how did I miss the point?   Regards1  
 Kerry Main Senior Consultantb Hewlett-Packard Canada! Consulting & Integration Servicesy Voice: 613-592-4660s Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----2 From: Bill Todd [mailto:billtodd@metrocast.net]=20 Sent: July 10, 2002 2:05 PM1 To: Info-VAX@Mvb.Saic.Com.4 Subject: Re: Only 20% drop in VMS systems (was: wow)      2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageH news:BE56C50EA024184DAF48F0B9A47F5CF402660813@kaoexc01.americas.cpqcorp. net. .t Bill,r  F >>> So, despite what you appear to believe, your explanation in no wayF refutes Andrew's statement (which was similar to one of mine):  an SMPA is the preferred (more efficient, better performing, simpler, ande; usually lower-cost) approach until it runs out of steam.<<<-  F And I still contend that it depends on the application and its design,C what the business expectations are for performance and availabilityoB considerations (including single vs multi-site and keeping data inH sync), the IT shops desire to balance the load (maximize use of existing1 HW) and the level of skill of the staff involved.x   ***.A You're still missing the point.  I've in no way asserted that SMPeH *substitutes* for clustering in terms of availability, just that *withinE whatever availability constraints exist* (e.g., that might require atrF least two clustered systems plus whatever additional systems one mightF need to keep performance acceptable should one fail) larger individualB cluster nodes are usually preferable to greater numbers of smallerH cluster nodes.  Andrew has been saying much the same thing, but tends toC emphasize SMP-ness without placing it in a cluster context as well.   ? Of course, many corporate systems (especially those not visibletF externally on the Web) *can* tolerate brief outages as long as they'reF rare and thus may not need to be clustered at all (redundant storage -G mirrored remotely if necessary - will protect their data) save when one F runs out of head-room in one's SMP.  The ability to cluster such largeG SMPs if it makes overall sense as an alternative to fork-lift-upgradingr% them eliminates that false dichotomy.D   - bill   ------------------------------  % Date: Wed, 10 Jul 2002 20:58:58 -0400a' From: "Main, Kerry" <Kerry.Main@hp.com>u4 Subject: RE: Only 20% drop in VMS systems (was: wow)T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF40266081E@kaoexc01.americas.cpqcorp.net>   Andrew -  G <<< I have been responsible for OPS clusters on Solaris which are based F on OPS to provide very high reliability, not scalability no tuning for< it but then it wasn't the reason that OPS was being used.>>>  + Let me guess .. Pre Oracle 8.1.x .. Right ?e  H Re : Oracle OPS (RAC) not being used for scalability .. You obviously doB not have much expereince with 8.1.7 or higher with OPS / RAC. EvenH Oracle types with expereience dating back pre 8.1.x have the same issuesA ie. lack of scalability etc. because they have not used the newere	 versions.u  G I do know of a particular OpenVMS Oracle site today that has approx 250 C HW RAID volumes (approx 8-900 physical disks) with 4 GS60's 8.1.7.x  w/OPS.  E <<< There is an Oracle 8i DBA course or a 9i course and then once youwH have done that if you want to administer a OPS or RAC cluster there is a$ 8i OPS course or a 9i RAC course.<<<  F Course, the alternative is to read the manual, talk to others who have8 done it etc, but perhaps these are out of date concepts?   :-)e   Regards   
 Kerry Main Senior Consultant  Hewlett-Packard Canada! Consulting & Integration Servicesa Voice: 613-592-4660t Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----' From: Andrew Harrison SUNUK Consultancya7 [mailto:andrew_nospam.harrison_remove_this@sun#.com]=20  Sent: July 10, 2002 1:43 PMO To: Info-VAX@Mvb.Saic.Com 4 Subject: Re: Only 20% drop in VMS systems (was: wow)         jlsue wrote:  J > On Tue, 09 Jul 2002 19:23:18 +0100, Andrew Harrison SUNUK Consultancy=206 > <andrew_nospam.harrison_remove_this@sun#.com> wrote: >=20 >=20 >=20 >>1.3 >>You need to be specially trained to manage an OPSs4 >>	environment for starters, this might in itself be1 >>	a hint to you that managing an OPS cluster hasr6 >>	additional skill requirements to a single instance. >> >=20E > I think you overstate the complexities involved.  Sure, there is=20 H > additional training, so what?  Every time you get a new version you=20H > need some bit of training for the changes/new features as well. And=20' > it's not as if there are no benefits.  >=20    + Come on you must understand the difference.   / There is an Oracle 8i DBA course or a 9i courses: and then once you have done that if you want to administerA a OPS or RAC cluster there is a 8i OPS course or a 9i RAC course.t  / Its an additional layer that you have to learn.     E > It's a trade-off.  Many, many customers find the availabilty and=20wG > scalability features of VMSclustering far outweigh this complexity=20y" > that you keep spouting on about. >=20    H Get back to the point, Kerrys claim is that from a DBA standpoint tuningF and running a DBMS on a large SMP system is as difficult as tuning and% running a DBMS on a cluster of nodes.v  < This is BS and no DBA will support his claim as we all know.     >=20 >>2.5 >>Tuning is very different, in addition to the normalo4 >>	tuning issues of data placement etc you also need4 >>	to consider partitioning data, something that you5 >>	may do on an SMP system but for different reasons.  >> >=20I > What?  Well, none of this was true in the VMSclusters I ran that had=20.I > Oracle clustered on them.  Tuning consisted of merely making the SGA=20uG > as large as we wanted to reduce real I/O to disk, and there was NO=20 ( > need whatsoever to partition the data. >=20    A Really, so what was your scalability, did the DBMS deliver 2x thei8 thoughput on two nodes when compared with one node ?????  F I have been responsible for OPS clusters on Solaris which are based onF OPS to provide very high reliability, not scalability no tuning for it6 but then it wasn't the reason that OPS was being used.      I > Now, I admit that these Oracle clusters were some time ago, and that=20 D > Oracle OPS has many of VMScluster's features built-in, and thus=20H > requires different attention.  However, from what I've seen from my=20D > colleagues, it's not all that difficult, and the trade-offs and=20+ > contraints are well-known and documented.r >=20    H Again what was the scalability. Oracle themselves are quite conservativeD about OPS scalability and much happier to talk about it as a HA toolB that can also deliver additional throughput if tuned or if you are lucky.     >=20 >>3.4 >>There are financial penalties for using OPS, which5 >>	any administrator in his/hers right mind will needp >>	to consider.u >> >=20H > Sure, as well as the financial penalties of increasing the size of the  H > single server when other apps on the same server grow.  I know this=20I > one from experience.  Oracle apps remained constant, other apps were=20-J > growing, requiring a server upgrade, which meant over $250K (USD) for=20( > Oracle upgrades we really didn't need. >=20    8 Nope sorry. Learn your charging model before responding.  : Oracle charge on a per CPU basis, so my 32 CPU server will8 cost 32x the price of a single CPU system. Then they add; OPS on top. So is you have 2 x 16 CPU nodes you get chargedp0 the standard 32 CPU charge plus the cost of OPS.  8 Many customers have global deals with Oracle that bypassG the per CPU cost, however none of the customers that I have worked withyH who have this type of deal have included OPS in this charging making the/ OPS costs relatively even higher than the norm.t  7 The deal looks even worse when you realise that currentt9 large GS boxes deliver less throughput per CPU than Sun's 9 and in a cluster will not unless we are talking somethinge> like TPC-C deliver linear or even remotely linear scalability.  5 The best case ratio which is if you don't have volumerH deal and are buying per CPU puts a cluster being used for scalability atG more than 2x the Oracle licensing costs to get the same throughput. And97 that ignores the increased TCO of the clustered option.o       >>4.7 >>Your storage requirements are entirely different, youl9 >>	now need a shared storage subsystem rather than singlew >>	host attach.i >> >=20G > Not a problem.  It's completely transparent in my VMSclusters.  In=20 I > fact, when these customers see how other servers & clusters work wrt=20s > storage, they cringe.u >=20    E You seem to miss the point again, the best and cheapest storage for aoE datawarehouse is often JBOD drives, the read-ahead caching in a arraytG just gets in the way. If you don't want a HA datawarehouse system and aaB lot of people don't then clustering to get scalability for DW willF increase your storage costs, you need dual pathing etc and reduce yourE choices. All drives/arrays within reason can be connected to a singleu  system subsets of this can be=20D connected to clusters. Compaqs own clustered TPC-H number for ES45'sH shows this with a relatively complex switch/array infrastructure imposedG by a need to have connectivity from all the nodes in the cluster to allu the storage.      H > Support/Schmupport.  You'd never imagine how many times vendors try to  H > pull that crap.  If problems occur with software, it's a simple matter  J > to reproduce the problem with only one server running the app... that=20G > shuts the vendor up as far as finger-pointing goes.  Rarely, and I=20eI > mean RARELY (as in, never in 17 years) have I had a problem that was=20d > cluster-specific.f >=20    B Akk, and you are customer facing, you would propose, configure andB install a platform to host an application that wasn't qualified or supported for that platform.  G Why would you do something so contrary to your customers best interestse ?b  E I can only assume that your attitude isn't shared either offically orc4 unofficially by your previous or current management.    J > If the vendor's software works in a cluster at all (i.e., they didn't=20> > pull some real bone-headed programming mistakes) then any=20H > cluster-related problems are not usually related to their software,=20	 > per se.s >=20    A So even the risk of this happening would not stop you proposing at@ cluster without having the vendors apps qualified for a cluster.  B Ouch again, interesting though it may explain the relative lack ofE interest that ISV's show in OpenVMS if thats the prevailing attitude.s  B Just the sort of think that would tee me off if I was an ISV, someH hardware vendor putting forward a "solution" to support my software that> I didn't support and then blaming any issues on my "bone" head programming.   Regards  Andrew HarrisonP   ------------------------------  # Date: Thu, 11 Jul 2002 01:34:35 GMTr* From: "Bill Todd" <billtodd@metrocast.net>4 Subject: Re: Only 20% drop in VMS systems (was: wow)A Message-ID: <Lg5X8.17034$iB1.1120847@bin4.nnrp.aus1.giganews.com>   2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF4023D92A8@kaoexc01.americas.cpqcorp.net. .. Bill,g  H <<< You're still missing the point. .... larger individual cluster nodesF are usually preferable to greater numbers of smaller cluster nodes.>>>  D Here is what I stated - "imho, the best approach is to cluster fewer? large SMP servers and only add additional servers as required."r   So how did I miss the point?   ***wI By responding in a manner that appeared to intend to refute the statementtG you quoted.  If your belief is that one should usually use the largest,e/ fewest servers compatible with availability andmE preservation-of-performance-after-a-failure requirements, then we allc7 (including Andrew) seem to be pretty much in agreement.n   - bill     Regardsw  
 Kerry Main Senior Consultantt Hewlett-Packard Canada! Consulting & Integration Servicesi Voice: 613-592-4660d Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----/ From: Bill Todd [mailto:billtodd@metrocast.net]' Sent: July 10, 2002 2:05 PMy To: Info-VAX@Mvb.Saic.Comn4 Subject: Re: Only 20% drop in VMS systems (was: wow)      2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageH news:BE56C50EA024184DAF48F0B9A47F5CF402660813@kaoexc01.americas.cpqcorp. net. .  Bill,s  F >>> So, despite what you appear to believe, your explanation in no wayF refutes Andrew's statement (which was similar to one of mine):  an SMPA is the preferred (more efficient, better performing, simpler, ando; usually lower-cost) approach until it runs out of steam.<<<m  F And I still contend that it depends on the application and its design,C what the business expectations are for performance and availabilityoB considerations (including single vs multi-site and keeping data inH sync), the IT shops desire to balance the load (maximize use of existing1 HW) and the level of skill of the staff involved.e   ***wA You're still missing the point.  I've in no way asserted that SMP H *substitutes* for clustering in terms of availability, just that *withinE whatever availability constraints exist* (e.g., that might require atoF least two clustered systems plus whatever additional systems one mightF need to keep performance acceptable should one fail) larger individualB cluster nodes are usually preferable to greater numbers of smallerH cluster nodes.  Andrew has been saying much the same thing, but tends toC emphasize SMP-ness without placing it in a cluster context as well.b  ? Of course, many corporate systems (especially those not visible F externally on the Web) *can* tolerate brief outages as long as they'reF rare and thus may not need to be clustered at all (redundant storage -G mirrored remotely if necessary - will protect their data) save when oneaF runs out of head-room in one's SMP.  The ability to cluster such largeG SMPs if it makes overall sense as an alternative to fork-lift-upgrading % them eliminates that false dichotomy.    - bill   ------------------------------  # Date: Thu, 11 Jul 2002 01:36:33 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>-4 Subject: Re: Only 20% drop in VMS systems (was: wow)' Message-ID: <3D2CE569.DCB9B8F0@fsi.net>r   Bob Ceculski wrote:: >  > Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<aghft6$eqh$1@new-usenet.uk.sun.com>...a > > David J. Dachtera wrote: > >u > > > Bill Todd wrote: > > >a > > >>[snip]  = Better catch your breath there, Bob! You're stuttering again!V   -- e David J. Dachtera  dba DJE Systemsa http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/n   ------------------------------    Date: 10 Jul 2002 16:00:12 -0700& From: jordan@ccs4vms.com (Rich Jordan)3 Subject: OpenSSL and certificates concept questionse= Message-ID: <cc5619f2.0207101500.14cde0fe@posting.google.com>   E I'll intrude here with this because we are working on a VMS project. 0C First off, if anyone has recommended reading thats good on covering:F concepts for using certificates in the context of secure http and SSL,F I'd love to know.  I'm going to have the O'Reilly book on SSL shortly,= but whatever else would be useful.  The project is the secure0F 'http/https' client server item I asked about in a previous thread (to: make sure the needed infrastructure was available on VMS).  F We installed the current OpenSSL and most recent VMS cURL on our AlphaD (VMS 7.2-1, TCPIP V5.1 eco 3) to do a proof of concept.  cURL works,E retrieves page data, header data, etc.  Per the demo guide, we are to-F contact a first http URL with a long string of parameters for its cgi;  the return data is an https URL.    We are then to do the following:A      verify the certificate ends with the domain name of the siteBE      verify the certificate URL matches the URL returned in the first  getl.      verify the certificate is from a valid CA*      verify the certificate is not expired*      verify the certificate is not revoked  @ The docs imply that the 'URL is the certificate'.  Attempting toE access the https URL with cURL or a browser errors out, with no validsA information in the headers (error 405, method not allowed, but noaF header listing usable methods).  cURL doesn't appear to have an optionF for displaying the certificate information associated with a URL, save	 as below.   E If I turn on cURL verbose, I see the SSL info where it lists most (ifnD not all) of the relevant data I need; it just doesn't 'return' it inF usable format.  I'm assuming I need to use one of the OpenSSL tools toF actually 'extract' the certificate data via that URL, so I can performE the steps?  Since the CA name is returned (verisign), I presume theretC is also a way to use that info to check that the certificate is notrE revoked?  Everything else is presumably there since verbose cURL seese it.(  C Sorry for asking for what is doubtless basic info, but I don't havepF any experience or docs on this stuff, save for all the lovely RFCs theB secretary is printing, the cURL man pages, and soon the SSL book.  Thanks.f   Rich Jordana   ------------------------------  # Date: Thu, 11 Jul 2002 00:39:50 GMTiL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")7 Subject: Re: OpenSSL and certificates concept questionsx8 Message-ID: <00A10BC1.41709D06@SSRL04.SLAC.STANFORD.EDU>  f In article <cc5619f2.0207101500.14cde0fe@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes:  F >I'll intrude here with this because we are working on a VMS project. D >First off, if anyone has recommended reading thats good on coveringG >concepts for using certificates in the context of secure http and SSL,   M I found a pretty good primer on SSL/certificate concepts at www.mod-ssl.org abO few months ago, but that site seems to be down at the moment.  (There's similariO discussion in my VMS webserver book, but that won't be out for a month or two.)   @ I have no experience with cURL so can't say much about the rest.   -- Alan     O ===============================================================================i0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056eM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210rO ===============================================================================-   ------------------------------    Date: 10 Jul 2002 20:01:55 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r7 Subject: Re: OpenSSL and certificates concept questions43 Message-ID: <3YQ3Fq44HtFJ@eisner.encompasserve.org>e  f In article <cc5619f2.0207101500.14cde0fe@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes:  " > We are then to do the following:C >      verify the certificate ends with the domain name of the sitenG >      verify the certificate URL matches the URL returned in the first  > getC0 >      verify the certificate is from a valid CA, >      verify the certificate is not expired, >      verify the certificate is not revoked  @ Verify the certificate is authorized for the sort of transaction for which you are using it.a   ------------------------------  % Date: Thu, 11 Jul 2002 00:27:15 -0500c( From: Rich Jordan <duodec@speakeasy.net>7 Subject: Re: OpenSSL and certificates concept questions., Message-ID: <3D2D1733.3000102@speakeasy.net>   Larry Kilgallen wrote:h > In article <cc5619f2.0207101500.14cde0fe@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes: >  > " >>We are then to do the following:C >>     verify the certificate ends with the domain name of the sitepG >>     verify the certificate URL matches the URL returned in the first  >>getC0 >>     verify the certificate is from a valid CA, >>     verify the certificate is not expired, >>     verify the certificate is not revoked >  > B > Verify the certificate is authorized for the sort of transaction > for which you are using it.m   Ok.  How? :)  F Apparently the method for 'getting the certificate info' is different F from the method for 'getting the URL page data'.  cURL doesn't return I it, though if you use verbose (effectively a debug mode) it displays the  H needed info (but doesn't return it in usable fashion short of capturing G sys$output/sys$error to a file).  I can see the info, it tells me that fD the first, second, and fourth items are valid, and I can assume the B third is (it'll be verisign or thawte).  How do you verify that a I certificate is not revoked?  That is after doing whatever it is you need  ? to do to get the relevant certificate info in a usable fashion?    Rich Jordano   ------------------------------    Date: 10 Jul 2002 20:18:43 -07001 From: susan_skonetski@hotmail.com (Sue Skonetski)  Subject: OpenVMS Ambassadors= Message-ID: <857e9e41.0207101918.68b3e393@posting.google.com>   B There have been some recent postings about the OpenVMS Ambassadors2 Program and I just wanted to set matters straight.  C I manage the OpenVMS Ambassadors program, it is the best part of my F job and I feel that I am an authority on the program since I have beenC running it for the last 7 years.  I am proud of the Ambassadors andiD the program, it has been used as a bench mark by Digital, Compaq andE now hp, and it is 15 years old.  We do have original members still in  the program.  < There are 205 OpenVMS Ambassadors from 33 countries. OpenVMSA Ambassadors are hp internal people only and they are master levela; folks.   To become an Ambassador requires that an extensiven= application be filled out and a number of references providedcB including another Ambassador.  If any one of the references do notE agree I do not accept the person.  If they do not have their managershC approval I do not accept, if they are not master level experts I do.E not accept, and if their application is not filled out correctly I doA not accept them.   I regularly refuse aplicants.   ! OpenVMS Ambassadors you may know.a  E Kerry Main, John Wisniewski, Mike Kier, Dave Foddy, Kevin FitzPatricki4 and Keith Parris now that he is back in the company.  @ If you have questions, I am happy to answer them, but to use the6 Ambassadors Program as rumor fodder is a waste of cpu.   suer   ------------------------------  % Date: Thu, 11 Jul 2002 01:36:33 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>I  Subject: Re: OpenVMS Ambassadors, Message-ID: <3D2D1958.951CF2AD@videotron.ca>   Sue Skonetski wrote:H > job and I feel that I am an authority on the program since I have been" > running it for the last 7 years.  J Sue, I am curious. Is it just me, or has the existence of the ambassador's) program only recently been made public ?    M On the other hand, I believe that a former DEC employee in montreal who was a N real VMS guru (his name escapes me at present) had one told me he was <specialL title> with the VMS folks. Perhaps he was an ambassador. He had given me theI "I <love> MY VAX" stickers that are proudly displayed above my all mighty I Microvax II. He was let go of Digital fairly early on during the layoffs.   N Since the takeover by Compaq, have the Ambassador's roles/visibility increasedM ? I would think that under Compaq, there would have been no sales rep pushingcL VMS, so would Ambassadors have become more important in generating sales for VMS ?,    > > There are 205 OpenVMS Ambassadors from 33 countries. OpenVMSC > Ambassadors are hp internal people only and they are master levels > folks.  J Is there a typical day job that the Ambassador has ? Sales rep ? Technical; support ? Engineering ? Consultant ? Or is it very varied ?o      6 >   To become an Ambassador requires that an extensive? > application be filled out and a number of references providedw > including another Ambassador.P  A So a box of belgium chocolates isn't enough to convince you ? :-)a    M Herre is a touchy question for you: would the official list of ambassadors bei1 something confidential or could it be published ?e  J Reason I ask: suppose I know someone who would have some VMS questions forG potential sale (or trying to fight removal of an installed VMS system),o9 knowing who the nearest ambassador is might help greatly.    ------------------------------  # Date: Thu, 11 Jul 2002 03:31:42 GMTe1 From: "David J. Dachtera" <djesys.nospam@fsi.net>   Subject: OpenVMS Polls are back!' Message-ID: <3D2D0065.4D46F8E6@fsi.net>   9 I am pleased to announce that the OpenVMS Polls are back!e  E We lost the existing statistics due to a snafu when Advantagenet spuneF off FreePolls, but then maybe that's the good news: we all get to vote all over again.>  * Here are the polls currently available at:' http://www.djesys.com/vms/vmspolls.htmlt  4 o Should Digital have completed the Emerald project?  = o Should HP (or OpenVMS's current owner) revive OpenVMS-IA32?s  8 o Could your company benefit from an AFFORDABLE OpenVMS?  F o If an affordable OpenVMS were available, how many licenses would you! or your company be likely to buy?   G o Do you believe the current OpenVMS ISV program is suited to the needss" of your customers or your company?  C Again, there's no guarantee that these will have any impact, but atl? least our voice can be heard by anyone who visits these polls. m   Better than nothing, eh?   -- e David J. Dachteraa dba DJE SystemsT http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  # Date: Wed, 10 Jul 2002 22:03:40 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)& Subject: Re: PURGE   version set to ;14 Message-ID: <0b2X8.210842$IR.3609249@news.chello.at>  U In article <uingpo82bc6940@corp.supernews.com>, "Vivek Soni" <visoni@bmc.com> writes:.; >With PURGE  I am left with the latest version of the file.o > 9 >Ex. XY.C;11   XY.C;10  XY.C;9 .... and so on till XY.C:1) >v >PURGE will give me XY.C;11d > F >Do we any options in PURGE so we get XY.C;11 but is renamed to XY.C;1  G No, but I wrote a .COM for this. (It was my first one, back in 1983 ;-)s   $ dirup	:== @sw$command:diruph  L $!*************************  Start of DIRUP.COM  *************************** $!C $       FILE    =F$PARSE(P1,"*.*",,,)-F$PARSE(P1,"*.*",,"VERSION",)o $       PURGE           'FILE': $       IF $status THEN RENAME          'FILE'/EXC=;1   ;1 $!L $!**************************  End of DIRUP.COM  ****************************   -- r Peter "EPLAN" LANGSTOEGER0% Network and OpenVMS system specialistu E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------    Date: 10 Jul 2002 16:28:31 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)& Subject: Re: PURGE   version set to ;1= Message-ID: <343f30ae.0207101528.7384b81f@posting.google.com>   a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3D2C768F.861D63A5@videotron.ca>... , > > Fortunately, Hoff provided the solution: > > - > >     $ RENAME  file.typ;*     TMPNAME.TMP; * > >     $ RENAME  TMPNAME.TMP;*  file.typ; >  > Nop:B > 	$ IF F$SEARCH("TMPNAME.TMP") .nes. "" THEN DELETE TMPNAME.TMP;*, >      $ RENAME  file.typ;*     TMPNAME.TMP;) >      $ RENAME  TMPNAME.TMP;*  file.typ;s    D Yes, you should of course check that you won't be renaming the files@ to something that already exists. But as a command procedure, ifB something interrupts the above, you can't fix the problem and just+ rerun it without losing some of your files.i   I'd recommend    (Check for name.TMP files)   then f   $ RENAME name.typ;*  .TMP; $ RENAME name.TMP;*  .typ;  > or the following full-blown procedure: (Usage: @RESEQUENCE.COM	 name.typ)M   $    STATUS = 1a $    ON WARNING THEN EXITq $e  $    NAME = F$PARSE(P1,,,"NAME")  $    TYPE = F$PARSE(P1,,,"TYPE") $    IF (F$SEARCH(P1).EQS."")r	 $    THENt! $        WSO "File(s) not found."e $        STATUS = 2u $        GOTO _EXITt
 $    ENDIF $e% $    SOURCE1 = F$PARSE(";*"      ,P1)r* $    TARGET1 = F$PARSE(".AEFTMP;",SOURCE1) $r* $    SOURCE2 = F$PARSE(";*"      ,TARGET1)* $    TARGET2 = F$PARSE(TYPE      ,SOURCE2) $  $    RENAME 'SOURCE1 'TARGET1r $    RENAME 'SOURCE2 'TARGET2a $a $    GOTO _EXITt $a $_ERROR: $    STATUS = $STATUS6 $    GOTO _EXIT9 $4 $_EXIT::  $    EXIT STATUS .OR. %X10000000  + Comments, bug reports, suggestions welcome.o Disclaimer: JMHO Alan E. Feldman2 afeldman gfigroup comV   ------------------------------    Date: 10 Jul 2002 16:45:28 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)& Subject: Re: PURGE   version set to ;1= Message-ID: <343f30ae.0207101545.5c43d6cb@posting.google.com>   a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3D2C768F.861D63A5@videotron.ca>...(, > > Fortunately, Hoff provided the solution: > > - > >     $ RENAME  file.typ;*     TMPNAME.TMP;m* > >     $ RENAME  TMPNAME.TMP;*  file.typ; >  > Nop:B > 	$ IF F$SEARCH("TMPNAME.TMP") .nes. "" THEN DELETE TMPNAME.TMP;*, >      $ RENAME  file.typ;*     TMPNAME.TMP;) >      $ RENAME  TMPNAME.TMP;*  file.typ;t  D Yes, you should of course check that you won't be renaming the files@ to something that already exists. But as a command procedure, ifB something interrupts the above, you can't fix the problem and just+ rerun it without losing some of your files.a   I'd recommend    (Check for name.TMP files)   then m   $ RENAME name.typ;*  .TMP; $ RENAME name.TMP;*  .typ;  > or the following full-blown procedure: (Usage: @RESEQUENCE.COM* name.typ)(Ignore my previous post on this)   $    STATUS = 1t $    ON WARNING THEN EXITl $h  $    NAME = F$PARSE(P1,,,"NAME")  $    TYPE = F$PARSE(P1,,,"TYPE") $    IF (F$SEARCH(P1).EQS."")r	 $    THENo! $        WSO "File(s) not found."h $        STATUS = 2d $        GOTO _EXITe
 $    ENDIF $i% $    SOURCE1 = F$PARSE(";*"      ,P1) * $    TARGET1 = F$PARSE(".AEFTMP;",SOURCE1) $h* $    SOURCE2 = F$PARSE(";*"      ,TARGET1)* $    TARGET2 = F$PARSE(";",TYPE  ,SOURCE2) $a $    RENAME 'SOURCE1 'TARGET1c $    RENAME 'SOURCE2 'TARGET2a $- $    GOTO _EXITl $f $_ERROR: $    STATUS = $STATUSa $    GOTO _EXITa $u $_EXIT:   $    EXIT STATUS .OR. %X10000000    + Comments, bug reports, suggestions welcome.l Disclaimer: JMHO Alan E. Feldmanf afeldman gfigroup comi   ------------------------------  % Date: Wed, 10 Jul 2002 19:57:57 -0400 K From: "Encompass - HP Enterprise Technology Symposium" <KilleenJ@toast.net>m& Subject: Re: Quick Poll - two sessions/ Message-ID: <uipigp8nhu4741@corp.supernews.com>C   As to why #2 and #3...  F #2 is by HP Engineering.  #3 is from a known technically knowledgeable user...o     --   Jeff Killeen   All Info: http://www.Killeen.ccs  ? ---------------------------------------------------------------l  . "John Smith" <a@nonymous.com> wrote in message? news:kOWW8.331$6DW1.125@news01.bloor.is.net.cable.rogers.com... # > I'd choose #1, but read on......./ >wI > Seems to me that Session 2 and session 3 *could* cover the same ground.. WhymI > not roll the two of them together, call it session #2, and lengthen the7 > merged session by an hour? >h$ > Then I'd attend session #1 and #2. >e >oG > "Encompass - HP Enterprise Technology Symposium" <KilleenJ@toast.net>e wrote 6 > in message news:uio9gsogenp0ac@corp.supernews.com...L > > 1) If you had to choose between session #1 or session #2 which would youK > > pick and why? (Please assume that session #3 is being offered no matterS
 > > what ) > >n8 > > 2) Would you attend any of these 3 sessions and why? > >h/ > > Thanks in advance (no HP replies please)...s > >0 > >t > > Session #1 > >rL > > Developing Secure Applications on OpenVMS Using the Common Data Security > > Architecture (CDSA). > >eJ > > This session provides details on the Common Data Security ArchitectureF > > (CDSA) including installing, configuring and administering CDSA on
 > OpenVMS.L > > Since CDSA is just being introduced on OpenVMS, this session provides anL > > overview of CDSA and it's components, how they fit together and how CDSA > isH > > related to SSL. Just having CDSA doesn't help much if you don't haveK > > applications using it, so we discuss how to build a CDSA application onu > > OpenVMS. > > 
 > > Session#2f > > : > > OpenSSL for OpenVMS: Application Development and ToolsI > > This technical session explains the components and tools available asy partF > > of the OpenSSL port to OpenVMS. The session covers the various SSL > protocolswL > > available and their use in application development that utilize OpenSSL.C > > Topics covered include an SSL application example, performance,p > certificatesH > > and other issues associated with developing along with deploying SSL > > applications.  > >t
 > > Session#3b > >I@ > > This session covers using OpenSSL to secure and encrypt data > communicationsD > > over the Internet or other open network. This is an introductory session,L > > but will cover creating and generating Digital Certificates, getting andC > > building the OpenSSL libraries from source code (under Unix and 	 OpenVMS),hE > > basic programming using the OpenSSL libraries in C, and a workingo exampleeL > > that is portable between Unix and OpenVMS.Experience with either Unix orJ > > OpenVMS, and familiarity with the C programming language is suggested. > ThoseoJ > > without programming experience may also benefit from the discussion of4 > > applications which can or should use encryption. > >h > > -- > >- > > Jeff Killeen > >-# > > All Info: http://www.Killeen.cc5 > >iC > > ---------------------------------------------------------------p > >r > >u >c >R   ------------------------------    Date: 10 Jul 2002 15:18:45 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)c( Subject: Re: Quorum discussion/questions= Message-ID: <cf15391e.0207101418.14488150@posting.google.com>    Just a minor correction.  Y Nic Clews <sendspamhere@127.0.0.1> wrote in message news:<3D1C363C.BF58C402@127.0.0.1>...eD > Remember you also have the option of adding the new ES45 as a zeroB > (non voting) member but this has other implications for locking.  > Nic was undoubtedly thinking here of the effect of setting theD LOCKDIRWT parameter to zero, as is often done on nodes when VOTES is> set to zero (e.g. satellite nodes).  But technically the VOTES( parameter itself doesn't affect locking.. ----------------------------------------------. Keith Parris | parris at encompasserve dot org   ------------------------------  % Date: Wed, 10 Jul 2002 14:19:44 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>P Subject: Re: RECALL suggestion, Message-ID: <3D2C7ABD.CF57022C@videotron.ca>   Guy Peleg wrote:J > In upcoming V7.3-1 we have modified recall/all to accept a parameter, soR > $RECALL/ALL dir will get you all the commands in the recall buffer starting with > dir.  = THANK YOU ! That will be most very usefull for so many cases.i  4 Is that going to be available on VAX or Alpha only ?   ------------------------------  % Date: Wed, 10 Jul 2002 17:58:08 -0400g; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>e Subject: Re: RECALL suggestion$ Message-ID: <3d2cae0d$1@news.si.com>  L >Not sure how implementable this is, but how about the ability to recall the' >last X directories that were visited ?v  / Fixed in a previously release freeware program.r -- hA Brian Tillman                   Internet: tillman_brian at si.comrA Smiths Aerospace                          tillman at swdev.si.coms= 3290 Patterson Ave. SE, MS      Addresses modified to prevent-< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------    Date: 10 Jul 2002 16:00:35 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman) Subject: Re: RECALL suggestion= Message-ID: <343f30ae.0207101500.3b8b6a4f@posting.google.com>   s SPAMSINK2001@YAHOO.COM (Alan E. Feldman) wrote in message news:<343f30ae.0207100754.39ecc9c9@posting.google.com>...fc > JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3D2BE6DB.DB068599@videotron.ca>...sO > > Not sure how implementable this is, but how about the ability to recall the * > > last X directories that were visited ? > > U > > What I would even like better is that this would be JOB and not process specific.  > > K > > EG: find out what directories were visited in the last couple of hours.n > > J > > Of course, having the ability of RECALL CD listing all the CD commands' > > previously issued would solve that.b > > R > > When one rummages through the SYS$COMMON tree on character cell, the directoryQ > > names can become rather tediious and it would really be nice to have featuresfC > > that allow easier going back/forth between various directories.  >  > G > You could try my TO.COM. I can e-mail to you if you like. It does not F > keep track of directories used by spawned processes though. I assume* > that's what you mean by "JOB specific").    > Actually, it would be a simple matter to change all the DEFINE< statements in TO.COM to DEFINE/JOB, if that's what you want.    B Also, the advantage of a separate default recall stack is that youF could do things like the following which saves you from having to type/ any stored default for subsequent DCL commands.   	 IDS03$ TOn Processing old default....   5 Enter   S   for  SAVE = DISK$DATA1:[FELDMAN.FTCOMMON]t, Enter   9   for 9BACK = DISK$DATA1:[FELDMAN]2 Enter   8   for 8BACK = DISK$DATA1:[FELDMAN.CHECK]3 Enter   7   for 7BACK = DISK$DATA1:[FELDMAN.MONREM]n3 Enter   6   for 6BACK = DISK$DATA1:[FELDMAN.FT.ILO]D3 Enter   5   for 5BACK = DISK$DATA1:[FELDMAN.CHKNOD]W/ Enter   4   for 4BACK = DISK$DATA1:[FELDMAN.FT].8 Enter   3   for 3BACK = DISK$DATA1:[FELDMAN.UTI.GENERAL]3 Enter   2   for 2BACK = DISK$DATA1:[FELDMAN.UTI.FT]h' Enter   1   for  LAST = XXDISK:[XX.DAT]a/ Press <RET> for  HERE = SYS$SYSDEVICE:[FELDMAN]t Or enter new default: F ====================================================================== Your default ise   SYS$SYSDEVICE:[FELDMAN]c    IDS03$ COPY 8BACK:ALL.OUT HERE:t? %COPY-S-COPIED, DISK$DATA1:[FELDMAN.CHECK]ALL.OUT;666 copied tol/ SYS$SYSDEVICE:[FELDMAN]ALL.OUT;666 (660 blocks)i $i   Disclaimer: JMHO Alan E. Feldman  afeldman gfigroup come   ------------------------------  # Date: Thu, 11 Jul 2002 01:54:23 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: RECALL suggestion' Message-ID: <3D2CE994.25105125@fsi.net>y   labadie wrote: > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3D2BE6DB.DB068599@videotron.ca...K > > Not sure how implementable this is, but how about the ability to recallr > theM* > > last X directories that were visited ? >  > HelloS > = > I remember a cd.com doing that, defining a bunch of symbolsk8 > (cd_1=dka200:<durand>, cd_2= dka300:<dupont.perso>...)/ > Of course, it could end using some resources.t  F O.k. I gotta look dumb here - can someone tell me what is the value ofF this stack? I've seen multiple people express that interest. Why wouldC you need to track where you've been? If you don't know where you'rea5 going, what's the point of trying to navigate at all?d  G ...or am I just spoiled from using Reflection and/or DECterm instead ofa a VT?o   -- a David J. Dachtera< dba DJE Systemsc http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/r   ------------------------------  % Date: Wed, 10 Jul 2002 23:22:50 -0400a- From: JF Mezei <jfmezei.spamnot@videotron.ca>n Subject: Re: RECALL suggestion+ Message-ID: <3D2CFA0A.5C232D4@videotron.ca>l   "David J. Dachtera" wrote:H > O.k. I gotta look dumb here - can someone tell me what is the value ofH > this stack? I've seen multiple people express that interest. Why wouldE > you need to track where you've been? If you don't know where you'reb7 > going, what's the point of trying to navigate at all?   H suppose you are doing much of your work in one directory. You then spawnE another process so you can rummage through system directories to find N something. You find it, log off to return to your original directory, but now,O you've forgotten in exactly which directory that file you had found really was.u  L OR: you don't spawn and go rummage and once you've found your file, you need% to return to your original directory.n  L However, the solution in 7.3 which allows RECALL/ALL string would be perfect since you could do:g  J RECALL/ALL CD  and that would list all your prevous CD commands, which youC could also include in a default recall buffer loaded at login time.,   ------------------------------  % Date: Thu, 11 Jul 2002 01:18:10 +0100t4 From: John Laird <john@laird-towers.freeserve.co.uk> Subject: Re: RRD42 Caddy8 Message-ID: <fdjpiu4nujrdtd8iin1amkdommclod6994@4ax.com>  F On Wed, 10 Jul 2002 16:51:32 +0100, Nic Clews <sendspamhere@127.0.0.1> wrote:   >John Laird wrote: >> hI >> On Wed, 10 Jul 2002 09:40:30 +0100, Nic Clews <sendspamhere@127.0.0.1>l	 >> wrote:y >> l >> >issinoho wrote:d >> >>oW >> >> Has anyone in the UK got one of these spare that they'd be willing to mail to me?H >> >>p  >> >> Any help MUCH appreciated. >> >I >> >I'm pretty sure I have a spare one around, may have broken side clips  >> >but works anyway.  >> aI >> It would be kinder to give away a spare RRD46 or the like ;-)  LoadingoM >> anything off a single-speed drive is fine if you like cooking a meal whileu >> you wait... >pB >The DEC 2000 has an RRD45 or 46 which uses one of those caddys...  K I though they were the boggo tray drives ?  The 40 had the grippy claw in aeL clear case device (case withdrew), and the 42 was the tray with flip up lid,F all of which remained in the drive.  Everything after that returned to sanity, I thought.   K >> I am notoriously mean, but have bought two 2nd hand 4x drives for my own N >> machines rather than use a spare 40 or 42 from the office supply !  Neither& >> are DEC, but both worked just fine. >.H >I think a caddy would survive a postal journey, I'm fairly sure the GPOG >or whatever they are calling themselves these days would do a good jobnE >of loosening anything more complex into more constituent bits than aXG >present day Alpha uses for addressing. That is, if they didn't lose it  >or misdeliver it first.  L I am just plain naive.  Both CDROMs were mailed to me, and both worked fine.L I have also had hard disks mailed, but admittedly these tend to be packed inI the box 20 times the volume with a 6" layer of foam between every part ofR" the drive and the world outside...     	John  --  
 John Laird Yezerski Roper Ltd   ------------------------------  # Date: Thu, 11 Jul 2002 01:42:19 GMT*1 From: "David J. Dachtera" <djesys.nospam@fsi.net>*) Subject: Re: SET FILE/TRUNCATE equivalent0' Message-ID: <3D2CE6C0.9B619C88@fsi.net>    Carl Perkins wrote:e > 7 > "David J. Dachtera" <djesys.nospam@fsi.net> writes...  > }Carl Perkins wrote: > }>I > }> "Antony Wardle" <antony.wardle@nospammmmm.optusnet.com.au> writes...M/ > }> }let me know if you find out how to do it.: > }> }4 > }> }I have a file that is a 2.2 million lines long- > }> }and I only want the first 11 characterse' > }> }from each line, I will have to don* > }> }this every few days, so I guess that* > }> }dcl is out for me, and I don't speak > }> }programming languages ;-(* > }> }	 > }> }ant  > }>. > }> Of course you can do it in DCL. Try this: > }> > }> $ create first11.sortspec) > }> /field=(name=textdata,pos:1,size:11)N > }> /key=1; > }> /data=textdata O > }> $ sort/spec=first11.sortspec/stable really_big_file.txt smaller_output.txtt > }>G > }> Tada. Well, technically it isn't exactly all in DCL - it also usesm* > }> the sort specification file langauge. > }>C > }> You only need to create the sort specification file once, then B > }> reuse it every time. The miracle of the nonsorting sort - theD > }> key is a constant, so the /STABLE on the command line makes theE > }> records all appear in the output in the same order they appeared  > }> in the input. > } - > }If you're gonna go that far, why not just:, > }E2 > }$ CONVERT/FDL=SYS$INPUT infile outfile/TRUNCATE > }FILEQ, > }       ORGANIZATION            SEQUENTIAL	 > }RECORD % > }       BLOCK_SPAN              YESt& > }       CARRIAGE_CONTROL        NONE' > }       FORMAT                  FIXED"$ > }       SIZE                    11 > }$ EOD > }David J. Dachtera > F > Because this discards records that are less than 11 characters long,, > issuing an informational message for each?  H The original post implied that the records were consistently longer thanC 11 bytes, else he'd have said something to the effect of "up to thet first 11 characters".m   > (And comes out really ugly$ > when you TYPE the resulting file.)  F He did not state that as a need, but you're quite right - and doing soD adds nothing to the size/content of the file, so why not? Go for it!  D > You can fix the discarding of short records by adding a /PAD=%X20,1 > or whatever your prefered padding character is.:  F Assuming there are any "short" records. Probably well to plan a ahead,E but it was not stated explicitly in the spec., and the statement that  was made implied otherwise..  C > If you want it to look good when you TYPE it, carriage_return forT > CARRIAGE_CONTROL is better.f    He did not state that as a need.  H > There is more than one way to skin a cat, or trim records from a file.  5 ...or even trim bytes from a record! True, as always.    --   David J. Dachterao dba DJE Systemsh http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/l   ------------------------------   Date: 10 Jul 2002 23:07:11 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Setting Terml* Message-ID: <agiemv$ncr$1@web1.cup.hp.com>  X In article <3d2c3d87.10421415@news.verio.net>, shawnm@cotepe.carenet.org (Shawn) writes:  C :We just received a new DS20E server pre-installed with DecWindows. F :When the server boots it boots to the GUI.  I know there is a commandC :which I can re-direct the boot to goto a VT220 Terminal but cannotSD :remember what it is.  Any help would be appreciated.  It is running :OpenVMS 7.2-1  F   The OpenVMS Frequently Asked Questions (FAQ) -- the FAQ has answers @   to many common questions -- contains a section with the title:  @     "How do I boot an AlphaStation without monitor or keyboard?"  F   which will provide details on the SRM console and also on the serialE   terminal settings expected by most SRM consoles.  Newer systems can G   variously run at different (faster) baud rates, check the SRM consolee4   and the system hardware documentation for details.  F   Also please see the documentation on the startup logging mechanisms,   and specifically:o  .     http://www.openvms.compaq.com/doc/73final/)       6017/6017pro_010.html#booting_verif        -- and--  .     http://www.openvms.compaq.com/doc/73final/.       6017/6017pro_010.html#sysman_startup_sec    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 10 Jul 2002 22:51:26 -040072 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: Setting TermoJ Message-ID: <rdeininger-1007022251260001@1cust58.tnt3.nashua.nh.da.uu.net>  H In article <3d2c3d87.10421415@news.verio.net>, shawnm@cotepe.carenet.org (Shawn) wrote:   >Hi All. > C >We just received a new DS20E server pre-installed with DecWindows.aF >When the server boots it boots to the GUI.  I know there is a commandC >which I can re-direct the boot to goto a VT220 Terminal but cannotaD >remember what it is.  Any help would be appreciated.  It is running >OpenVMS 7.2-1  H You just received a NEW DS20E, and it's running 7.2-1 ?!?!  Good Grief! G Whose warehouse has it been sitting in?  Or has it been lost in transitt for 2 or 3 years?   H If it's a new system, make sure you got everything listed on the Bill ofH Materials, which should include a manual.  As others have indicated, theH manual will suggest the console command SET CONSOLE SERIAL to solve your problem.  E And if you bought this system new from HP/Compaq, you should probably'H investigate whether you have the right to run current VMS without payingJ for an upgrade license.  V7.3 is current, and I _think_ you're entitled toI run the version that's current when you buy the system, even if it has anhJ older one pre-installed.  The Factory-Installed-Software disks tend to lag$ by a few months, but this is absurd.  F Regardless of the system's history, you should be upgrading to V7.3 or. (soon) V7.3-1.  V7.2-1 is no longer supported.  O And if the OS is that old, I'd assume the firmware needs to be updated as well.c   ------------------------------  # Date: Wed, 10 Jul 2002 19:35:28 GMT ( From: Don Sykes <annonymous@pacbell.net>' Subject: Re: SMTP 8bit hack not workinga+ Message-ID: <3D2C8CFE.4A5DBACF@pacbell.net>r   Brian Tillman wrote: > K > >FYI "Wire Paladin" is not my pseudonym, it is a reference to a 1950's TVl	 > westerne1 > >(Have Gun, Will Travel) staring Richard Boone.f > 2 > "HAVE GUN, WILL TRAVEL," read the card of a man.* > A knight without armor in a savage land.. > His fast gun for hire heed the calling wind;1 > A soldier of fortune is the man called Paladin.i > & > Paladin, Paladin, where do you roam?' > Paladin, Paladin, far, far from home.h > > > I still remember this theme song 50 years later.  Go figure. > --M As do I. And each episode would begin with Paladin in his San Francisco hotel-J where he would be reading a letter or newspaper. Then he'd call "boy", theC Chinese guy and hand him his card with a message to be telegraphed. N I think we remember these things, cause we saw them over and over when we wereN kids our brains were like dry sponges. I think every 50's & early 60's TV showN has a permanent place in my head. Even if I get Alzheimer's and don't rememberP my own name, I'll still remember the theme songs of every single show I saw back then.<   Cheers -- 0   Have VMS. Will Travel. Wire Paladin (@alphase.com)f
 San Francisco    ------------------------------  # Date: Thu, 11 Jul 2002 02:03:26 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>e$ Subject: Re: Stupid storage question' Message-ID: <3D2CEBB5.7186D5D0@fsi.net>w   "John N." wrote: > D > Will someone please explain the difference between a SAN, and just1 > connecting a few Alpha Servers to a few HSG80s.-  @ I would suspect that it's six of one, half-a-dozen of the other.  M > More specifically, I currently have four Alpha servers with two FC adaptersmN > each.  We plug into a SAN that is managed by our Unix storage group.  In theL > SAN are two HSG82 arrays that are dedicated to us.  We are not getting theK > support we need, so I want to pull our four HSG80s (two redundant HSG82s) K > out of the SAN and connect directly to them and manage them ourselves.  IHM > suppose I will need to buy two 8-port switches (since each Alpha has two FCo7 > adapters, and there are four HSG80s) and some cables.- > 2 > Will we also need a PC with SWCC to manage them?  F Well, technically, *NO ONE* "needs" SWCC. The CLI is quiet sufficient,1 but some folks *(not me) find GUIs easier to use..  @ SWCC is a particular headache because for it to work right, yourF workstation's IP address must back-translate. Not sure why, but it's a
 real *BITCH*!e  C ...and yes, I still need to be able to command HSx's from within an  OpenVMS batch job!   -- m David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/i   ------------------------------  # Date: Thu, 11 Jul 2002 01:33:23 GMTo1 From: "David J. Dachtera" <djesys.nospam@fsi.net>-A Subject: Re: System Disk Initialization Parameter recommendationsK' Message-ID: <3D2CE4AA.E0E25178@fsi.net>:   Phillip Helbig wrote:e > : > > I am about to migrate from a 4.1GB SCSI system disk to> > > an 18GB SAN-based system disk (with OpenVMS Alpha V7.2-2). > E > My rule of thumb for my hobbyist disks is /HEADERS=<25,000 per GB>,pE > double that for a system disk, and keep the defaults for everythingc@ > else.  The defaults seem OK for "normal" usage (for a suitableI > definition of "normal" :-) ) except for /HEADERS, which is much too low  > for modern disks.i > F > I post this not so much as a response, but to collect comments on my2 > "rule of thumb for normal-usage hobbyist disks".  D I would caution that the formula you suggest would be suitable for aG large disk with many small files. On the 70GB (or so) disk mentioned by H another poster, this formula would result in tying up more than 1/2GB ofD disk space for file headers - more than 1.7 million blocks. That's a *LOT* of files!f  B Another caution is that /MAXIMUM_FILES (the size of the INDEXF.SYSF bitmap) should be as large as /HEADERS if /HEADERS exceeds the default size for /MAXIMUM_FILES.   -- o David J. Dachterab dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/t   ------------------------------  # Date: Thu, 11 Jul 2002 02:17:26 GMT'1 From: "David J. Dachtera" <djesys.nospam@fsi.net>rA Subject: Re: System Disk Initialization Parameter recommendationso' Message-ID: <3D2CEEFF.D767667F@fsi.net>    "David J. Dachtera" wrote: >  > Phillip Helbig wrote:T > >r< > > > I am about to migrate from a 4.1GB SCSI system disk to@ > > > an 18GB SAN-based system disk (with OpenVMS Alpha V7.2-2). > >nG > > My rule of thumb for my hobbyist disks is /HEADERS=<25,000 per GB>, G > > double that for a system disk, and keep the defaults for everything.B > > else.  The defaults seem OK for "normal" usage (for a suitableK > > definition of "normal" :-) ) except for /HEADERS, which is much too lowT > > for modern disks.] > >eH > > I post this not so much as a response, but to collect comments on my4 > > "rule of thumb for normal-usage hobbyist disks". > F > I would caution that the formula you suggest would be suitable for a  " Naturally, I meant "not suitable".  I > large disk with many small files. On the 70GB (or so) disk mentioned bytJ > another poster, this formula would result in tying up more than 1/2GB ofF > disk space for file headers - more than 1.7 million blocks. That's a > *LOT* of files!n > D > Another caution is that /MAXIMUM_FILES (the size of the INDEXF.SYSH > bitmap) should be as large as /HEADERS if /HEADERS exceeds the default > size for /MAXIMUM_FILES. -- - David J. Dachtera1 dba DJE Systemsa http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/5   ------------------------------  # Date: Wed, 10 Jul 2002 19:05:02 GMT ' From: Stuart Fuller <stufuller@usa.net>X Subject: Re: TIMAe0 Message-ID: <v7thga.pm1.ln@dadsys1.fuller.local>   Nic Clews wrote:   > What does TIMA mean? > 9 > http://www.openvms.compaq.com/openvms/supportchart.htmlc > 7 > e.g. Top line is ES45, saying we need VMS 7.3 + TIMA.o > F > An ex DECcie told me that TIMA was "Technical Information ManagementE > Architecture" but it doesn't make sense in that context. Now, if ithH > meant patches, then ECO (Engineering Change Order) would make a little
 > more sense.v > 1 > The "Digital Dictionary" doesn't have an entry.f >  > So, does anyone know?d  0 Yes, the acronym expansion mentioned is correct.  K TIMA was how a lot of tech information was distributed.  Patches, firmware pL updates, etc. were all distributed via TIMA.  Our local office used to have E a MicroVAX 3100 with (for its time) lotsa disk space to "cache" this aK information.  Other information could be downloaded or mailed to one's VAX a& mail (there was no Alpha then) account  @         Stu (working in at the same job, now in his 3rd company)   ------------------------------  % Date: Wed, 10 Jul 2002 22:27:28 -0400 2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: TIMA J Message-ID: <rdeininger-1007022227280001@1cust58.tnt3.nashua.nh.da.uu.net>  D In article <3D2C689D.1020502@dskwld.zko.dec.compaq>, "John.Malmberg"' <Malmberg@dskwld.zko.dec.compaq> wrote:r   >Nic Clews wrote:n& > > Is it a FLA (Four Letter Acronym)? >r" >No, it is an ETLA - Extended TLA.  , Wouldn't it be trendier to use XTLA instead?   ------------------------------  % Date: Wed, 10 Jul 2002 18:07:44 -0400'; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>c# Subject: Re: Used Alpha's and Vax's3" Message-ID: <3d2cb04f@news.si.com>  A >I am trying to locate used OpenVMS systems (Alpha and VAX). DoesiB >anyone know of any websites where I can find sellers?  Thank you.   www.glcomp.com www.compunetinc.comi www.computerclearinghouse.com- www.geminidigital.com  www.decparts.comF www.matchcomputer.com (actually, IGS owns the above and this one, too) www.newman.com  % How many do you want?  I could go on.o -- @A Brian Tillman                   Internet: tillman_brian at si.comhA Smiths Aerospace                          tillman at swdev.si.comu= 3290 Patterson Ave. SE, MS      Addresses modified to preventw< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  # Date: Wed, 10 Jul 2002 23:03:04 GMTt4 From: "Andy Bustamante" <a_c_bustamante@bigfoot.com># Subject: Re: Used Alpha's and Vax'saF Message-ID: <I23X8.12668$A43.1311999@newsread2.prod.itd.earthlink.net>  L We will have several VAX 3100's available in the next few months, and have a4 VAX 7000-730 (3 CPU) we're retiring.  Any interrest?  A The 7000 is my (internal) web server, formerly a customer system.o --   Andy Bustamanteh( remove the ascii-95's to reply by e-mail      < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3D2B9156.508C17F6@fsi.net...y > Jim Rizzolo wrote: > >oD > > I am trying to locate used OpenVMS systems (Alpha and VAX). DoesE > > anyone know of any websites where I can find sellers?  Thank you.0 >  > A few listings at: > 1 > http://www.djesys.com/vms/hobbyist/support.html : > http://www.djesys.com/vms/hobbyist/support.html#hardware >  > -- > David J. Dachterat > dba DJE Systemsm > http://www.djesys.com/ >g* > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/X >p   ------------------------------  % Date: Wed, 10 Jul 2002 22:29:38 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)# Subject: Re: Used Alpha's and Vax'soJ Message-ID: <rdeininger-1007022229390001@1cust58.tnt3.nashua.nh.da.uu.net>  F In article <I23X8.12668$A43.1311999@newsread2.prod.itd.earthlink.net>,5 "Andy Bustamante" <a_c_bustamante@bigfoot.com> wrote:   M >We will have several VAX 3100's available in the next few months, and have a 5 >VAX 7000-730 (3 CPU) we're retiring.  Any interrest?a   Located where?   ------------------------------  # Date: Thu, 11 Jul 2002 04:17:11 GMT3- From: "John E. Malmberg" <wb8tyw@qsl.network>e8 Subject: Re: Using GNU C on OpenVMS FAQ (Looking for it)( Message-ID: <3D2D0288.90007@qsl.network>   Frits A.M. Storms wrote:< > "John E. Malmberg" <wb8tyw@qsl.network> schreef in bericht   >>J >>I am not sure how your link statement succeeded.  I suspect that you are> >>using a LNK$LIBRARY* logical name to bring in the VAX C RTL. >  m > Well. Not really.v* > $ LINK/NOTRACE hello,gnu_cc_library:crt1I > works as well, so the options-file with sys$share:decc$shr/share had nog > function in this. 5 > (A good deal of ignorance on my part, I am afraid.)y? > And forgot to mention : I am working on OpenVMS Alpha V7.1-2.e  G First do a show log LNK$*.  Someone may have done you a "favor" so thatc> you are in effect linking against the VAXCRTL.OLB selectively.  F I have not worked with GCC on ALPHA, so I do not know if it generates  prefixes or not.  H When the OpenVMS Hobbyist program offered the DEC/COMPAQ C compilers as   part of it, I stopped using GCC.  4 I did use GCC quite a bit before then on VAX though.  J >>The SYS$SHARE:DECC$SHR.EXE image has prefixes on all of the modules, andJ >>the output from GCC, (unless there has been a major change) does not put# >>prefixes on the external symbols.e >>I >>Because the VAXCRTL is old, it is much better to be using the DEC C RTLf
 >>instead. >>C >>There is a special DECC shared image that does not have prefixes.   L That image is VAXC2DECC.EXE, and it only seems to be present on OpenVMS VAX.  D I thought that the GZIP example that I provided a URL to was linked 7 against the VAXC2DECC.EXE image, but now I am not sure.s  ( All my files from that era are archived.   -John  wb8tyw@qsl.network Personal Opinion Only?   ------------------------------  % Date: Wed, 10 Jul 2002 19:28:03 +0100nE From: Jamie Stallwood <this.no.work.try.something.else@project76.net> % Subject: Re: VS 3100/38: Boot messageo8 Message-ID: <e4voiu8f9r2356ioq3tnoinfhhninhh7l3@4ax.com>  D On Wed, 10 Jul 2002 11:23:07 -0600, Kevin Handy <kth@srv.net> wrote:  4 >I'd guess that you're missing a network terminator.= >Items marked with a single '?' usually don't keep the systema? >from booting, those with double '??' do.  A missing terminator > >is set up (for some reason) to cause a boot fail on power up. >a9 >Plug in a terminator on the network port, and it it willa$ >probably boot without any problems. >o9 >You can also just try typing 'boot' at the '>>>' prompt.t< >It should boot from there even with the missing terminator.  5 Thanks for the info. Now if I can find the manual...!w   ------------------------------  % Date: Thu, 11 Jul 2002 01:11:27 +0100m4 From: John Laird <john@laird-towers.freeserve.co.uk>% Subject: Re: VS 3100/38: Boot message.8 Message-ID: <lsipiusub28aokra12t647rmpo8g9orcvl@4ax.com>  3 On Wed, 10 Jul 2002 19:28:03 +0100, Jamie Stallwood-6 <this.no.work.try.something.else@project76.net> wrote:  E >On Wed, 10 Jul 2002 11:23:07 -0600, Kevin Handy <kth@srv.net> wrote:  >i5 >>I'd guess that you're missing a network terminator.1> >>Items marked with a single '?' usually don't keep the system@ >>from booting, those with double '??' do.  A missing terminator? >>is set up (for some reason) to cause a boot fail on power up.s >>: >>Plug in a terminator on the network port, and it it will% >>probably boot without any problems.  >>: >>You can also just try typing 'boot' at the '>>>' prompt.= >>It should boot from there even with the missing terminator.  >w6 >Thanks for the info. Now if I can find the manual...!  J The NI message is simply "no network found".  Either connect to a network,C or stick a terminator on the selected port (AUI/BNC), or ignore it.h  K The DZ error suggests the printer port is not well (the 4th status value) - I unless you need to use it desperately, you could manage without it.  I am J curious that the keyboard and mouse status values seem to indicate each isG connected.  In serial console mode, the mouse won't do anything and the 1 keyboard is usually connected to the terminal :-)   J I have the manual in Bookreader format (it used to be on the Condist CDs).G If you would like a copy, let me know.  MGBOOK is a terminal-based toolsK which will access Bookreader files in the absence of a graphics display, of-: course the tables and figures might be difficult to read !     	Johnm   ------------------------------  + Date: Wed, 10 Jul 2002 21:25:28 +0000 (UTC)-* From: bleau@umtof.umd.edu (Lawrence Bleau)4 Subject: What does UCX 6.2 SMTP intermittently fail?0 Message-ID: <agi8o8$rgp$1@grapevine.wam.umd.edu>  M Hi, I'm helping a colleague.  He has a VAXstation 4000-90 running OpenVMS VAX:L 6.2.  I just installed UCX 4.2 and put on the latest, greatest patch - ECO 53 (yes, I rebooted).  Here's the UCX SHO VERS output:s  >   Digital TCP/IP Services for OpenVMS VAX Version V4.2 - ECO 52   on a VAXstation 4000-90 running OpenVMS V6.2      M I configured it okay, and we can ftp and telnet both ways, as well as receive L email.  Sending email is problematic, though.  When sent to one of the otherO systems I administer, it works, but when sent to another one, it fails.  Here'sV. the contents of the file UCX$SMTP_LOGFILE.LOG:    D %%%%%%%%%%%%                   10-JUL-2002 15:49:23.80  %%%%%%%%%%%%  N %UCX-I-SMTP_LOGSUC, Using log file SYS$SPECIFIC:[UCX_SMTP]UCX$SMTP_LOGFILE.LOG    D %%%%%%%%%%%%                   10-JUL-2002 15:49:24.33  %%%%%%%%%%%%  D %UCX-I-SMTP_SYMBRUN, Symbiont is running the queue UCX$SMTP_SURYA_01H smtp_sender_close sclose R0 status = -1, errno = 65535, vaxc$errno = 316H smtp_sender_close sclose R0 status = -1, errno = 65535, vaxc$errno = 316H smtp_sender_close sclose R0 status = -1, errno = 65535, vaxc$errno = 316H smtp_sender_close sclose R0 status = -1, errno = 65535, vaxc$errno = 316  L Outgoing email failed 4 times, so I'm assumming that is the reason for the 4; error messages.  If 316 is a VMS status code, then it meansg  % %SYSTEM-F-IVCHAN, invalid I/O channel   O Any idea what is happenning and how to go about debugging this?  And why should-O it work to one system (sampex.umd.edu) but not another (umtof.umd.edu), both ofiM which are on the same Ethernet cable with the sending system (surya.umd.edu)?  Thanks.u   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edu-   ------------------------------  % Date: Wed, 10 Jul 2002 18:28:19 -0400i; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>s8 Subject: Re: What does UCX 6.2 SMTP intermittently fail?$ Message-ID: <3d2cb51f$1@news.si.com>  F >I configured it okay, and we can ftp and telnet both ways, as well as receivet. >email.  Sending email is problematic, though.  I Solve all of your SMTP problems on VMS by getting MX from www.madgoat.comu --A Brian Tillman                   Internet: tillman_brian at si.comrA Smiths Aerospace                          tillman at swdev.si.comc= 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 10 Jul 02 20:02:39 +0200) From: p_sture@elias.decus.ch (Paul Sture)o' Subject: Re: Where to put startup stuffe) Message-ID: <bORaRjo$2eIE@elias.decus.ch>n  n In article <343f30ae.0207091220.5dedd931@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes:\ > p_sture@elias.decus.ch (Paul Sture) wrote in message news:<hTbIPfr3W$Yd@elias.decus.ch>...q >> In article <343f30ae.0207081456.547ddd17@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes:m_ >> > p_sture@elias.decus.ch (Paul Sture) wrote in message news:<kG$Jt4wZPKgK@elias.decus.ch>... g >> >> In article <l0djiuc7vmvor6stp3b2iq46m65t1k5c8g@4ax.com>, jlsue <jlsuexxxz@screaminet.com> writes:nM >> >> > I've always made use of all the SY*.com files - though SYSECURITY and N >> >> > SYCONFIG are not needed most of the time (well, "always" is since they >> >> > were made available).r >> >> > M >> >> > It helps to put a write sys$output statement at the beginning and endgI >> >> > of each of these files (I also include f$time()) so that you knowaF >> >> > which procedure is being executed.  Likewise, before I executeJ >> >> > external procedures I use write sys$output to let me know where myI >> >> > startup is processing... this is very useful if things die in thes: >> >> > middle, helps me get right to the problem quicker. >> >> > J >> >> > One thing to note:  Some of the files - e.g., SYLOGICALS.COM - getN >> >> > executed even during a "minimum" boot.  You need to put logic there toN >> >> > skip over things you don't want to happen during a minimum boot (e.g.,2 >> >> > mounting all of the system/cluster disks). >> >> > L >> >> It is worth noting that the startup calls these procedures with P1 set >> >> appropriately. >> > dK >> > Could you please explain what you mean by this? Give an example maybe?J >> tP >> For example, in SYLOGICALS, on a normal boot, P1 is "FULL". I honestly forgetL >> what the value is on a minimum boot ( "MINI" ?), bur have used it to good >> effect in the past. m >> __d
 >> Paul Stured >> Switzerland >  >  >  c > $ SEAR SYLOGICALS.COM P1& > $!      P1 - Name of device to check2 > $       If .NOT. F$Getdvi(P1,"EXISTS") Then Exit2 > $       DSENT   'P2' 'F$Getdvi(P1,"FULLDEVNAM")' > $  >  > ; > That P1 is for a device and is only for the subroutine in F > SYLOGICALS.COM. I tried a minimum boot (set STARTUP_P1 to "MIN") andA > it ran SYLOGICALS.COM, SYPAGSWPFILES.COM and SYSECURITY.COM buts- > skipped SYCONFIG.COM and SYSTARTUP_VMS.COM.  > G > Your post made it sound like you could pick and choose which get run. ; > AFAIK, you can only pick between MIN and regular booting.  >   = I think we must be careful here wrt version and Alpha or VAX.   : The following example is from Alpha V7.3 (plus most ECOs).F (Come to think of it I can only remember doing tests for P1 on Alphas)  2 I inserted the following lines in SYSLOGICALS.COM:  
 $ set noon) $ open/write x sys$manager:sylogicals.logi $ write x "P1 = ", p1f $ write x "P2 = ", p2e $ write x "P3 = ", p3i $ write x "P4 = ", p4  $ write x "P5 = ", p5t $ write x "P6 = ", p6X $ write x "P8 = ", p7  $ write x "P1 = ", p8 	 $ close xs  O And did a normal reboot (sorry, not near the system at the time, so didn't wanto to do a minimum reboot).   Here's the result:    $type sys$startup:sylogicals.log	 P1 = FULL  P2 = g P3 = p P4 = d P5 = o P6 = o P8 =   P1 = -  N On inspecting my (originally V7.2) SYLOGICALS.COM further, it mentions testingN that F$GETSYI("STARTUP_P1") is not "MIN" in order to determine whether clusterA logicals will be set up. Perhaps that is a "more supported" test.D  < And I see that the V7.3 .TEMPLATE has added a "Set NoOn" :-) __
 Paul Sture Switzerlandm   ------------------------------    Date: 10 Jul 2002 18:07:41 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)' Subject: Re: Where to put startup stuff5= Message-ID: <b096a4ee.0207101707.2da6a597@posting.google.com>   Z p_sture@elias.decus.ch (Paul Sture) wrote in message news:<bORaRjo$2eIE@elias.decus.ch>...p > In article <343f30ae.0207091220.5dedd931@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes:^ > > p_sture@elias.decus.ch (Paul Sture) wrote in message news:<hTbIPfr3W$Yd@elias.decus.ch>...s > >> In article <343f30ae.0207081456.547ddd17@posting.google.com>, SPAMSINK2001@YAHOO.COM (Alan E. Feldman) writes:ma > >> > p_sture@elias.decus.ch (Paul Sture) wrote in message news:<kG$Jt4wZPKgK@elias.decus.ch>... i > >> >> In article <l0djiuc7vmvor6stp3b2iq46m65t1k5c8g@4ax.com>, jlsue <jlsuexxxz@screaminet.com> writes:  [SNIP]L > >> >> > One thing to note:  Some of the files - e.g., SYLOGICALS.COM - getP > >> >> > executed even during a "minimum" boot.  You need to put logic there toP > >> >> > skip over things you don't want to happen during a minimum boot (e.g.,4 > >> >> > mounting all of the system/cluster disks).
 > >> >> > N > >> >> It is worth noting that the startup calls these procedures with P1 set > >> >> appropriately. > >> > 1M > >> > Could you please explain what you mean by this? Give an example maybe?: > >>  R > >> For example, in SYLOGICALS, on a normal boot, P1 is "FULL". I honestly forgetN > >> what the value is on a minimum boot ( "MINI" ?), bur have used it to good > >> effect in the past. y > >> __e > >> Paul Stureo > >> Switzerland > >  > >  > >  A > > $ SEAR SYLOGICALS.COM P1( > > $!      P1 - Name of device to check4 > > $       If .NOT. F$Getdvi(P1,"EXISTS") Then Exit4 > > $       DSENT   'P2' 'F$Getdvi(P1,"FULLDEVNAM")' > >  > > = > > That P1 is for a device and is only for the subroutine infH > > SYLOGICALS.COM. I tried a minimum boot (set STARTUP_P1 to "MIN") andC > > it ran SYLOGICALS.COM, SYPAGSWPFILES.COM and SYSECURITY.COM butP/ > > skipped SYCONFIG.COM and SYSTARTUP_VMS.COM.b > > I > > Your post made it sound like you could pick and choose which get run.e= > > AFAIK, you can only pick between MIN and regular booting.m > >  > ? > I think we must be careful here wrt version and Alpha or VAX.a > < > The following example is from Alpha V7.3 (plus most ECOs).H > (Come to think of it I can only remember doing tests for P1 on Alphas) > 4 > I inserted the following lines in SYSLOGICALS.COM: >  > $ set noon+ > $ open/write x sys$manager:sylogicals.log2 > $ write x "P1 = ", p1a > $ write x "P2 = ", p2a > $ write x "P3 = ", p3d > $ write x "P4 = ", p4o > $ write x "P5 = ", p5s > $ write x "P6 = ", p6e > $ write x "P8 = ", p7  > $ write x "P1 = ", p8t > $ close x  > Q > And did a normal reboot (sorry, not near the system at the time, so didn't want  > to do a minimum reboot). >  > Here's the result: > " > $type sys$startup:sylogicals.log > P1 = FULLb > P2 = o > P3 = h > P4 =   > P5 = R > P6 = l > P8 =   > P1 = e    -: Yes, P1 has the value "FULL" but it doesn't do anything in SYLOGICALS.COM.p   Your original statement,   [cut and pasted]N > >> >> It is worth noting that the startup calls these procedures with P1 set > >> >> appropriately.  F is still kind of vauge. Set appropriately for what? It never gets used in SYLOGICALS.COM at least.n    P > On inspecting my (originally V7.2) SYLOGICALS.COM further, it mentions testingP > that F$GETSYI("STARTUP_P1") is not "MIN" in order to determine whether clusterC > logicals will be set up. Perhaps that is a "more supported" test.     7 Sorry, I should have mentioned my VMS v is 6.1 and 6.2.4    > > And I see that the V7.3 .TEMPLATE has added a "Set NoOn" :-)     Cool.      Disclaimer: JMHO Alan E. Feldman- afeldman gfigroup com    ------------------------------  % Date: Wed, 10 Jul 2002 22:04:08 -0400f2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: Re: Where to put startup stuff2J Message-ID: <rdeininger-1007022204080001@1cust58.tnt3.nashua.nh.da.uu.net>  F In article <bORaRjo$2eIE@elias.decus.ch>, p_sture@elias.decus.ch (Paul
 Sture) wrote:     > >I think we must be careful here wrt version and Alpha or VAX. >n; >The following example is from Alpha V7.3 (plus most ECOs). G >(Come to think of it I can only remember doing tests for P1 on Alphas)9 >.3 >I inserted the following lines in SYSLOGICALS.COM:  >m >$ set noonc* >$ open/write x sys$manager:sylogicals.log >$ write x "P1 = ", p1 >$ write x "P2 = ", p2 >$ write x "P3 = ", p3 >$ write x "P4 = ", p4 >$ write x "P5 = ", p5 >$ write x "P6 = ", p6 >$ write x "P8 = ", p7 >$ write x "P1 = ", p8
 >$ close x >#P >And did a normal reboot (sorry, not near the system at the time, so didn't want >to do a minimum reboot).r >l >Here's the result:  >t! >$type sys$startup:sylogicals.log 
 >P1 = FULL >P2 =  >P3 =  >P4 =  >P5 =  >P6 =  >P8 =  >P1 =  >0  H IIRC, each command file in the startup database is called by STARTUP.COM with 8 parameters.  H If the entry in the database supplies a particular parameter, that value* is used in the call to the component file.  P Otherwise, STARTUP.COM uses its own value of Pn when calling the component file.  J STARTUP.COM's Pn is either the Pn it was called with, or failing that, the* corresponding STARTUP_Pn system parameter.  J A component file, like SYLOGICALS.COM or SYSTARTUP_VMS.COM, is free to use the Pn, or ignore them.M  I I specifically investigated what happens on Alpha; VAX may be different. o> Also, there are some differences in different versions of VMS.  I And my memory isn't perfect, and I don't feel like opening up STARTUP.COMk at the moment to check.>   ------------------------------  % Date: Thu, 11 Jul 2002 01:58:07 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>i Subject: Which thread am I in ?o+ Message-ID: <3D2D1E65.DCBBA47@videotron.ca>r  M Say I have a subroutine that has various RAB blocks allocated. I want to makenI sure that if the calling program is multithreaded, one thread will not beiK using a RAB from another thread even though that RAB may be marked "free tos reuse" for that thread.o  M Is there an easy way in C to find out which thread is calling my subroutine ?    ------------------------------  % Date: Wed, 10 Jul 2002 20:01:11 +0200t9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>S  Subject: Re: Who Knows Asterix ?' Message-ID: <3D2C7667.5F8E0437@aaa.com>p   Well, getting a bit of topic...t  < I was in Britany a couple of weeks ago. Britteny is the only@ area in France there you don't have to pay on the highway. Why ?  A Well, sometimes around 1250 (yes in the 13'th century !), BritanysC was still an independent "country". Then the king of France offereduD the Queen of Brittany ("Anne de Bretagne") a marriage. At that time,G there *was* road tolls on France, but not in Brittany. Anne de BretagneCC managed to get a paragraph into the marriage "contract" saying thathF Britany would still be freed from road tolls. And the folks in BritanyC still shows this document whenever the question is raised nowadays.h  0 Talk about not letting your history pass away...   Jan-Erik Sderholm.   -   JF Mezei wrote:- >  > labadie wrote:! > > Any idea of what is Asterix ? 7 > > For the french people, it is only a famous cartoon.6 > P > "Only" ????? Astrix is tantamount to a history book to most who grew up on itP > and learned about how Les Gaulois put up a fight against the Romans... Without( > Asterix, most wouldn't know about it !   ------------------------------  # Date: Thu, 11 Jul 2002 02:54:27 GMTp# From: ualski <ualski@earthlink.net>t  Subject: Re: Who Knows Asterix ?- Message-ID: <3D2CF351.51786F55@earthlink.net>    JF Mezei wrote:u >  > labadie wrote:! > > Any idea of what is Asterix ? 7 > > For the french people, it is only a famous cartoon.c > P > "Only" ????? Astrix is tantamount to a history book to most who grew up on itP > and learned about how Les Gaulois put up a fight against the Romans... Without( > Asterix, most wouldn't know about it !  L Count me amongst that group. Without Asterix I also wouldn't know the origin of a certain English tradition.0   ------------------------------   End of INFO-VAX 2002.378 ************************