0 INFO-VAX	Tue, 03 Feb 2004	Volume 2004 : Issue 67      Contents: Re: $SETUAI() Query/Problem  Re: $SETUAI() Query/Problem & Re: (OT) FS: Digital Network Equipment& Re: (OT) FS: Digital Network Equipment) Re: Adding internal disks to ES40 Model 2  Re: AlphaServer 21000 Re: Canadian Troll infestation in rec.travel.air Re: DQDRIVER on a 500au  Re: DQDRIVER on a 500au 0 Re: Help with SSH, any tricks to get it to work? RE: Hobbyist questions? P Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performance on ItaniuP Re: Intel and Microsoft provide higher 32-bit applications performanceon Itanium7 Re: is the current SCO denial of service attack bogus ? 7 Re: is the current SCO denial of service attack bogus ? : Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !! MyDoom.B RE: MyDoom.B Need help solving SSH problemsP Re: OpenVMS vs unix security ... Andrew, the IBM guy awaits your  response!  res0 Re: PC print server devices compatible with VMS?0 Re: PC print server devices compatible with VMS? Re: Renaissance of VAX-VMS ? Re: Rumours of (CPU) Wars  Re: Rumours of (CPU) Wars  Re: Rumours of (CPU) Wars  Re: Rumours of (CPU) Wars  Re: Rumours of (CPU) Wars  Re: Rumours of (CPU) Wars , Re: Slightly OT: Code Red on the rise again?, Re: Slightly OT: Code Red on the rise again? Re: stupid smtp/mapi question  Re: stupid smtp/mapi question  To buy Alphaserver 1000 4/233   Re: Why PERL on VMS? We have DCL  Re: Why PERL on VMS? We have DCL  Re: Why PERL on VMS? We have DCL  RE: Why PERL on VMS? We have DCL	 ZIP/UNZIP 
 Re: ZIP/UNZIP  [DS15] New sound card supported   F ----------------------------------------------------------------------   Date: 2 Feb 2004 22:36:59 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) $ Subject: Re: $SETUAI() Query/Problem3 Message-ID: <3y7iwnRm2v5E@eisner.encompasserve.org>   o In article <IrqdnRt5p8O-JIPdRVn-jw@brightview.com>, "covendotartdottalk21dotcom" <postmaster@127.0.0.1> writes:  > < > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message/ > news:+Cdy84lFaBVk@eisner.encompasserve.org... 6 >> In article <KZednRzI0rm6FYDdRVn-hQ@brightview.com>,= > "covendotartdottalk21dotcom" <postmaster@127.0.0.1> writes: ? >> > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message 2 >> > news:kzO4pYdQXv9J@eisner.encompasserve.org... >>G >> >> And I think your network consisting entirely of secured computers E >> >> with no connections to the outside world but yet with some risk ' >> >> of password sniffing is unlikely.  >> >6 >> > They don't have connections to the outside world. >> >H >> > Some risk of password sniffing?  Well, anyone with a legitimate boxG >> > on the network (or physical access to disconnect a legitimate box, H >> > and plug in their own one), and can put an adapter into promiscuous >> > mode can do just that.  >>F >> Then the network does not consist of secured computers if attackers >> have physical access. > C > One has to place some level of trust in one's employees.  If they E > subsequently then decide (for whatever reason) to become malicious, E > there's little that can be done about that - having security people C > monitoring and frisking people is a non-sequitur - Quis custodiet  > ipsos custodies?  @ If malicious people with access to the network is not a problem,? there should be no need to encrypt passwords on the wire (which ( was your originally stated requirement).   ------------------------------  % Date: Tue, 03 Feb 2004 09:25:19 +0000 0 From: Chris Sharman <chris.sharman@sorry.nospam>$ Subject: Re: $SETUAI() Query/Problem4 Message-ID: <bvnpdv$1e7$1$830fa79f@news.demon.co.uk>   Chris Scheers wrote:H > "Normally", logical names are checked for in the tables: PROCESS, JOB, > GROUP, SYSTEM, in that order.  > 8 > LNM$FILE_DEV is the logical which controls this order. > G > You can redefine it in your LNM$PROCESS_DIRECTORY table to change the  > order used by your process.  > I > The trick is that the executive mode definition of LNM$FILE_DEV says to I > only look in the SYSTEM table, so process logicals have no affect on an " > executive mode name translation.  E Not true on my systems: the two (system) definitions of lnm$file_dev  > differ only in their inclusion (or not) of decw$logical_names.# This is in a vms 7.3 Alpha cluster. I On our standalone Alpha 7.3 system (no decwindows) there's no difference.   6 Can't find any definitions of lnm$file_dev, except in G sys$update:pcsi$*.com - I presume the original definitions are part of   the bootstrap, not dcl.    Chris    ------------------------------   Date: 3 Feb 2004 05:58:19 -0800 . From: fabiopenvms@yahoo.com.br (Fabio Cardoso)/ Subject: Re: (OT) FS: Digital Network Equipment = Message-ID: <f30679fb.0402030558.16f805e8@posting.google.com>    "B i t N i s s e n ;-)" <bitnissen#spammer_format_your_harddisk#@hotmail.com> wrote in message news:<uoat10906v905eu6n6fb876mfs48k30f80@4ax.com>...  > Hi   > 5 > I have a lot of Digital Network Equipment for sale.  > < > If you are interested in some of the parts - send me a bid >  > Buyer pays shipping  > 0 > Product                       PART-NO      Pcs2 > ----------------------------- ------------ -----0 > DecBridge 90                  DEWGB-M        20 > DecBrouter 90T1               DEWB1-N        20 > DECHUB 900 Powersupply        H7874-MA       30 > DECHUB One (PSU)              DEHVA-NB       20 > DecRepeater 90T-16            DETML-M        30 > DecRepeater FA                DEFAR-M        20 > Decserver 90L+                DSRVG-M       110 > DECSwitch 900EE                              10 > DECSwitch 900EF               DEFBA-M        10 > Multiswitch HUB 624T          DLMR2-M        10 > MuxServer 90                  DSRZF-E        60 > PortSwitch 900TP/12           DETPX-M        30 > Powersupply for X90 modules   H7027-AB       50 > RouteAbout Access EI          DEXBR-M       110 > RouteAbout access EW          DEXZR-M        40 > RouteAbout Central Router EW  DEZ8R-P        20 > WanRouter 90 EW	              DEWAZ-E        2 >  > Best regards >  > Jimmi Aakjr > 3 > Remove everything between ## in my e-mail address  > C > (Cross posted in (comp.sys.dec.micro,comp.os.vms,vmsnet.networks)     H Why dont you put all these items in an auction site ? Like E-bay etc...     Regards    FC   ------------------------------  % Date: Tue, 03 Feb 2004 19:27:32 +0100 S From: "B i t N i s s e n ;-)" <bitnissen#spammer_format_your_harddisk#@hotmail.com> / Subject: Re: (OT) FS: Digital Network Equipment 8 Message-ID: <qfmv101qt9ps0hit597vvssm6rmf0agi23@4ax.com>   Hi Fabio  F I have tried these newsgroups because the hardware is almost only used by TRU64 and OpenVMS customers.   E In Denmark we use the newsgroups a lot for trading used hardware and  E I dont think Ebay is the best media for selling DEC network hardware.   	 ;O) Jimmi   I >Why dont you put all these items in an auction site ? Like E-bay etc...  >  >  >Regards >  >FC    ------------------------------  $ Date: Tue, 3 Feb 2004 12:51:45 -0500; From: "Island Computers USA" <dbturner@islandco.com.nospam> 2 Subject: Re: Adding internal disks to ES40 Model 20 Message-ID: <101vnmkrbqe0ncc@news.supernews.com>  2 I would use the 3X-KZPEA-DB ( oour PN IC-KZPEA-DB)7 The BA610-6E is the 6 slot cage and runs at U160 speeds @ The controller is a dual channel with 2 external connections too   David      --   David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180  Savannah GA 31404  Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.nospam.net  http://www.hpaq.net   1 "Scott Vieth" <svieth@wi.rr.com> wrote in message 7 news:5a85bce2.0402021231.371adc57@posting.google.com...  > All: > F > I would like to add an internal disk cage (BA610-6D) to my ES40 so IE > can have some "local" disk storage in addition to the connection to 
 > our EVA. > F > I have been looking through the Golden Eggs and the ES40 Quickspecs.F > There seems to be a few options for SCSI controllers that could work# > in an ES40 running OpenVMS 7.3-1.  > E > Does anyone have recommendations for which SCSI controller I should A > use in my ES40?  I'd like to do some controller-based mirroring H > (especially if I could get write-back caching!) but I would settle for> > a "dumb" controller and then use VolShad or host-based RAID. > + > Any suggestions/info greatly appreciated.  > 	 > thanks, 6 > -Scott, SCSI plumber in the snowy/frozen Midwest :^)   ------------------------------  ! Date: Tue, 03 Feb 04 12:00:52 GMT  From: jmfbahciv@aol.com  Subject: Re: AlphaServer 2100 3 Message-ID: <401f9ffa$0$8815$61fed72c@news.rcn.com>   H In article <13t1f1-slq.ln1@deep.bit.bucket>, BAH <bah@bit.bucket> wrote: <snip>   This isn't me.     /BAH  ' Subtract a hundred and four for e-mail.    ------------------------------  # Date: Tue, 03 Feb 2004 16:09:10 GMT - From: "neil tupper" <neil_tupper@hotmail.com> 9 Subject: Re: Canadian Troll infestation in rec.travel.air - Message-ID: <GIPTb.385577$JQ1.32473@pd7tw1no>   % "edo" <nobody@cryptorebels.net> wrote    > H > Canadian trolls are like cockroaches, when you see one there's about a million K > other ones lurking in the dark.  Unfortunately, we have the Leader of the B > Canadian Trolls here, JF Mezei, and he has brought out his slave cockroaches to > help him do his dirty work.  >    Speaking of trolls . . .   Neil   ------------------------------  $ Date: Tue, 3 Feb 2004 10:45:12 +0100E From: "Eberhard Heuser-Hofmann" <vaxinf@chclu.chemie.uni-konstanz.de>   Subject: Re: DQDRIVER on a 500auB Message-ID: <000a01c3ea3a$6c3dd460$ad072286@chemie.unikonstanz.de>   Charles,  H There are newer versions of the dqdriver available on the HP-patch-site.   eberhard ----- Original Message -----  / From: "Charles J. Fisher" <cfisher@rhadmin.org>  To: <Info-VAX@Mvb.Saic.Com> ( Sent: Monday, February 02, 2004 10:11 PM Subject: DQDRIVER on a 500au    K > I am trying to install and boot OpenVMS v7.3 on a Seagate 40gb IDE drive. 5 > This is on a 500au that has the Cypress controller.  > K > I am able to run through the entire install procedure successfully, and I G > can break into DCL from the CD and mount the drive, but when I try to  > "boot -flags 0,0 dqa0" I get:  >  > ...  > jumping to bootstrap code & > %APB-F-MOUNT, failed to mount volume >  > halted CPU 0 >  > halt code = 5  > ...  > 4 > I have found some documentation for DQDRIVER here: > , > http://vms.tuwien.ac.at/freeware/DQDRIVER/ > J > Supposedly, booting a 500au from IDE is possible: "DQDRIVER is supportedL > with and autoconfigures on and can boot via the Cypress IDE PCI PeripheralG > Controller found on various Personal Workstation -au series systems."  > @ > Is my booting problem explained by differences in the DQDRIVERK > configuration between the CD-ROM drive and the installed VMS image on the J > hard drive? Can I insure that ENABLE-IDE.COM is called (or that DQDRIVER- > is otherwise initialized properly at boot)?  > 8 > Will I be able to boot my 500au in this configuration? > J > p.s. I have had intermittent problems with this box. Does the 500au have* >      any reputation for (un)reliability? > L >     ---------------------------------------------------------------------- ----- H >    / Charles J. Fisher                   | "Waste no more time arguing what  / L >   /  cfisher@rhadmin.org                 |  a good man should be. Be one." / ? >  /   http://rhadmin.org                  |   -Marcus Aurelius  / L > -------------------------------------------------------------------------- -    ------------------------------  * Date: Tue, 3 Feb 2004 13:59:12 +0000 (UTC), From: lewis@mazda.mitre.org (Keith A. Lewis)  Subject: Re: DQDRIVER on a 500au. Message-ID: <bvo9fg$bjg$1@newslocal.mitre.org>   "Charles J. Fisher" <cfisher@rhadmin.org> writes in article <Pine.BSO.4.53.0402021458480.28126@bart.rhadmin.org> dated Mon, 02 Feb 2004 21:11:42 GMT: J >I am trying to install and boot OpenVMS v7.3 on a Seagate 40gb IDE drive.4 >This is on a 500au that has the Cypress controller.  H I added a large IDE drive to my 500au, even though it's unsupported.  ItK works, but the performance really bites.  I would never consider this drive < for use as a system device because of the performance issue.  K Recently I have found that the large drive has real problems if served over E NFS!  I am preparing to move it to a Linux box and use the Alpha as a L client.  (It will probably be faster to send the data over 100baseT than the 500au's IDE interface!)   H Spring for a 2-4 GB SCSI drive, and keep your speed-critical files thereH (sys$sysdevice, sys$login, sys$scratch, and whatever else).  Use the IDE0 drive for your files where speed isn't an issue.  M Large IDE drives are fully-supported on the newer models such as the DS10L.     I >p.s. I have had intermittent problems with this box. Does the 500au have ) >     any reputation for (un)reliability?   K I ran a pair of them here at Mitre for several years, and one at home since J 2002.  On rare occasions I've had problems with video (Decwindows) and theF PS2 keyboard interface, but the network services have been rock-solid.  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  * Date: Tue, 3 Feb 2004 18:30:56 +0000 (UTC)* From: bleau@umtof.umd.edu (Lawrence Bleau)9 Subject: Re: Help with SSH, any tricks to get it to work? 0 Message-ID: <bvopcv$jj7$1@grapevine.wam.umd.edu>  S In article <bvmm4m$j97$1@news3.tilbu1.nb.home.nl>, Dirk Munk <munk@home.nl> writes: L >Yes, it is very easy. You can *only* use Stream-lf files with SSH/SCP/SFTP.     Thanks, that explains a lot.  F Btw, apologies for posting my questions second time under subject lineN "Need help solving SSH problems".  Years ago I had set my newsreader to ignoreM subject lines contianing "SSH", and hence blocked seeing responses to my call  for help.  That's been fixed.    Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edu    ------------------------------  $ Date: Tue, 3 Feb 2004 08:13:09 -0700B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com>  Subject: RE: Hobbyist questions?O Message-ID: <11721EF39C7D7F47A55447158274CAF79006E4@cossmgmbx01.email.corp.tld>    Didier Morandi wrote:   - > Jim, Encompass and Interex is the same org.   D Tell that to Encompass US and Interex.  Where did you get this crazy idea?  --=0D  Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991 > Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0D : I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain= D  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to= H  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do= G  not necessarily represent those of the company.  The recipient should= I  check this e-mail and any attachments for the presence of viruses.  The= A  company accepts no liability for any damage caused, directly or= 4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------   Date: 2 Feb 2004 23:32:12 -0800 ' From: icerq4a@spray.se (David Svensson) Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu = Message-ID: <734da31c.0402022332.651aad8a@posting.google.com>    Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bvlvae$iqd$1@new-usenet.uk.sun.com>...  > David Svensson wrote:  > > Andrew Harrison SUNUK Consultancy <Andrew No.Harrison No@nospamn.sun.com> wrote in message news:<bvl8sf$aqi$1@new-usenet.uk.sun.com>...  > >  > >>Keith Parris wrote:  > >>H > >>>Intel and Microsoft release new technology for HP Integrity servers > >>>by Jeff Kyle  > >>>  > >>L > >>"IA-32 EL is intended for use with supporting applications that are not K > >>performance-sensitive (for example for administration tools and system  M > >>monitoring). Primary applications, such as database software or business   > >>M > >>applications, should be native 64-bit applications designed for use with   > >>8 > >>Windows Server 2003 for 64-Bit Itanium-based System" > >>C > >>This is what Microsoft say about the product looks like a niche 
 > >>solution.  > >> > >> > >>Regards  > >>Andrew Harrison  > >> > >  > > 1 > > It does not look like a niche solution to me. J > > But you must ofcourse have something to say about it in a dull manner. > > > > > BTW. used it last week, and it looks good. For normal userD > > applications I don't experience any difference in performance toI > > running those on a normal Pentium 4 computer. It would be interesting ! > > to see how games work though.  >  > F > Funny, the white paper which shows the performance of the translatedC > code vs native shows that the performance of a 1.5 GHZ Itanium is A > 105% of a 1.6 GHz Xeon for SPECint and 98% of that of a 1.6 GHz  > Xeon for SPECfp.  C Funny, that you don't understand. I experience no difference when I B run for example Word or Excel on a 2GHz P4 or a 3GHz P4. I had the; same experience with IA32-EL. And, many interactive Windows E applications also run a good share of code that already is native, so   the speed is actually very good.  $ > Few people will be prepared to pay= > 4x or more for a CPU to get 40% or less of the performance.   ; No, I don't expect people to buy Itanium to mostly run IA32 
 applications.   + > You won't see any gamers using it either.   # Funny that you think I expect that. C Although I still think it would be interesting to see how games do.    ------------------------------   Date: 2 Feb 2004 23:35:25 -0800 ' From: icerq4a@spray.se (David Svensson) Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu = Message-ID: <734da31c.0402022335.4ade9a43@posting.google.com>    Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bvltm4$i7b$2@new-usenet.uk.sun.com>...  > David Svensson wrote:  > > Andrew Harrison SUNUK Consultancy <Andrew No.Harrison No@nospamn.sun.com> wrote in message news:<bvl8sf$aqi$1@new-usenet.uk.sun.com>...  > >  > >>Keith Parris wrote:  > >>H > >>>Intel and Microsoft release new technology for HP Integrity servers > >>>by Jeff Kyle  > >>>  > >>L > >>"IA-32 EL is intended for use with supporting applications that are not K > >>performance-sensitive (for example for administration tools and system  M > >>monitoring). Primary applications, such as database software or business   > >>M > >>applications, should be native 64-bit applications designed for use with   > >>8 > >>Windows Server 2003 for 64-Bit Itanium-based System" > >>C > >>This is what Microsoft say about the product looks like a niche 
 > >>solution.  > >> > >> > >>Regards  > >>Andrew Harrison  > >> > >  > > 1 > > It does not look like a niche solution to me. J > > But you must ofcourse have something to say about it in a dull manner. > >  > 6 > Hang on you were the person who claimed that Itanium. > was the fastest DBMS platform you have used. > 5 > MS tell you not to use the IA32 emulator for DBMS's + > so even in your book its a niche utility.  > 	 > Regards  > Andrew Harrison   F I am clueless, I don't think anyone would primarily want to run a DBMSE in IA32-EL. But there are tons of user applications that now run much  better on Itanium than before.   ------------------------------  $ Date: Tue, 3 Feb 2004 01:34:34 -0500  From: John Santos <JOHN@egh.com>Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu 6 Message-ID: <1040203010733.15796A-100000@Ives.egh.com>  < On Mon, 2 Feb 2004, Andrew Harrison SUNUK Consultancy wrote:   > David Svensson wrote:  > > Andrew Harrison SUNUK Consultancy <Andrew No.Harrison No@nospamn.sun.com> wrote in message news:<bvl8sf$aqi$1@new-usenet.uk.sun.com>...  > >  > >>Keith Parris wrote:  > >>H > >>>Intel and Microsoft release new technology for HP Integrity servers > >>>by Jeff Kyle  > >>>  > >>L > >>"IA-32 EL is intended for use with supporting applications that are not K > >>performance-sensitive (for example for administration tools and system  M > >>monitoring). Primary applications, such as database software or business   > >>M > >>applications, should be native 64-bit applications designed for use with   > >>8 > >>Windows Server 2003 for 64-Bit Itanium-based System" > >>C > >>This is what Microsoft say about the product looks like a niche 
 > >>solution.  > >> > >> > >>Regards  > >>Andrew Harrison  > >> > >  > > 1 > > It does not look like a niche solution to me. J > > But you must ofcourse have something to say about it in a dull manner. > >  > 6 > Hang on you were the person who claimed that Itanium. > was the fastest DBMS platform you have used. > 5 > MS tell you not to use the IA32 emulator for DBMS's + > so even in your book its a niche utility.  > 	 > Regards  > Andrew Harrison   ; For Andrew, some rhetorical questions.  I'm sure he'll spin 7 his usual circular reasoning about these, but unless he : actually says something original (a word, by the way, that: Andrew consistently misspells), I probably won't bother to9 answer.  Don't take my failure to respond as agreement or 
 surrender.  ? 1) Define "niche utility".  This sounds like classic FUD to me.   B 2) The VAX 11/780 (announced in 1977) was, in PDP-11 compatibility< mode, about the same performance as a PDP-11/70 (1975.)  The: VAX 8600  (1984) was significantly slower than an 11/70 in; compatibility mode.  (These were both the fastest available = VAX model when released.)  Clearly by not being significantly > faster than a much older system of half the word-size, the VAX. must have been a dismal failure in the market.  B 3) How fast does the fastest Sparc run x86 emulation?  If it can't2 run the Windows version of Oracle with competitive: performance, then by your criterion it is a niche product.     -------   ; For everyone else, an aside about migration tools.  (I know 1 a little about this subject, having written one.)   ; A migration tool doesn't have to be remotely as fast as the < original to be useful.  People port the performance-critical= bits of their applications first, and then port the remainder  at their leisure.        --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Tue, 3 Feb 2004 10:03:29 -0500* From: "Bill Todd" <billtodd@metrocast.net>Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu 2 Message-ID: <_NmdnRNHfLK2JYLdRVn-iQ@metrocast.net>  - "John Santos" <JOHN@egh.com> wrote in message 0 news:1040203010733.15796A-100000@Ives.egh.com...   ...   D > 2) The VAX 11/780 (announced in 1977) was, in PDP-11 compatibility> > mode, about the same performance as a PDP-11/70 (1975.)  The< > VAX 8600  (1984) was significantly slower than an 11/70 in= > compatibility mode.  (These were both the fastest available ? > VAX model when released.)  Clearly by not being significantly @ > faster than a much older system of half the word-size, the VAX0 > must have been a dismal failure in the market.  K You appear to be somewhat confused.  We're not discussing whether an Itanic D with a full and robust complement of native software that was firmlyL established as the market leader but that ran x86 software at less than halfL the speed of the then-fastest x86 would be a viable product in 2008 (6 yearsE after the first usable Itanic appeared, which would be the applicable F comparison with your 8600 example above):  we're discussing whether anB Itanic with relatively little native software and near-zero marketE penetration that runs x86 software at less than half the speed of the ( currently-fastest x86 is viable *today*.  G Furthermore, we're asking that question in an environment where the x86 J architecture has already been extended by AMD to 64 bits and appears aboutK to be similarly extended by Intel as well, eliminating the main reason that I VAX was created to supplement and eventually (after a decade or so on the $ market) start to replace the PDP-11.  G In other words, your analogy above appears at first glance to be poorly J conceived and utterly inapplicable.  But feel free to explain why it might not be.    - bill   ------------------------------  $ Date: Tue, 3 Feb 2004 10:13:27 -0500* From: "Bill Todd" <billtodd@metrocast.net>Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu 2 Message-ID: <L6WdnTpG5OgdJ4Ld4p2dnA@metrocast.net>  4 "David Svensson" <icerq4a@spray.se> wrote in message7 news:734da31c.0402022332.651aad8a@posting.google.com...    ...   E > Funny, that you don't understand. I experience no difference when I D > run for example Word or Excel on a 2GHz P4 or a 3GHz P4. I had the= > same experience with IA32-EL. And, many interactive Windows G > applications also run a good share of code that already is native, so " > the speed is actually very good.  K It's possible that you're the one who doesn't understand.  The matter under A discussion would seem to be whether IA32-EL makes any substantive F difference, not whether it runs Word and Excel (or similar interactiveB applications) acceptably.  For example, if Word and Excel also ranL acceptably even under the sluggish Itanic *hardware* emulator, your analysis above would be irrelevant.  H Intel appears to believe that at least some significant x86 applicationsI that Itanic needs to be attractive do *not* run at acceptable speed under K the hardware emulator.  So *those* would the the applications to examine to K see whether the somewhat increased but still uncompetitive speed of IA32-EL  makes any real difference.   - bill   ------------------------------   Date: 3 Feb 2004 08:10:41 -0800 ' From: icerq4a@spray.se (David Svensson) Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu = Message-ID: <734da31c.0402030810.475aaf64@posting.google.com>    Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bvnrmu$8uq$1@new-usenet.uk.sun.com>...  > David Svensson wrote:  > > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bvlvae$iqd$1@new-usenet.uk.sun.com>...  > >  > >>David Svensson wrote:  > >> > >>>Andrew Harrison SUNUK Consultancy <Andrew No.Harrison No@nospamn.sun.com> wrote in message news:<bvl8sf$aqi$1@new-usenet.uk.sun.com>... > >>>  > >>>  > >>>>Keith Parris wrote:  > >>>> > >>>>J > >>>>>Intel and Microsoft release new technology for HP Integrity servers > >>>>>by Jeff Kyle  > >>>>>  > >>>>N > >>>>"IA-32 EL is intended for use with supporting applications that are not M > >>>>performance-sensitive (for example for administration tools and system  O > >>>>monitoring). Primary applications, such as database software or business   > >>>>O > >>>>applications, should be native 64-bit applications designed for use with   > >>>>: > >>>>Windows Server 2003 for 64-Bit Itanium-based System" > >>>>E > >>>>This is what Microsoft say about the product looks like a niche  > >>>>solution.  > >>>> > >>>>
 > >>>>Regards  > >>>>Andrew Harrison  > >>>> > >>>  > >>> 2 > >>>It does not look like a niche solution to me.K > >>>But you must ofcourse have something to say about it in a dull manner.  > >>> ? > >>>BTW. used it last week, and it looks good. For normal user E > >>>applications I don't experience any difference in performance to J > >>>running those on a normal Pentium 4 computer. It would be interesting" > >>>to see how games work though. > >> > >>H > >>Funny, the white paper which shows the performance of the translatedE > >>code vs native shows that the performance of a 1.5 GHZ Itanium is C > >>105% of a 1.6 GHz Xeon for SPECint and 98% of that of a 1.6 GHz  > >>Xeon for SPECfp. > >  > > G > > Funny, that you don't understand. I experience no difference when I F > > run for example Word or Excel on a 2GHz P4 or a 3GHz P4. I had the? > > same experience with IA32-EL. And, many interactive Windows I > > applications also run a good share of code that already is native, so $ > > the speed is actually very good. > >  > 7 > Funny but no one buying a server is going to care how # > well Excel runs on the processor.   B Funny, that you as usual don't understand, Excel was an example, ID also think Itanium is a server platform, but I just pointed out that> IA32-EL was pretty good. But that is something you can't standE apparently. In a server it is good to know that you from time to time 5 can run user applications too, especially on Windows.    > 7 > I thought you were the one boosting Itanium precisely  > because of its performance !   Yes, I do, for database work.    ------------------------------   Date: 3 Feb 2004 10:27:10 -0800 ' From: icerq4a@spray.se (David Svensson) Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu = Message-ID: <734da31c.0402031027.2cfed005@posting.google.com>   d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<L6WdnTpG5OgdJ4Ld4p2dnA@metrocast.net>...6 > "David Svensson" <icerq4a@spray.se> wrote in message9 > news:734da31c.0402022332.651aad8a@posting.google.com...  >  > ...  > G > > Funny, that you don't understand. I experience no difference when I F > > run for example Word or Excel on a 2GHz P4 or a 3GHz P4. I had the? > > same experience with IA32-EL. And, many interactive Windows I > > applications also run a good share of code that already is native, so $ > > the speed is actually very good. > M > It's possible that you're the one who doesn't understand.  The matter under C > discussion would seem to be whether IA32-EL makes any substantive H > difference, not whether it runs Word and Excel (or similar interactiveD > applications) acceptably.  For example, if Word and Excel also ranN > acceptably even under the sluggish Itanic *hardware* emulator, your analysis > above would be irrelevant.  @ OK..., one of my points was that IA32-EL does make a difference.> Although you can/could run many applications with the hardware? emulator you did notice that it was slow. With IA32-EL I didn't D noticed any slowdown, it was like running a say 2.4GHz P4. (I cannotF make a difference between say 1.5GHz to 3Ghz P4 unless I try to play a3 game where it is more easy to spot the difference.)   J > Intel appears to believe that at least some significant x86 applicationsK > that Itanic needs to be attractive do *not* run at acceptable speed under M > the hardware emulator.  So *those* would the the applications to examine to M > see whether the somewhat increased but still uncompetitive speed of IA32-EL  > makes any real difference.  : The hardware emulator was in my opinion always a bit slow.   ------------------------------  % Date: Tue, 03 Feb 2004 10:04:13 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniue0 Message-ID: <bvnrmu$8uq$1@new-usenet.uk.sun.com>   David Svensson wrote:r > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<bvlvae$iqd$1@new-usenet.uk.sun.com>...i >  >>David Svensson wrote:h >> >>>Andrew Harrison SUNUK Consultancy <Andrew No.Harrison No@nospamn.sun.com> wrote in message news:<bvl8sf$aqi$1@new-usenet.uk.sun.com>... >>>g >>>  >>>>Keith Parris wrote:e >>>> >>>>H >>>>>Intel and Microsoft release new technology for HP Integrity servers >>>>>by Jeff Kyle  >>>>>f >>>>L >>>>"IA-32 EL is intended for use with supporting applications that are not K >>>>performance-sensitive (for example for administration tools and system sM >>>>monitoring). Primary applications, such as database software or business r >>>>M >>>>applications, should be native 64-bit applications designed for use with f >>>>8 >>>>Windows Server 2003 for 64-Bit Itanium-based System" >>>>C >>>>This is what Microsoft say about the product looks like a nicheu
 >>>>solution.  >>>> >>>> >>>>Regards3 >>>>Andrew Harrisonr >>>> >>>  >>>u0 >>>It does not look like a niche solution to me.I >>>But you must ofcourse have something to say about it in a dull manner.- >>>p= >>>BTW. used it last week, and it looks good. For normal userdC >>>applications I don't experience any difference in performance totH >>>running those on a normal Pentium 4 computer. It would be interesting  >>>to see how games work though. >> >>F >>Funny, the white paper which shows the performance of the translatedC >>code vs native shows that the performance of a 1.5 GHZ Itanium islA >>105% of a 1.6 GHz Xeon for SPECint and 98% of that of a 1.6 GHz4 >>Xeon for SPECfp. >  > E > Funny, that you don't understand. I experience no difference when I.D > run for example Word or Excel on a 2GHz P4 or a 3GHz P4. I had the= > same experience with IA32-EL. And, many interactive WindowstG > applications also run a good share of code that already is native, so " > the speed is actually very good. >   5 Funny but no one buying a server is going to care howf! well Excel runs on the processor.W  5 We have already discussed Itaniums unsuitability as al desktop platform.   7 So are you now telling me that having a server platformp8 which takes a 60% performance hit is also not a problem.  5 I thought you were the one boosting Itanium preciselya because of its performance !   Regards  Andrew HarrisonM   ------------------------------  % Date: Tue, 03 Feb 2004 10:20:47 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>eY Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itaniu$0 Message-ID: <bvnsm0$9bh$1@new-usenet.uk.sun.com>   John Santos wrote:  = > For Andrew, some rhetorical questions.  I'm sure he'll spinn9 > his usual circular reasoning about these, but unless he < > actually says something original (a word, by the way, that< > Andrew consistently misspells), I probably won't bother to; > answer.  Don't take my failure to respond as agreement ort > surrender. > A > 1) Define "niche utility".  This sounds like classic FUD to me.o > : Its the same as FX!32, without support from the ISV's whos< applications are being translated there is no point in using it except for niche apps.   @ On the best Itanium CPU you take a 60% performance hit for using@ it vs the best IA32 and you have a platform that isn't supported by you ISV.e   Does that help you.o  D > 2) The VAX 11/780 (announced in 1977) was, in PDP-11 compatibility> > mode, about the same performance as a PDP-11/70 (1975.)  The< > VAX 8600  (1984) was significantly slower than an 11/70 in= > compatibility mode.  (These were both the fastest available ? > VAX model when released.)  Clearly by not being significantlyf@ > faster than a much older system of half the word-size, the VAX0 > must have been a dismal failure in the market. >   9 x86 is a commodity you are in a market that has different  rules.  D > 3) How fast does the fastest Sparc run x86 emulation?  If it can't4 > run the Windows version of Oracle with competitive< > performance, then by your criterion it is a niche product. >   > You seem to have forgotten that SPARC has ~12000+ applications< supported on it Itanium has ~500, that is the reason why the Intel emulator exists.  ; You mentioned Oracle, currently there are 4 Oracle products : certified for HP-UX Itanium while there are ~290 certified5 for Solaris/SPARC. I don't need an emulator to deploy35 Oracle on SPARC you almost certainly need an emulatorr to do so on Itanium. > 	 > -------  > = > For everyone else, an aside about migration tools.  (I know 3 > a little about this subject, having written one.)e > = > A migration tool doesn't have to be remotely as fast as theo> > original to be useful.  People port the performance-critical? > bits of their applications first, and then port the remaindern > at their leisure.  > = If people only use the translator to move their internal appsn: then it will have failed. One of Itaniums many problems is; a vanishingly tiny ISV portfolio without a solution to thatn8 problem Itanium is destined to languish in niche markets6 like HPC where much of the code is developed in house.   Regardse Andrew Harrisone >  >    ------------------------------  % Date: Tue, 03 Feb 2004 17:10:00 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>hY Subject: Re: Intel and Microsoft provide higher 32-bit applications performance on Itanium0 Message-ID: <bvokl9$hm8$1@new-usenet.uk.sun.com>   David Svensson wrote:m  D > Funny, that you as usual don't understand, Excel was an example, IF > also think Itanium is a server platform, but I just pointed out that@ > IA32-EL was pretty good. But that is something you can't standG > apparently. In a server it is good to know that you from time to timea7 > can run user applications too, especially on Windows.s >   A Sorry but it is you who seem to be having a serious comprehension  issue.  = Excel was just the kind of app that should do ok with IA32-EL ; its an interactive application, you spend more time lookingr= at it and thinking than it spends working and it runs fine on  200-300 Mhz Pentium boxes.  < That said, realtime updates, large worksheets lots of macros9 complex recalc and you will want a faster system but that = wasn't what you did was it, you loaded Excel looked at it and 9 went hey thats cool, opened a file, saved a file and said ! thats cool again and that was it.-  : Currently there are only 4 of the ~300 Oracle applications8 certified for HP-UX/Linux Itanium, do you honestly think7 that say the DW/DM components are going to work well ono	 IA32-EL ?e  6 And does it matter if Oracle won't support it anyway ?  7 Incedentally if you can come up with examples of me not : understanding your point in the past then you are entirely welcome to air them.  4 The problem with Itanium isn't user applications its5 applications full stop as you can see from the Oracled example.  6 Without performance and without ISV support IA32-EL is6 destined to become another FX!32 except that FX!32 was quite quick.   Regardsa Andrew Harrisona   ------------------------------  % Date: Tue, 03 Feb 2004 12:37:42 -05007* From: JF Mezei <jfmezei.spamnot@istop.com>Y Subject: Re: Intel and Microsoft provide higher 32-bit applications performanceon Itanium ) Message-ID: <401FDC64.3D1AA885@istop.com>7  ( Andrew Harrison SUNUK Consultancy wrote:@ > You seem to have forgotten that SPARC has ~12000+ applications> > supported on it Itanium has ~500, that is the reason why the > Intel emulator exists.  C Does anyone really expect Windows to be more than a demo on IA64 ?    L Will the 8086 softaware emulator on IA64 be of any use for HP-UX ? (could itM provide an environment where you could boot a 8086 version of windows ? (aka:- Insignia solutions) ?u  N Or is the emulator really targetted only at a windows instance on IA64 wishing+ to run windows software compiled for 8086 ?t   ------------------------------  # Date: Tue, 03 Feb 2004 03:08:08 GMTl> From: Michael Austin <maustin@no-more-spam.firstdbasource.com>@ Subject: Re: is the current SCO denial of service attack bogus ?= Message-ID: <sgETb.12658$RY1.4992@newssvr22.news.prodigy.com>i  
 Daeron wrote:d  * > Innocents Caught in SCO-Linux Cross Fire > Peter Galli Aug 27 20030 >  > [..] > H > "Each DDoS attack aimed at SCO over the past 4 months has crippled not > only SCO, but Centershift asF > well," James Hafen, Centershift's chief technical officer and senior7 > vice president, told eWEEK in an e-mail on Wednesday.t > E > "Stepping aside from the issues of how, architecturally, this wouldo > haveA > spilled over into Centershift's domain, it should be known thatnB > bystanders are being injured as this war rages on," Hafen added. >  > [..] > F > Hafen responded by admitting that his company did indeed have issues? > with its hosting company and was taking action on that front.o > 5 > http://www.eweek.com/article2/0,3959,1233231,00.aspr >  > A > 'Stepping aside .. architecturally' ???  translation: I made anuH > unfounded statement and cannot back it up with hard facts. I wonder is) > the current DDOS attack equally bogus ?tG quite obvious the author of this thread has no clue what a DDoS attack zC can do to entire network segments.  I have identified many  forged  H emails that are indeed the MyDoom.B virus -- but since I read all email 6 on my VMS box in text mode only, they don't bother me.  C Transaltation: he is not going to discuss his network topology and tC security in a public forum and how it directly affects his company.    ------------------------------   Date: 3 Feb 2004 09:29:13 -08000' From: doug_mentohl@yahoo.co.uk (Daeron)i@ Subject: Re: is the current SCO denial of service attack bogus ?= Message-ID: <da46811d.0402030929.5f4d7824@posting.google.com>    Michael Austin <maustin@no-more-spam.firstdbasource.com> wrote in message news:<sgETb.12658$RY1.4992@newssvr22.news.prodigy.com>...   I > quite obvious the author of this thread has no clue what a DDoS attack  E > can do to entire network segments.  I have identified many  forged oJ > emails that are indeed the MyDoom.B virus -- but since I read all email 8 > on my VMS box in text mode only, they don't bother me.  E > Transaltation: he is not going to discuss his network topology and eE > security in a public forum and how it directly affects his company.0  F My point was that Centershift were having problems previously and onlyE used the alleged SCO attack as an excuse. Were anyone other companiesaA on the same network having problems ? If SCo had not announced ith would anyone else noticed ?A   ------------------------------  $ Date: Tue, 3 Feb 2004 12:09:40 +0100  From: "Dr. Dweeb" <dr@dweeb.com>C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!s, Message-ID: <bvnvhk$93n$1@news.cybercity.dk>   Bill Todd wrote:6 > "David Svensson" <icerq4a@spray.se> wrote in message9 > news:734da31c.0401311240.12ebb42a@posting.google.com...i8 >> "Bill Todd" <billtodd@metrocast.net> wrote in message0 > news:<ssKdnXdIn9OlN4fdRVn-vg@metrocast.net>...6 >>> "Brian Chase" <bdc@world.std.com> wrote in message& >>> news:bvcrr0$bco$1@pcls4.std.com...7 >>>> In article <bvbham$ald$1@news1.tilbu1.nb.home.nl>,v% >>>> Dirk Munk  <munk@home.nl> wrote:  >>>>G >>>>> What many of us have been predicting for the last couple of yearsiE >>>>> now is  going to be a reality. Intel will have 64bit x86 cpu's.l >>>>> Please read the  >>>  following+ >>>>> article from the news agency Reuters:m >>>>>n >>>>>e >>>l >>>  >dL http://www.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=4233918 >>>  ) >>>>6 >>>> A similar story is being carried on news.com too:5 >>>> ( http://news.com.com/2100-1006_3-5150336.html )l >>>>% >>>> ->  Intel shifts 64-bit emphasise >>>> ->  By Stephen Shanklandg$ >>>> ->  Staff Writer, CNET News.com >>>> ->oC >>>> ->  Intel plans to demonstrate a 64-bit revamp of its Xeon andu >>>> PentiumF >>>> ->  processors in mid-February--an endorsement of a major rival's
 >>>> strategy-> >>>> ->  and a troubling development for Intel's Itanium chip. >>>aG >>> Thanks - I hadn't seen that second article, with the explicit rumorn4 >>> about a planned Intel demo in a couple of weeks. >>> G >>> I'm not about to start prancing around in a flight suit in front of-G >>> a "Mission Accomplished!" banner, but it really is starting to look-E >>> like Truth, Justice, and at least what I'd like to think is stillrA >>> The American Way may triumph after all over strong-arm marketDB >>> tactics and spin - a lesson in the power of "Never surrender!"G >>> grassroots activism which I hope will not be lost on both corporatehF >>> and political entities in this country (though I fear it will take: >>> a few more repetitions before it *really* takes hold). >>> 
 >>> - bill >>F >> I thought you were more interested in technology (and competition). >uF > I'm certainly interested in technology, and I think that competitionF > is a good thing.  Both of those inform my attitude toward Itanic:  IB > don't believe it's very useful technology (at least not the EPICC > part:  some surrounding chip features like the cache architectureaA > merit respect), and I strongly resent the fact that despite itsaD > technological inferiority it managed to kill off superior productsF > (though that's more the fault of those who did so than of Itanic per > se). >s > ...e >y3 >> I don't know what mission you are talking about?E >rF > The mission to make sure that cHumPaq paid the appropriate price forC > the manner in which it killed Alpha - though I've always said I'daE > settle for appropriate apologies coupled with real compensation (in G > the form of dramatically increased VMS support to make up for the hiteG > it took from the Alphacide).  If if weren't for the Alphacide, Itanic G > would just have been a not-very-interesting side-show for me, since I E > doubt that anyone would still be considering it a viable competitor G > (after all, there's no indication that Intel had a clue where to take C > it after McKinley and the subsequent shrinks and cache-increases:0< > they needed the Alpha team to have had any future at all). >) > - bill  
 Spot on Bill.H  K But I sort of (shudder) have to agree with Andrew, HP and some of the execsdJ will be in deep poo very shortly, and while anything is possible, it seemsI that egos will stop a change in HighEnd Server direction at HP until longiG after its too late.  They have "personal investment" (as distinct frompL money) in the Itanic and once having stood up and shouted to the world, theyK have no where to go.  In negotiation, this is known as a "position", and itgI is very hard to move people from a "position" in which they have investedeI their "person".  The usual result is that the person has to be (to use anPL old vet saying) "disappeared", with often disastrous follow-on consequences.  H An AMD-64 strategy for VMS has always made more sense in my mind, ratherH than IA-64 - given that the head has already been removed from the AlphaD body.  This was however NEVER on the cards at HP, because HP and itsJ executives has as much invested (both money and reputation) in the EPIC asK Intel.  While Intel will not founder with the EPIC (they still own the cashaI cow), there is a very real chance that HP HighEnd division will - and thes1 only ones who will really care are the VMS folks.r  I By the way, I too would like to see a real EV79 and EV8, but I think thatoK unlikely, sadly.  Not because there are no engineers to do it (no one has a*J monopoly on brains, and there are planty of talented people at Intel & AMDG and popping out of MIT etc. who might consider going "back to the fold"yC given the right incentives), but because people like Carley have nodL conception of admission of error, and therefore of winding back to the pointD of error and moving forward in a different direction.  She is not anK engineer, she is a manipulative, eleoquent and misguided powermonger with aeJ broken vision, and she will be around for a while yet I fear - much to our
 detriment.  
 Dr. Dweeb.   ------------------------------  % Date: Tue, 03 Feb 2004 15:37:22 +0000tO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!a0 Message-ID: <bvof7i$fnk$1@new-usenet.uk.sun.com>   Dr. Dweeb wrote:  J > An AMD-64 strategy for VMS has always made more sense in my mind, ratherJ > than IA-64 - given that the head has already been removed from the AlphaF > body.  This was however NEVER on the cards at HP, because HP and itsL > executives has as much invested (both money and reputation) in the EPIC asM > Intel.  While Intel will not founder with the EPIC (they still own the cashoK > cow), there is a very real chance that HP HighEnd division will - and thea3 > only ones who will really care are the VMS folks.w >   D HP have more invested in EPIC than Intel it was after all their idea which they pitched to Intel.  D I can almost feel sorry of Intel, caught between the sharp suits andF spiffy powerpoint presentations of the HP folks and the hard realities of the market.   Regardsl Andrew Harrison    ------------------------------   Date: 3 Feb 2004 09:03:30 -0800r' From: icerq4a@spray.se (David Svensson)aC Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!a= Message-ID: <734da31c.0402030903.1ea2b572@posting.google.com>>  d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<uoidnc8RSvJE-4Hd4p2dnA@metrocast.net>...6 > "David Svensson" <icerq4a@spray.se> wrote in message9 > news:734da31c.0401311240.12ebb42a@posting.google.com...>9 > > "Bill Todd" <billtodd@metrocast.net> wrote in message 1 >  news:<ssKdnXdIn9OlN4fdRVn-vg@metrocast.net>... 8 > > > "Brian Chase" <bdc@world.std.com> wrote in message( > > > news:bvcrr0$bco$1@pcls4.std.com...: > > > > In article <bvbham$ald$1@news1.tilbu1.nb.home.nl>,( > > > > Dirk Munk  <munk@home.nl> wrote: > > > >nK > > > > > What many of us have been predicting for the last couple of yearsu	 >  now ise >  goingK > > > > > to be a reality. Intel will have 64bit x86 cpu's. Please read thes >  following/ > > > > > article from the news agency Reuters: 	 > > > > >a	 > > > > >5 > > >a > > > O >  http://www.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=4233918S > > >  ) > > > >a9 > > > > A similar story is being carried on news.com too: 8 > > > > ( http://news.com.com/2100-1006_3-5150336.html ) > > > > ( > > > > ->  Intel shifts 64-bit emphasis  > > > > ->  By Stephen Shankland' > > > > ->  Staff Writer, CNET News.com 
 > > > > ->N > > > > ->  Intel plans to demonstrate a 64-bit revamp of its Xeon and PentiumI > > > > ->  processors in mid-February--an endorsement of a major rival's  >  strategy A > > > > ->  and a troubling development for Intel's Itanium chip.e > > >aI > > > Thanks - I hadn't seen that second article, with the explicit rumors
 >  about a. > > > planned Intel demo in a couple of weeks. > > >-K > > > I'm not about to start prancing around in a flight suit in front of aaL > > > "Mission Accomplished!" banner, but it really is starting to look likeF > > > Truth, Justice, and at least what I'd like to think is still The >  AmericannK > > > Way may triumph after all over strong-arm market tactics and spin - as	 >  lessonmN > > > in the power of "Never surrender!" grassroots activism which I hope will >  notN > > > be lost on both corporate and political entities in this country (though >  IN > > > fear it will take a few more repetitions before it *really* takes hold). > > >e > > > - bill > >lG > > I thought you were more interested in technology (and competition).e > K > I'm certainly interested in technology, and I think that competition is atG > good thing.  Both of those inform my attitude toward Itanic:  I don'ttH > believe it's very useful technology (at least not the EPIC part:  someM > surrounding chip features like the cache architecture merit respect), and IpH > strongly resent the fact that despite its technological inferiority itN > managed to kill off superior products (though that's more the fault of those$ > who did so than of Itanic per se).  D I am though interested too see how EPIC will do in the future, and IF want to see Montecito and especially Tukwila. I think one ofcourse canF argue about if EPIC is good or bad, but I don't think that EPIC is theF biggest "problem", the biggest problem for Itanium is that it is a new
 architecture.e   > 4 > > I don't know what mission you are talking about? > J > The mission to make sure that cHumPaq paid the appropriate price for theJ > manner in which it killed Alpha - though I've always said I'd settle forF > appropriate apologies coupled with real compensation (in the form ofL > dramatically increased VMS support to make up for the hit it took from theM > Alphacide).  If if weren't for the Alphacide, Itanic would just have been aIN > not-very-interesting side-show for me, since I doubt that anyone would stillN > be considering it a viable competitor (after all, there's no indication thatM > Intel had a clue where to take it after McKinley and the subsequent shrinksrL > and cache-increases:  they needed the Alpha team to have had any future at > all).b >  > - bill  D OK, then I understand you a little better, and I agree the AlphacideF was not a good thing, but I don't think everyone takes it at seriouslyF as you do. ;) I do think however that Intel could go along without theC Alpha team, Intel has the engineering capacity, and even though thepF Alpha team was great, there emerge new people too, which can be great.= The future of CPU engineering does not have to rely on Alpha.l   ------------------------------  % Date: Tue, 03 Feb 2004 12:46:02 -0500>* From: JF Mezei <jfmezei.spamnot@istop.com>C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!t) Message-ID: <401FDE58.A7D823A4@istop.com>    "Dr. Dweeb" wrote:M > Intel.  While Intel will not founder with the EPIC (they still own the cashnK > cow), there is a very real chance that HP HighEnd division will - and them3 > only ones who will really care are the VMS folks.l  I The minute Intel singals to HP that it wants out of that disaster of EPICtJ proportions, HP will then negotiate a deal where HP will pay for continuedN development of IA64 for X number of years in exchange for Intel paying for theS port of HP-UX, Tandem/NSK and possibly VMS to the Intel version of the 64 bit 8086.   G In my mind, when intel abandons IA64, its enterprise 8086 offering willrL include many large-systems features that are necessary and which AMD doesn't$ (yet) have. (for instance, lockstep)  M IA64 will live for at least 3 more years even if Intel were to announce today - that it is going foll hog on the 64 bit 8086.D  N The danger of this rumour is that it may become a self fulfilling prophecy: ifG customers currently on Alpha or PaRisc think that IA64 will be replaced N shortly with the 64 bit 8086, they may decide to stick with their old platformN and not bother with the port to IA64, wait a couple of years and port directly from Alpha to 8086.h  M Now, if customers do wait and hold off the purchanses of IA64 gears, the IA64nN sales will be dismal wich will quicken Intel's announcement of the 64 bit 8086 replacing IA64.e   ------------------------------   Date: 3 Feb 2004 05:04:36 -0800o* From: jeremyshum@hotmail.com (Jeremy Shum) Subject: MyDoom.Bo= Message-ID: <e9b82ade.0402030504.7f2fd2b8@posting.google.com>n  A i am not the creator of MyDoom.B although i thnk whoever made theeB MyDoom.B virus cuz i TOTALLY HATE SCO... they think their top shitC although they suck... one thing though... is that they shuldnt haveo> aimed to hit Microsoft... Microsoft are known as the bad guyz,! although i reckon their alrite...u  ? on the other hand... SCO sucks... JUST REMEMBER:: NEVER BUY SCO.D SOFTWARE... they just wanna hit everybody down... even their frends!  F if anybody disagrees w/ me... thats fine, although that is my point of view...J  F yet again, I LOVE THE MyDoom.B virus... why wuld i wanna visit SCO.comA or MICROSOFT.com anyway? just makes it easier for me NOT to go to9 those websites...c   ------------------------------  $ Date: Tue, 3 Feb 2004 08:32:23 -0700B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com> Subject: RE: MyDoom.BhO Message-ID: <11721EF39C7D7F47A55447158274CAF7900700@cossmgmbx01.email.corp.tld>    Jeremy Shum wrote:  ? > i am not the creator of MyDoom.B although i thnk whoever made , > the MyDoom.B virus cuz i TOTALLY HATE SCO.  F Here we have a child who could not even graduate from grade school, as? evidenced by the lack of ability in using the English language.  --=0Dm Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991a> Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0De: I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain=aD  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to=kH  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do= G  not necessarily represent those of the company.  The recipient should=tI  check this e-mail and any attachments for the presence of viruses.  The=-A  company accepts no liability for any damage caused, directly or=d4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  * Date: Tue, 3 Feb 2004 18:14:41 +0000 (UTC)* From: bleau@umtof.umd.edu (Lawrence Bleau)' Subject: Need help solving SSH problemsg0 Message-ID: <bvooeh$ivq$1@grapevine.wam.umd.edu>  E I just upgraded to OpenVMS AXP V7.3-2, referred to at vmshost.umd.edueC in this message.  The main reason to upgrade was to use the new SSHLE tools.  The TCPIP version I'm running is V5.4.  A user of that systemoC is having problems, and the online manuals weren't too informative..  B I can only reproduce his efforts, since I have only one VMS systemE running ssh, so cannot do VMS-to-VMS tests.  I've edited out specifica= system and username references and substituted generic names.l  E First, my user is able to log in using ssh and slogin.  However, when L he tried to do a secure copy of a file from the VMS system to his Mac using 1 scp, he got an error about an unexpected newline:r   Mac command:5 	scp myuser@vmshost.umd.edu:hehratio.dat hehratio.dati   Output:b" myuser@vmshost.umd.edu's password:$ protocol error: unexpected <newline>  3 Note: hehratio.dat is a normal VMS ascii text file.   C Also, he could not copy a file from his Mac to the VMS system usingA0 scp either, but the error message was different:   Mac command:3 	scp 4flares.gif myuser@vmshost.umd.edu:4flares.gifm   Output:p" myuser@vmshost.umd.edu's password: OpenVMS$tty_gettty() failede  < 4flares.gif is a binary file, fixed-length 512 byte records.  ) He then tried sftp and had more problems:c   Command: 	sftp myuser@vmshost.umd.edu   Output:H  Connecting to vmshost.umd.edu..." myuser@vmshost.umd.edu's password:# sftp> get hehratio.dat hehratio.datn# Couldn't get handle: Unknown statuso	 sftp> pwdA( Remote working directory: /diskf1/lasley sftp> ls ID mismatch (4 != 5)  3 The sftp session ended after the ID mismatch error.o  F He was, though, able to copy a file from another Mac to the VMS system using sftp.o  E Also, when he tries to exit an sftp session, the session hangs and he F says he has to kill it with Ctrl-C.  This happens even when he doesn't try to copy files.  F Finally, he tells me that BBEdit has a feature that allows you to editF files on a  remote server using sftp.  He uses this feature all of theF time to edit files on his work Mac from home, so he knows it works, atF least between Macs.  He tried connecting to the VMS system with BBeditE and the program crashed.  (Yes, I know, not a VMS problem per se, bute= maybe there are Mac users out that that've encountered this.)d   ---v The next day he did more tests:S  < He is able to log in to his Mac from the VMS system via ssh.  B He is able to connect to his Mac from the VMS system via sftp, butA when he tries to transfer a file, or even do a dir the connectiontF hangs and he is unable to get out with CTRL-C or even CTRL-Y.  He getsC the same result trying to transfer to another (non-Mac) system thata	 has sftp.d  F When he tries to copy a file from the VMS system to his Mac using scp,@ the  connection hangs and i am unable to get out using CTRL-C orE CTRL-Y; his password appears to have been accepted.  Same result witho3 trying to  transfer to the other (non-Mac) account.o  ? Here's a little bit of the verbose output after his password is 	 accepted:a  D debug: Ssh2Common/SSHCOMMON.C:289: Received SSH_CROSS_AUTHENTICATED   packet from connection protocol.4 tcpip$ssh_scp2.exe:Scp2/SCP2.C:1807: notification: 0U tcpip$ssh_scp2.exe:SshFCTransfer/SSHFC_TRANSFER.C:1673: No connection yet. Waiting...25 debug: Ssh2Common/SSHCOMMON.C:722: num_channels now 1hb tcpip$ssh_scp2.exe:SshFileXferClient/SSHFILEXFERC.C:1092: ssh_file_client_receive_proc: bad STATUS  D I haven't changed the configuration files since they were installed;C let me know if you want any of my config files or settings.  I also + checked for SSH patches and didn't see any.l  " Any suggestions on how to proceed?   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edus   ------------------------------  % Date: Tue, 03 Feb 2004 12:10:52 +0000oO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> Y Subject: Re: OpenVMS vs unix security ... Andrew, the IBM guy awaits your  response!  res 0 Message-ID: <bvo34d$bg0$1@new-usenet.uk.sun.com>   Dirk Munk wrote:	 > Hi Bob,  > I > I suppose this is a reaction on some earlier message by Andrew ? Which tG > one is it ? (there are some many). If you can gibe me date and time,   > that would be nice.. >   - I posted a number of Multinet vunerabilites..d  2 For reasons better known to himself Bob has chosen4 to respond with some TCPWARE examples. Now why isn't$ that the surprise that it should be.     Regardsa Andrew Harrison1 >  > Bob Ceculski wrote:  > F >> Yes, but first redesign and rewrite your unix to cleanly catagorize >> and separate I >> Kernel Mode from Supervisor Mode and from User Mode. Three modes are aa
 >> minimumG >> for a correct ring protection system. The use of three or more rings 
 >> happens to F >> be a fully patented methodology by OpenVMS Engineering. OpenVMS has >> four.H >> OpenVMS also has 40 groups of higher mode functionality classified as >> requiring >> special named privileges.   >  >    ------------------------------  % Date: Tue, 03 Feb 2004 00:10:14 -0700r( From: Lorin Ricker <lorin@locktrack.com>9 Subject: Re: PC print server devices compatible with VMS? 8 Message-ID: <2hgu101vb08b91g37btif8p54g1g8b1v2b@4ax.com>  F Progress report, FWIW:  I bit the bullet and dropped ~$70 on a HawkingE Technologies 1-port USB Printer Server HPS1U.  It's the cutest littles@ gizmo, about 2"x2"x3/4" (the power transformer dongle is huge byD comparison!), mostly for the USB, RJ45 and power connectors.  SeemedD like a cheap enough experiment... and there seems to be enough partsF of an OS and a printer daemon in that miniscule package to do the job!  B Installation and configuration on my W2K PC was simple, and workedD right out of the box without a hitch.  It's so small, though, that IC had to stick some double-sticky velcro tape to it to dock it to thee> back of my printer so I'd not loose the darn thing on my rack.  F Nearly the same results when configuring it for VMS... defined a print queue as follows:u  0 $ INITIALIZE /QUEUE /PROCESSOR=TCPIP$TELNETSYM -     /SCHEDULE=NOSIZE  - 4     /PROTECTION=(SYSTEM:E,OWNER:D,GROUP,WORLD:RW)  -     /ON="10.0.3.101:9100"  -     /LIBRARY=SYSDEVCTL_LJ  -     LORINPRINTER $ SHOW QUEUE /FULL LORINPRINTERh= Printer queue LORINPRINTER, idle, on FOXY::"10.0.1.101:9100",bB mounted form DEFAULT /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT)0 /LIBRARY=SYSDEVCTL_LJ Lowercase  /OWNER=[SYSTEM]7 /PROCESSOR=TCPIP$TELNETSYM /PROTECTION=(S:M,O:D,G,W:RS)e /SCHEDULE=(NOSIZE)  F The /PROCESSOR=TCPIP$TELNETSYM /ON="ip_address:port" are the importantF bits.  And, it *almost* worked first time... If the test-print job wasF > 1 page, the first page(s) printed, but I had to hit the "big" buttonB on the printer's front-panel to force it to (re)gurgitate the lastC page... and for a 1-page job, the "big" button had to be pressed toi! get it to print that page at all.l  C After some puzzling with a buddy, we remembered something about thenF logical name TCPIP$TELNETSYM_SUPPRESS_FORMFEEDS, which have to do withB whether or not the TELNETSYM includes or suppresses a job-trailingA <FF> character, which most PC printers require to eject that lastiD page.  Sure enough, a bit of research in the "HP TCP/IP Services forE OpenVMS - Management" manual, Chapter 25, and specifically in Section F 25.6.4.1 "Controlling Form Feed Suppression" proves to be the relevant7 bit... and this logical name's value seems to have beenmE enhanced/expanded with TCP/IP v5.3 &ff, so do check this section out,>C even if you think you know all about this logical in past versions.i  D Once I got the right value ("18") set for this logical, VMS printingB on this Hawking Printer Server (attached to an HP2300d) is workingD great.  When considering my purchase, I also looked at comparable 1-F and 3-port devices from LinkSys, DLink, IOGear and a couple of others;F I settled on the Hawking because it was cheap(est) and looked simplestF to install from the outside of the box.  However, all the others claimA to support the same protocols (TCP/IP, IPP, NetWare, IPX/SPX, and ? more), interfaces (USB, parallel port, Ethernet 10/100 UTP) and B general operational feature so that, now that I've this experienceD under my belt, I'm pretty sure they'd all work acceptably in similar$ ways for this application/situation.  ! Hope this helps someone else out.    Lorinc   ------------------------------  $ Date: Tue, 3 Feb 2004 08:55:57 -0500  From: John Santos <JOHN@egh.com>9 Subject: Re: PC print server devices compatible with VMS?e6 Message-ID: <1040203083840.15796H-100000@Ives.egh.com>  ' On Tue, 3 Feb 2004, Lorin Ricker wrote:o  H > Progress report, FWIW:  I bit the bullet and dropped ~$70 on a HawkingG > Technologies 1-port USB Printer Server HPS1U.  It's the cutest little4B > gizmo, about 2"x2"x3/4" (the power transformer dongle is huge byF > comparison!), mostly for the USB, RJ45 and power connectors.  SeemedF > like a cheap enough experiment... and there seems to be enough partsH > of an OS and a printer daemon in that miniscule package to do the job! > D > Installation and configuration on my W2K PC was simple, and workedF > right out of the box without a hitch.  It's so small, though, that IE > had to stick some double-sticky velcro tape to it to dock it to theb@ > back of my printer so I'd not loose the darn thing on my rack. > H > Nearly the same results when configuring it for VMS... defined a print > queue as follows:  > 2 > $ INITIALIZE /QUEUE /PROCESSOR=TCPIP$TELNETSYM - >     /SCHEDULE=NOSIZE  -s6 >     /PROTECTION=(SYSTEM:E,OWNER:D,GROUP,WORLD:RW)  - >     /ON="10.0.3.101:9100"  - >     /LIBRARY=SYSDEVCTL_LJ  - >     LORINPRINTER! > $ SHOW QUEUE /FULL LORINPRINTERh? > Printer queue LORINPRINTER, idle, on FOXY::"10.0.1.101:9100",TD > mounted form DEFAULT /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT)2 > /LIBRARY=SYSDEVCTL_LJ Lowercase  /OWNER=[SYSTEM]9 > /PROCESSOR=TCPIP$TELNETSYM /PROTECTION=(S:M,O:D,G,W:RS)1 > /SCHEDULE=(NOSIZE) > H > The /PROCESSOR=TCPIP$TELNETSYM /ON="ip_address:port" are the importantH > bits.  And, it *almost* worked first time... If the test-print job wasH > > 1 page, the first page(s) printed, but I had to hit the "big" buttonD > on the printer's front-panel to force it to (re)gurgitate the lastE > page... and for a 1-page job, the "big" button had to be pressed to # > get it to print that page at all.f > E > After some puzzling with a buddy, we remembered something about thehH > logical name TCPIP$TELNETSYM_SUPPRESS_FORMFEEDS, which have to do withD > whether or not the TELNETSYM includes or suppresses a job-trailingC > <FF> character, which most PC printers require to eject that last F > page.  Sure enough, a bit of research in the "HP TCP/IP Services forG > OpenVMS - Management" manual, Chapter 25, and specifically in SectionPH > 25.6.4.1 "Controlling Form Feed Suppression" proves to be the relevant9 > bit... and this logical name's value seems to have beenoG > enhanced/expanded with TCP/IP v5.3 &ff, so do check this section out, E > even if you think you know all about this logical in past versions.n > F > Once I got the right value ("18") set for this logical, VMS printingD > on this Hawking Printer Server (attached to an HP2300d) is workingF > great.  When considering my purchase, I also looked at comparable 1-H > and 3-port devices from LinkSys, DLink, IOGear and a couple of others;H > I settled on the Hawking because it was cheap(est) and looked simplestH > to install from the outside of the box.  However, all the others claimC > to support the same protocols (TCP/IP, IPP, NetWare, IPX/SPX, and-A > more), interfaces (USB, parallel port, Ethernet 10/100 UTP) andrD > general operational feature so that, now that I've this experienceF > under my belt, I'm pretty sure they'd all work acceptably in similar& > ways for this application/situation. > # > Hope this helps someone else out.n >  > Lorin   7 A second data point that might be helpful to someone...m  < A few days ago, I got a similar setup working with an HP5550; printer and my Airport Extreme basestation, which has a USBi< printer port on it.  USB cable to the airport, and I can nowF print from my MicroVAX 3600 using TCPWare.  (VMS V7.3, TCPWARE V5.6-2)  E In my system startup file that defines print and batch queues, I put:u  / $ DEFINE/SYSTEM TCPWARE_TSSYM_HP5550 "rtr,9100" ? $ ! rtr is the IP address of the airport on my LAN, 192.168.5.1aG $ INITIALIZE/QUEUE/START/PROCESSOR=TCPWARE_TSSYM/FORM_MOUNTED=LN_FORM -kB                 /RETAIN=ERROR/DEFAULT=(NOFLAG,NOBURST,NOTRAILER) -1                 /ENABLE_GENERIC/RECORD_BLOCKING -oH                 /SEPARATE=(NOFLAG,NOBURST,NOTRAILER,RESET=(FF,DECSTR)) -$                 /library=hp_devctl -                         HP5550  @ HP_DEVCTL is a device control library I created years ago for anC HP laserjet 2 or 3, IIRC, that still works fine.  We gave the setupeB modules the same names as the equivalent ones in our SYSDEVCTL.TLBC (which I just noticed has vanished - I'll have to check into that),i> which had previously been set up for an LN02.  By creating twoG device control libraries, with the same modules but different contents, D we can use the same forms for different printers.  That's the origin! of the LN_FORM mentioned above...   C I can post the forms definitions and contents of hp_devctl.tlb, buthA I think that is pretty much irrelevent, except you might need theoD formfeed to get it to flush the printer before printing the VAX job.   -- i John Santos- Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   Date: 3 Feb 2004 18:12:19 GMT:, From: bill@gw5.cs.uofs.edu (Bill Gunshannon)% Subject: Re: Renaissance of VAX-VMS ?e9 Message-ID: <bvooa2$u4n8q$1@ID-135708.news.uni-berlin.de>   ( In article <bvh0sr$eha$1@pcls4.std.com>,( 	bdc@world.std.com (Brian Chase) writes: > E > I personally think a 64-bit extended VAX architecture incorporatinge4 > current processor technologies would be dreamy :-) >   G Personally, I would love to have a "64-bit extended PDP-11 architecture4F incorporating current processor technologies".  But I figure both have( about the same chance of happening.  :-)   bill   -- lJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   i   ------------------------------   Date: 2 Feb 2004 23:56:26 -0800e' From: icerq4a@spray.se (David Svensson)D" Subject: Re: Rumours of (CPU) Wars= Message-ID: <734da31c.0402022356.4ba5618b@posting.google.com>-  d "Bill Todd" <billtodd@metrocast.net> wrote in message news:<24CdnQ4SrrRdSYPdRVn-ug@metrocast.net>...K > 1.  The p690 off-chip cache is pretty slow (90+ ns. or so, IIRC, based onsL > that HP white paper from a year ago that I mentioned):  it helps some, butL > nowhere nearly as much as 8x as much fast *on-chip* Tukwila cache would (aI > single-core POWER4/4+ benchmark does benefit from being able to use the'J > second core's on-chip L2, but at only an additional 0.75 MB that doesn't > help *all* that much either).   ? SPECcpu (especially SPECfp) seems to gain pretty much with moredC off-chip L3 cache. I may have misunderstood this, but I thought thetC other major benefit of using the L3 cache is to reduce (memory) bus  traffic?   ------------------------------  % Date: Tue, 03 Feb 2004 10:58:10 +0000eO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>S" Subject: Re: Rumours of (CPU) Wars0 Message-ID: <bvnus3$9vt$1@new-usenet.uk.sun.com>   Rob Young wrote: > In article <bvlti0$i7b$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >  >>Rob Young wrote: >> >>>In article <bvl9sd$ba2$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:h >>>  >>>t >>>>Rob Young wrote: >>>  >>>-7 >>>	But you aren't acknowledging what Fister is saying:0 >>>rL >>>"The Santa Clara, Calif.-based chipmaker is working on chipsets and otherK >>>products and technologies that will make an Itanium-based server no more.O >>>expensive than a similar machine powered by its Xeon chip by 2007, said Mike I >>>Fister, senior vice president of the server products group at Intel. "t >>>i >>>h >>? >>Why would a reduction in the cost of doing Itanium to make it A >>competitive with Xeon (even ignoring the simple physics issues)s >>in 2007 be interesting.. >>> >>You don't get it do you. Itanium doesn't have until 2007, atA >>best it has another 12-18 months after which Intel will have to > >>think of someting else to do with it. Give it to HP etc etc. >> >  > B > 	But you are contradicting yourself.  Of course they have longerB > 	than that.  As UltraSparc sure isn't going away.  Where is the  > 	contradiction?  See below.o >   ! Do I have to spell it out to you.l  6 US is a much cheaper processor to do than Itanium, Sun5 designs the CPU's and TI FABS them. The combined cost33 base to Sun and TI is rather lower than Itanium. No.4 one in the history of CPU development has ever spent5 5 billion on a new architecture (except HP and Intel)b that is.  8 US has a huge ISV portfolio compared with Itanium 12000+3 vs ~500. Take Oracle 4 Oracle apps are certified on@6 HP-UX/Itanium the same 4 in fact that have always been" available, SPARC/Solaris has ~290.  3 So Sun doesn't have to fund porting/tuning centers,e. ISV development programs etc, Intel and HP do.  1 Finally SPARC is the market leader Itanium is thea0 market laggard and because of that our cost base is lower than Itaniums.  > : >>The last time IDC estimated the potential Itanium market7 >>they put it at 8 billion by 2007, this was before thel= >>Intel CT announcements which will only serve to reduce thiso >>number even further. >> >  > ? > 	But even at that , given Sun's declining revenue and numberss; > 	won't Itanium be a much larger market than UltraSparc in> > 	2007?  ( No or are you making another prediction. >   A >>12-18 months tops is what HP have and its very much HP on their>9 >>own. It will be interesting to see how long people likee; >>NEC and Unisys stay on Itanium once they can get CT based  >>procs. >  > ; > 	Ha.  Right.  How about bloat?  Won't an Itanium core be i7 > 	much smaller than CT, therefore more fit on a die?  f >   C Many people think that the bloat as you call it has already arrivedI in the Prescott core.>  A But the answer is that it doesn't make that much difference 64bit 6 support for Opteron added ~5% to the transistor count.  > And Xeon/Opteron are not as cache dependant as Itanium so what; you may or may not lose by adding CT you gain by not having " to build such large onchip caches.  < An AMD64 with a 1 MB cache outperformes a 6MB cached Itanium@ on SPECint, it also outperforms 1.5MB cached Itaniums on SPECfp.      A > 	Correct.  And Intel will undercut AMD like they did a year and ) > 	a half ago, a repeat of this strategy:o >   B You seemed to have missed the point, currently Intel is responding) to AMD pricing not the other way arround.r  / It also isn't a winning strategy for 2 reasons.>= 1.	Intel needs to substantially undercut AMD because the costg8 	base for building a Xeon based system is higer than the7 	equivalent AMD based system (excluding the CPU costs)..  4 	Do you honestly think that potential Intel/AMD OEMS7 	don't tally up the cost of the whole mother board/CPU.h  : 2.	Its no longer a commodity, Athlon-64 has something that5 	Xeon doesn't have, 64bit support. There is plenty of 3 	evidence that people are prepared to pay for 64bitt* 	support particularly in the server space.  9 	In the past AMD used to market their CPU's as being moret; 	cost effective than Intel with Intel marketing their CPU'sn9 	as the gold standard. The Intel Inside marketing programa
 	for example.t  9 	Intel are used to people being prepared to pay a premiumh: 	for an Intel based box, that is going to disappear unless  	they can get CT to market ASAP.  : A whole load of previous Rob posts snipped for brevity and to preserve everyones sanity.o   ------------------------------  % Date: Tue, 03 Feb 2004 13:26:53 +0000uO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>n" Subject: Re: Rumours of (CPU) Wars0 Message-ID: <bvo7iu$cvg$1@new-usenet.uk.sun.com>   David Svensson wrote: f > "Bill Todd" <billtodd@metrocast.net> wrote in message news:<24CdnQ4SrrRdSYPdRVn-ug@metrocast.net>... > K >>1.  The p690 off-chip cache is pretty slow (90+ ns. or so, IIRC, based on L >>that HP white paper from a year ago that I mentioned):  it helps some, butL >>nowhere nearly as much as 8x as much fast *on-chip* Tukwila cache would (aI >>single-core POWER4/4+ benchmark does benefit from being able to use the J >>second core's on-chip L2, but at only an additional 0.75 MB that doesn't >>help *all* that much either).d >  > A > SPECcpu (especially SPECfp) seems to gain pretty much with morehE > off-chip L3 cache. I may have misunderstood this, but I thought theaE > other major benefit of using the L3 cache is to reduce (memory) buss
 > traffic?  = SPECint and SPECfp run in ~200MB of memory, thats not the sumw: total for all the indevidual int and fp benchmarks but the> size that the benchmarks with the largest working set grow to.  ( SPECint and SPECfp are run sequentially.   You memory hierarchy goes.	 Registersa L1 L2 optionaln L3 optional  Main memory - Locals              - Remoteh   L1/L2/L3 caches are tradeoffs.  : L1 cache is generally the fastest and its generally onchip1 L2 cache is next fastest and is also often onchipa) L3 cache is slowest and generally offchipw  = Ideally you would build a dirty great L1 cache and not botherr; with L2/L3 caches but that would be difficult because cacheb= would consume die space, doing a 128MB L1 on-chip cache wouldg be impossible today.  ? So instead you balance your L1/L2/L3 caches to give the optimall< performance for the right price and with the onchip caches a< size that still allows you to FAB your processor at sensible prices etc.l  < All caches are designed to reduce the frequency that the CPU> has to access mainmemory and take a stall in the process. This. has the effect of reducing memory bus traffic.  @ Many large systems are now NUMA so memory can be local or remote; in the past some systems like the Sequent NUMA-Q have had a 9 "tertiary cache" which sits between the local NUMA memory- subsystem and remote memory.   Regards0 Andrew Harrisont   ------------------------------  $ Date: Tue, 3 Feb 2004 10:36:01 -0500* From: "Bill Todd" <billtodd@metrocast.net>" Subject: Re: Rumours of (CPU) Wars2 Message-ID: <N8ydncaOLp5XIoLdRVn-hg@metrocast.net>  4 "David Svensson" <icerq4a@spray.se> wrote in message7 news:734da31c.0402022356.4ba5618b@posting.google.com...v7 > "Bill Todd" <billtodd@metrocast.net> wrote in message . news:<24CdnQ4SrrRdSYPdRVn-ug@metrocast.net>...J > > 1.  The p690 off-chip cache is pretty slow (90+ ns. or so, IIRC, based onJ > > that HP white paper from a year ago that I mentioned):  it helps some, butsK > > nowhere nearly as much as 8x as much fast *on-chip* Tukwila cache wouldi (aK > > single-core POWER4/4+ benchmark does benefit from being able to use thePL > > second core's on-chip L2, but at only an additional 0.75 MB that doesn't! > > help *all* that much either).f > A > SPECcpu (especially SPECfp) seems to gain pretty much with moren > off-chip L3 cache.  G It's not entirely clear to me what generalization you're trying to makeoF there, especially as there aren't all that many off-chip L3 systems to examine to generalize from.i  K My point was that the speed of p690's off-chip L3 is much closer to that of!H main memory than it is to that of an on-chip L2/L3, hence increasing itsJ size has far less effect than a similar increase in size to a fast on-chipE cache.  When you compare the p690's single-CPU SPECint/fp scores withsL Itanic's, you're comparing a highly-scalable system configuration with 250 -K 290 ns. local main memory latency to a small-system-optimized configurationvI with 150 ns. main memory latency, and the p690's off-chip L3 (with 90 ns.aE latency) mostly just compensates for that fact rather than offers anymL advantage (at least if you're comparing processor performance, which is what SPECint/fp do).   1  I may have misunderstood this, but I thought thepE > other major benefit of using the L3 cache is to reduce (memory) busl
 > traffic?  I Not an issue for SPECint/fp, and probably not for anything else either (IgL don't think that the bandwidth of IBM's memory bus, which IIRC hangs off theJ off-chip L3 cache rather than off the processor, is the limiting factor inJ any workloads).  If you meant that a large L3 can reduce inter-MCM trafficH in large (more than 8-processor) systems, that's indeed true - but againC completely irrelevant to SPECint/fp (which don't stress that area).    - bill   ------------------------------   Date: 3 Feb 2004 10:54:46 -0600e+ From: young_r@encompasserve.org (Rob Young) " Subject: Re: Rumours of (CPU) Wars3 Message-ID: <pB3gBq4Z1fQw@eisner.encompasserve.org>   _ In article <N8ydncaOLp5XIoLdRVn-hg@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> writes:n >    > 3 >  I may have misunderstood this, but I thought the F >> other major benefit of using the L3 cache is to reduce (memory) bus >> traffic?c > K > Not an issue for SPECint/fp, and probably not for anything else either (I N > don't think that the bandwidth of IBM's memory bus, which IIRC hangs off theL > off-chip L3 cache rather than off the processor, is the limiting factor inL > any workloads).  If you meant that a large L3 can reduce inter-MCM trafficJ > in large (more than 8-processor) systems, that's indeed true - but againE > completely irrelevant to SPECint/fp (which don't stress that area).  >   = 	It depends on the fp footprint, if it fits inside the cache	y@ 	it will of course run faster (mostly the case).  You can googleB 	groups.google.com - this has been discussed in one form or other.   				Robt   ------------------------------   Date: 3 Feb 2004 10:52:21 -0600B+ From: young_r@encompasserve.org (Rob Young):" Subject: Re: Rumours of (CPU) Wars3 Message-ID: <oeFNjKzfKbwD@eisner.encompasserve.org>u   In article <bvnus3$9vt$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > Rob Young wrote:  C >> 	But you are contradicting yourself.  Of course they have longeroC >> 	than that.  As UltraSparc sure isn't going away.  Where is the o >> 	contradiction?  See below. >> a > # > Do I have to spell it out to you.y > 8 > US is a much cheaper processor to do than Itanium, Sun7 > designs the CPU's and TI FABS them. The combined costo5 > base to Sun and TI is rather lower than Itanium. Non6 > one in the history of CPU development has ever spent7 > 5 billion on a new architecture (except HP and Intel)a
 > that is. >   = 	While Cost of Manufacturing is fun to talk about, that isn't > 	the issue.  Total system cost that the end user pays will be.  : > US has a huge ISV portfolio compared with Itanium 12000+5 > vs ~500. Take Oracle 4 Oracle apps are certified onr8 > HP-UX/Itanium the same 4 in fact that have always been$ > available, SPARC/Solaris has ~290. > 5 > So Sun doesn't have to fund porting/tuning centers,w0 > ISV development programs etc, Intel and HP do. > 3 > Finally SPARC is the market leader Itanium is thel2 > market laggard and because of that our cost base > is lower than Itaniums.r    @ 	Wonderful.  But you are side-stepping the point - you - brought= 	up.  You brought up the fact that IDC is lowering Itanium tod= 	an $8 billion a year business in 2007.  At the rate SPARC is @ 	sliding it will be doing less than the $1.4 billion per quarter< 	I quoted by 2007.  Let's be generous and say that SPARC has; 	slide off to $4 billion per year by 2007.  That is 1/2 thegF 	size of the Itanium market by 2007.  You'll be selling faster/cheaper? 	Opteron boxes - UltraSparc won't make sense in a month or two,- 	right?-   >>  ; >>>The last time IDC estimated the potential Itanium markete8 >>>they put it at 8 billion by 2007, this was before the> >>>Intel CT announcements which will only serve to reduce this >>>number even further.b >>>t >>   >> .@ >> 	But even at that , given Sun's declining revenue and numbers< >> 	won't Itanium be a much larger market than UltraSparc in	 >> 	2007?  > * > No or are you making another prediction.  < 	See above.  Sun is sliding in market share, IBM is gaining,% 	HP is static.  Unix space of course.n     >  > B >> 	Correct.  And Intel will undercut AMD like they did a year and* >> 	a half ago, a repeat of this strategy: >>   > D > You seemed to have missed the point, currently Intel is responding+ > to AMD pricing not the other way arround.n > 1 > It also isn't a winning strategy for 2 reasons.r? > 1.	Intel needs to substantially undercut AMD because the costh: > 	base for building a Xeon based system is higer than the9 > 	equivalent AMD based system (excluding the CPU costs).n  < 	I wouldn't think so.  Especially with Dell/HP/IBM discounts? 	in hand.  Besides, they obviously can make a boatload of money . 	selling them or they would rush to the exits.  6 > 	Do you honestly think that potential Intel/AMD OEMS9 > 	don't tally up the cost of the whole mother board/CPU.m    , 	Sure.  Mom and Pop shops do.  Dell doesn't.   > ; > 	In the past AMD used to market their CPU's as being moret= > 	cost effective than Intel with Intel marketing their CPU's6; > 	as the gold standard. The Intel Inside marketing programn > 	for example.0 > ; > 	Intel are used to people being prepared to pay a premium < > 	for an Intel based box, that is going to disappear unless" > 	they can get CT to market ASAP. >   @ 	No.  Dell PCs are very affordable as are Dell based Xeon boxes.  B 	Show us how much cheaper a 4-way Sun Opteron box will be comparedG 	to a 4-way Dell Xeon box.  That will be a data point.  Do that and you A 	will watch your 4-way UltraSparc's be marginalized (and not thato1 	Celeron equivalent box you trotted out earlier).h   				Robh   ------------------------------  % Date: Tue, 03 Feb 2004 10:04:16 +0100U* From: Paul Sture <nospam@sture.homeip.net>5 Subject: Re: Slightly OT: Code Red on the rise again?s/ Message-ID: <401F7220.BC99B51@sture.homeip.net>f   Richard L. Dyson wrote:L >  > Martin P.J. Zinser wrote:dH > > just noticed in my logs that since Jan 8th I am getting requests forJ > > default.ida, which is the way Code Red attacks IIS servers. Since I doJ > > run my server on VMS I am not affected (besides of the fact that theseG > > idiotic requests spoil my log). Is this an isolated occurance or do9 > > others see this too? > L > I just checked a few of my servers that are connected to the open InternetN > and was surprised to see that they are free of Code Red fingerprints for theP > past few months.  I had been seeing lots of them up until about April of 2003. >   C I've just checked and see 20 entries since 9th Jan, most days once,r other days nothing or twice.     -- e
 Paul Sture   ------------------------------  % Date: Tue, 03 Feb 2004 10:26:29 +0000mO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 5 Subject: Re: Slightly OT: Code Red on the rise again?*0 Message-ID: <bvnt0l$9ce$1@new-usenet.uk.sun.com>   Paul Sture wrote:h > Richard L. Dyson wrote:c >  >>Martin P.J. Zinser wrote:t >>G >>>just noticed in my logs that since Jan 8th I am getting requests foreI >>>default.ida, which is the way Code Red attacks IIS servers. Since I doeI >>>run my server on VMS I am not affected (besides of the fact that these F >>>idiotic requests spoil my log). Is this an isolated occurance or do >>>others see this too?  >>L >>I just checked a few of my servers that are connected to the open InternetN >>and was surprised to see that they are free of Code Red fingerprints for theP >>past few months.  I had been seeing lots of them up until about April of 2003. >> >  > E > I've just checked and see 20 entries since 9th Jan, most days once,b > other days nothing or twice. >  > C I am still getting about 1 a day. Our Mail virus filter removes the*& offending virus but forwards the text.  . I am also getting about 1 SVEN a week as well.   Regards0 Andrew Harrisonl   ------------------------------   Date: 3 Feb 2004 04:15:51 -0800o From: rutledj@rjrt.com (Rut)& Subject: Re: stupid smtp/mapi question= Message-ID: <d88f2c7d.0402030415.3af6175f@posting.google.com>n  E So if I write a program that uses smtp calls am I talking directly to.. the mail server and bypassing exchange server?       Michael Austin <maustin@no-more-spam.firstdbasource.com> wrote in message news:<XCETb.12664$L02.4728@newssvr22.news.prodigy.com>...l > Rut wrote: > H > > Could someone explain the differences between smtp and mapi calls toI > > send mail? If my company used Microsoft exchange and I have a programsF > > making smtp calls, what exactly is going on? Do the calls still goI > > through the exchange server or do they somehow get routed through theo > > mainframe? > >  > > THanks,i > > Ruta >  > K > SMTP is a mail protocol -- servers speak this "language" to other serversr > I > MAPI is an Mail Application Programming Interface for a mail client -- rJ > your mail client (Outlook, Netscape, etc...) takes the message you have J > written and uses a client that talks to an interface on the mail server J > which then transmits the message via the SMTP protocol to the recipient  > domain mail server.s > I > So, in short MAPI is a client to/from server, SMTP is server to server.a >  >  > Michael Austin   ------------------------------  $ Date: Tue, 3 Feb 2004 13:29:56 +0100- From: "Winfried Bergmann" <dummy@empuron.com>o& Subject: Re: stupid smtp/mapi question9 Message-ID: <bvo446$u5gpa$1@ID-170759.news.uni-berlin.de>.  / "Rut" <rutledj@rjrt.com> schrieb im Newsbeitragw7 news:d88f2c7d.0402030415.3af6175f@posting.google.com...DG > So if I write a program that uses smtp calls am I talking directly too0 > the mail server and bypassing exchange server?  I No. If you want to talk to a mail server, you have to establish a network K connect and use SMTP to exchange data with this server. And the mail server F must understand SMTP. Don't know the protocols, the exchange server is using.   ------------------------------   Date: 3 Feb 2004 10:21:59 -0800   From: aleurink@chello.nl (riedo)& Subject: To buy Alphaserver 1000 4/233= Message-ID: <bec4f58c.0402031021.4a08e668@posting.google.com>   F HP/Compaq Alpha-Server 1000 4/233 with 18 GNB and 4,3 GB disks and 230A MB memory. Also a VT510 terminal, and 2 keyboards for Alpha seriea> LK46W.A2 and LK411-AA. And a Decserver 90 M with adapter. Site Holland, Price  180,00    ------------------------------  # Date: Tue, 03 Feb 2004 05:02:00 GMTe6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>) Subject: Re: Why PERL on VMS? We have DCL @ Message-ID: <a9e683885f3908cd77e67de4d48d0afc@news.teranews.com>  = In article <6ksTb.4297$SC5.39478514@news-text.cableinet.net>,r2  "David Barnes" <david@nospam-bitsolve.com> wrote:  9 nifty Perl example snipped (originallly from Hein, IIRC).n  K > Now this makes a case in point.. Programmers write code.. Operations guysYL > manage the systems on a day-to-day basis.. When the programmer has writtenD > the code above leaves. How on earth is the systems guy supposed toK > understand it? There is too much to do installing, running and maintiningM= > systems for systems managers to even start to learn pearl..c  F Or, for that matter, learn how to spell.  The name of the language is E Perl, not pearl.  Perl first really took off as a scripting language WG for system administration tasks on Unix systems, and I'm quite certain $F that VMS system managers are on the whole smart enough to use Perl as 7 well as DCL.  All the really good ones I know use both.o  H > Yes I agree that nifty little pearl scripts can be used to perform keyG > components within program suites and modules, but for controlling andlC > running the system on a day-to-day basis DCL needs to be used foro > maintainability..e  C It's true, you can't maintain code very well if you don't know the -G language it's written in, but that's as true of DCL as it is of Perl.  nE If you have a body of code in DCL and people who know it and need to  H stick with it for everything, that's fine, but don't claim that there's 8 anything other than arbitrary historical reasons for it.  K > As an MD running my company I don't want to have to pay the extra $ for ao, > pearl programmer just to run the system..   H Consider that a lot more people know Perl than know DCL and people with 6 DCL experience are probably on average more expensive.   ------------------------------  $ Date: Tue, 3 Feb 2004 10:31:05 +0100  From: "Dr. Dweeb" <dr@dweeb.com>) Subject: Re: Why PERL on VMS? We have DCL , Message-ID: <bvnpos$1lk$1@news.cybercity.dk>   Craig A. Berry wrote: ? > In article <6ksTb.4297$SC5.39478514@news-text.cableinet.net>,a4 >  "David Barnes" <david@nospam-bitsolve.com> wrote: >r; > nifty Perl example snipped (originallly from Hein, IIRC).f >aG >> Now this makes a case in point.. Programmers write code.. OperationseF >> guys manage the systems on a day-to-day basis.. When the programmerE >> has written the code above leaves. How on earth is the systems guyaA >> supposed to understand it? There is too much to do installing, G >> running and maintining systems for systems managers to even start to  >> learn pearl.. >iG > Or, for that matter, learn how to spell.  The name of the language islF > Perl, not pearl.  Perl first really took off as a scripting languageH > for system administration tasks on Unix systems, and I'm quite certainG > that VMS system managers are on the whole smart enough to use Perl asi9 > well as DCL.  All the really good ones I know use both.  > E >> Yes I agree that nifty little pearl scripts can be used to performi< >> key components within program suites and modules, but forE >> controlling and running the system on a day-to-day basis DCL needs # >> to be used for maintainability..n > D > It's true, you can't maintain code very well if you don't know theG > language it's written in, but that's as true of DCL as it is of Perl.tF > If you have a body of code in DCL and people who know it and need toA > stick with it for everything, that's fine, but don't claim that B > there's anything other than arbitrary historical reasons for it. >PF >> As an MD running my company I don't want to have to pay the extra $2 >> for a pearl programmer just to run the system.. >:D > Consider that a lot more people know Perl than know DCL and people= > with DCL experience are probably on average more expensive.:  1 People who know DCL are on the whole unemployed !R   ------------------------------  $ Date: Tue, 3 Feb 2004 11:56:50 -0500, From: "Steve Harris" <steveharris46@cox.net>) Subject: Re: Why PERL on VMS? We have DCL . Message-ID: <GpQTb.17460$u_6.10548@lakeread04>  H Anybody using Regina (REXX)?  It's simpler than Perl, but has more power than DCL ... -- s  ________________________________   Steve Harris  + "Dr. Dweeb" <dr@dweeb.com> wrote in messagep& news:bvnpos$1lk$1@news.cybercity.dk... > Craig A. Berry wrote:aA > > In article <6ksTb.4297$SC5.39478514@news-text.cableinet.net>,t6 > >  "David Barnes" <david@nospam-bitsolve.com> wrote: > >e= > > nifty Perl example snipped (originallly from Hein, IIRC)._ > >EI > >> Now this makes a case in point.. Programmers write code.. OperationssH > >> guys manage the systems on a day-to-day basis.. When the programmerG > >> has written the code above leaves. How on earth is the systems guyvC > >> supposed to understand it? There is too much to do installing,iI > >> running and maintining systems for systems managers to even start toe > >> learn pearl.. > >oI > > Or, for that matter, learn how to spell.  The name of the language is&H > > Perl, not pearl.  Perl first really took off as a scripting languageJ > > for system administration tasks on Unix systems, and I'm quite certainI > > that VMS system managers are on the whole smart enough to use Perl asH; > > well as DCL.  All the really good ones I know use both.l > >oG > >> Yes I agree that nifty little pearl scripts can be used to performa> > >> key components within program suites and modules, but forG > >> controlling and running the system on a day-to-day basis DCL needsv% > >> to be used for maintainability..e > > F > > It's true, you can't maintain code very well if you don't know theI > > language it's written in, but that's as true of DCL as it is of Perl.aH > > If you have a body of code in DCL and people who know it and need toC > > stick with it for everything, that's fine, but don't claim that D > > there's anything other than arbitrary historical reasons for it. > >-H > >> As an MD running my company I don't want to have to pay the extra $4 > >> for a pearl programmer just to run the system.. > >SF > > Consider that a lot more people know Perl than know DCL and people? > > with DCL experience are probably on average more expensive.  > 3 > People who know DCL are on the whole unemployed !1 >r >S   ------------------------------  $ Date: Tue, 3 Feb 2004 09:24:17 -0800# From: "Tom Linden" <tom@kednos.com>g) Subject: RE: Why PERL on VMS? We have DCLJ9 Message-ID: <NDEMLKKEBOIFBMJLCECIIECNCMAA.tom@kednos.com>e  F I have thought for some time of installing it but we have so much dcl,G that it would only be used for new activity.  Now if there were also ani emacs command shell with it...  D Actually, with Regina you could write your own lexical functions and you get decimal arithmetic.s  3 People who know REXX are probably not unemployed:-|      -----Original Message-----3   From: Steve Harris [mailto:steveharris46@cox.net]m*   Sent: Tuesday, February 03, 2004 8:57 AM   To: Info-VAX@Mvb.Saic.Como+   Subject: Re: Why PERL on VMS? We have DCLP      J   Anybody using Regina (REXX)?  It's simpler than Perl, but has more power   than DCL ...   --"   ________________________________     Steve Harris  -   "Dr. Dweeb" <dr@dweeb.com> wrote in messageR(   news:bvnpos$1lk$1@news.cybercity.dk...   > Craig A. Berry wrote:9C   > > In article <6ksTb.4297$SC5.39478514@news-text.cableinet.net>,E8   > >  "David Barnes" <david@nospam-bitsolve.com> wrote:   > >O?   > > nifty Perl example snipped (originallly from Hein, IIRC).=   > >)K   > >> Now this makes a case in point.. Programmers write code.. OperationstJ   > >> guys manage the systems on a day-to-day basis.. When the programmerI   > >> has written the code above leaves. How on earth is the systems guy E   > >> supposed to understand it? There is too much to do installing,aK   > >> running and maintining systems for systems managers to even start tot   > >> learn pearl..   > > K   > > Or, for that matter, learn how to spell.  The name of the language isoJ   > > Perl, not pearl.  Perl first really took off as a scripting languageL   > > for system administration tasks on Unix systems, and I'm quite certainK   > > that VMS system managers are on the whole smart enough to use Perl as =   > > well as DCL.  All the really good ones I know use both.a   > >tI   > >> Yes I agree that nifty little pearl scripts can be used to perform @   > >> key components within program suites and modules, but forI   > >> controlling and running the system on a day-to-day basis DCL needs '   > >> to be used for maintainability..    > >iH   > > It's true, you can't maintain code very well if you don't know theK   > > language it's written in, but that's as true of DCL as it is of Perl.aJ   > > If you have a body of code in DCL and people who know it and need toE   > > stick with it for everything, that's fine, but don't claim thatcF   > > there's anything other than arbitrary historical reasons for it.   > >iJ   > >> As an MD running my company I don't want to have to pay the extra $6   > >> for a pearl programmer just to run the system..   > >eH   > > Consider that a lot more people know Perl than know DCL and peopleA   > > with DCL experience are probably on average more expensive.a   >t5   > People who know DCL are on the whole unemployed !b   >I   >e       ---e(   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).B   Version: 6.0.566 / Virus Database: 357 - Release Date: 1/22/2004   ---s& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.566 / Virus Database: 357 - Release Date: 1/22/2004   ------------------------------   Date: 3 Feb 2004 05:35:31 -0800C. From: dieter.rossbach@gmx.de (dieter rossbach) Subject: ZIP/UNZIP= Message-ID: <e1d40caf.0402030535.1c53f602@posting.google.com>W  > I have a problem with ZIP 2.3 on my VMS/ALPHA system. I try do4 compress savesets and not all of them are processed:   $ dir *.bckA  = 19_B5_V_040201.BCK;1                     5084730   1-FEB-2004E 19:08:04.36,= 19_B5_V_040201.ZIP;1                           1   3-FEB-2004O 13:07:22.04 = 21_A5_V_040131.BCK;1                     9608067  31-JAN-2004R 19:07:57.04 = 21_A5_V_040131.ZIP;1                      333616   3-FEB-2004  13:21:56.57a  @ file 21_A5-V_040131.BCK went into 21_A5_V_040131.ZIP without any@ problems, trying the same with 19_B5_V_040201.BCK;1 resulted in:  0 $ zip "-V" 19_A5_V_040131.zip 19_A5_V_040131.BCK(   adding: 19_A5_V_040131.BCK (stored 0%)  B The backup file is OK, I can do BACKUP/LIST and /RESTORE from thatB file and it was NOT accessed by any other process when I tried the ZIP.   My zip version:n   Zip 2.3 (November 29th 1999).    Regards    Dieter   ------------------------------  # Date: Tue, 03 Feb 2004 16:22:23 GMTf9 From: Hein van den Heuvel <hein_netscape@eps.zko.dec.com>n Subject: Re: ZIP/UNZIP/ Message-ID: <401FCA3C.65663D63@eps.zko.dec.com>   C fyi... Dieter also submitted this problem to the OpenVMS itrc forumS where . it received soem comments and further details.  K http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=418827i   Hein..   dieter rossbach wrote:  @ > I have a problem with ZIP 2.3 on my VMS/ALPHA system. I try do6 > compress savesets and not all of them are processed:   ------------------------------   Date: 3 Feb 2004 11:04:43 +0100 K From: pmoreau@ath.cena.fr (Patrick MOREAU, CENA Athis, Tel: 01.69.57.68.40)n( Subject: [DS15] New sound card supported! Message-ID: <lpyKrXZZDQae@sinead>t   Hi,a  E Looking at HP site, I see there is now a sound card supported on DS15eN alphastation/server, but only for Tru64 Unix (with Multimedia services v 3.4).& Good news. Part Number is 3X-AVH15-AA.  L Is a VMS support planned in the future ? I plan to replace my home DS10 by a' DS15 but I need a supported sound card.    TIAi   Patrickt --O ===============================================================================-N pmoreau@ath.cena.fr  (CENA)      ______      ___   _          (Patrick MOREAU)4 moreau_p@decus.fr (DECUS)       / /   /     / /|  /|J CENA/Athis-Mons FRANCE         / /___/     / / | / |   __   __   __   __  N BP 205                        / /         / /  |/  |  |  | |__| |__  |__| |  |N 94542 ORLY AEROGARE CEDEX    / /   ::    / /       |  |__| | \  |__  |  | |__|N http://www.ath.cena.fr/~pmoreau/            http://www.multimania.com/pmoreau/O ===============================================================================o   ------------------------------   End of INFO-VAX 2004.067 ************************