1 INFO-VAX	Thu, 11 Apr 2002	Volume 2002 : Issue 200       Contents: 4000/60 or 90 HW Manual  Re: 4000/60 or 90 HW Manual  Re: 4000/60 or 90 HW Manual  RE: 4000/60 or 90 HW Manual  a new server Re: a new server Re: a new server Re: a new server Re: a new server RE: a new server  Re: Alphaserver EV7 announcement  Re: Alphaserver EV7 announcement  Re: Alphaserver EV7 announcement  Re: Alphaserver EV7 announcement& Re: ANN: HGFTP V3.0-1 is now available. Announcing ht://Dig 3.1.6 Release1 for OpenVMS2 Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMS2 Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMS2 Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMS0 Re: another hint of post-merger direction, maybe0 Re: another hint of post-merger direction, maybe0 Re: another hint of post-merger direction, maybe) Re: Best DCL programm you've ever written  Re: C issues... ! Re: can VMS use these SCSI disks? * Re: Can't change resolution on ELSA Gloria Re: CLD problem (flight), determine display server number for SSH X11? Re: Evolution of the DEC mouse RE: Evolution of the DEC mouse Re: Evolution of the DEC mouse EWA-bootdriver messages  Re: EWA-bootdriver messages   Re: Fearless VMS Prognostication  Re: Fearless VMS Prognostication  Re: Fearless VMS Prognostication  Re: Fearless VMS Prognostication  Re: Fearless VMS Prognostication  Re: Fearless VMS Prognostication% Re: file structure allocation failure 1 free speech (was: RE: More positive OpenVMS news)  Re: Gartners paid off by IBM!  Re: Gartners paid off by IBM!  GOV GRANTS, NEVER REPAY.* Re: Help! VT100 emulators - some problems.* Re: Help! VT100 emulators - some problems.* Re: Help! VT100 emulators - some problems.* Re: Help! VT100 emulators - some problems.( HP/Compaq - smoking gun or grassy knoll?, Re: HP/Compaq - smoking gun or grassy knoll?- Re: I was right!  Alpha will live in Itanium! - Re: I was right!  Alpha will live in Itanium! - Re: I was right!  Alpha will live in Itanium!  Re: IA64 is not the VAX  Re: IA64 is not the VAX  Re: IA64 is not the VAX % Re: Insufficient virtual memory error % Re: Insufficient virtual memory error % Re: Insufficient virtual memory error  Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles' Re: Just another linux nightmare day #2 3 Mail system scalability (Was: Re: Itanium troubles)  Re: Memory Corruption  Re: Memory Corruption  Re: Memory Corruption  Re: Memory Corruption 7 Merger problems was Re: WIreless networking + VMS/Tru64 ; Re: Merger problems was Re: WIreless networking + VMS/Tru64  Re: mice infestation Re: mice infestation RE: mice infestation Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news Re: More positive OpenVMS news7 NEW Import Department - for DEPB  DFRC  SIL & DUTY FREE  northernlight.com response time " Polycenter Agents for Windows 20000 Re: Poor Ebay ... should have got a VMS cluster!0 Re: Poor Ebay ... should have got a VMS cluster!) Re: Predictions - just for the hell of it ) Re: Predictions - just for the hell of it ) Re: Predictions - just for the hell of it ) Re: Predictions - just for the hell of it 2 Re: Problems writing blank tape in TZ86 tape drive2 Re: Problems writing blank tape in TZ86 tape drive/ Q: BROTHER Laser via DCPS+LAT - Any Experience?  Reboot after adding quorum disk ( Redirect SYS$OUTPUT for detached process Re: remote mailbox Re: remote mailbox4 Re: Spring Forward, Fall Back - unfortunately not :- Start Windows program from VMS? * Re: SW Command Console for Raid array 3000I Re: TCPIP V5.0A VERSUS V5.1 coding problem with "network protocol  error" # Re: WIreless networking + VMS/Tru64 # Re: WIreless networking + VMS/Tru64 # Re: WIreless networking + VMS/Tru64 # Re: WIreless networking + VMS/Tru64   F ----------------------------------------------------------------------  % Date: Thu, 11 Apr 2002 06:34:46 -0700 # From: "Tom Linden" <tom@kednos.com>   Subject: 4000/60 or 90 HW Manual9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEOMELAA.tom@kednos.com>   1 Is there one online?  Anybody know the purpose of 6 the D15 connector next to the Mouse and KB connectors?   ------------------------------  # Date: Thu, 11 Apr 2002 13:54:59 GMT # From: "John Smith" <a@nonymous.com> $ Subject: Re: 4000/60 or 90 HW ManualE Message-ID: <TAgt8.1353$nA1.658@news02.bloor.is.net.cable.rogers.com>   " From the S&O catalog, Oct-Dec. 91,, that port is a 'Remote Mouse/Keyboard Port'.  8 You'd use a BC13M-10 keyboard/mouse extender cable - 10'        . "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJIEEOMELAA.tom@kednos.com... 3 > Is there one online?  Anybody know the purpose of 8 > the D15 connector next to the Mouse and KB connectors?   ------------------------------  % Date: Thu, 11 Apr 2002 10:12:10 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> $ Subject: Re: 4000/60 or 90 HW Manual, Message-ID: <a945v4$bup5$1@lead.zk3.dec.com>   Tom Linden wrote in message ... 2 >Is there one online?  Anybody know the purpose of7 >the D15 connector next to the Mouse and KB connectors?   J Hmmm.  If I remember correctly (and the VS4000 is going back a long while)D there is a cable that can be plugged in there that terminates with aK retangular box, to which you can plug in the KB and mouse.  That allows the + system box to be further from the KB/Mouse.    ------------------------------  % Date: Thu, 11 Apr 2002 08:16:03 -0700 # From: "Tom Linden" <tom@kednos.com> $ Subject: RE: 4000/60 or 90 HW Manual9 Message-ID: <CIEJLCMNHNNDLLOOGNJICEPCELAA.tom@kednos.com>   E I thought so, it is the same as on the 3000/300 AXP boxes.  BTW, this K all because we support PL/I back to VAX VMS 5.5-2 and Ultrix 4.2, otherwise  this gear we be gone.    > -----Original Message-----< > From: Fred Kleinsorge [mailto:kleinsorge@star.zko.dec.com]( > Sent: Thursday, April 11, 2002 7:12 AM > To: Info-VAX@Mvb.Saic.Com & > Subject: Re: 4000/60 or 90 HW Manual >  >  > ! > Tom Linden wrote in message ... 4 > >Is there one online?  Anybody know the purpose of9 > >the D15 connector next to the Mouse and KB connectors?  > L > Hmmm.  If I remember correctly (and the VS4000 is going back a long while)F > there is a cable that can be plugged in there that terminates with aB > retangular box, to which you can plug in the KB and mouse.  That > allows the- > system box to be further from the KB/Mouse.  >  >  >  >    ------------------------------  % Date: Thu, 11 Apr 2002 15:14:52 +0200 , From: "Hamid Bourchi" <hbourchi@hotmail.com> Subject: a new server - Message-ID: <a94282$6j$1@odysseus.uci.kun.nl>    Hello,  J After experincing some serius performance problems during last 2 weeks, we: are considering to buy a new Alphaserver. Our options are:? 1 - a second hand Alpha 1200. 2 processor 400 MHZ and 1 GB mem. = 2 - a new Alphaserver DS20E 1 processor 833 MHZ and 1 GB mem.   7 Right now we are running with following configurations: 	 hardware: ' Alpha 1000A 233 Mhz processor, 512 mem.    sofware: OpenVMS 7.2  Oracle 8.1.7 as database% Oracle designer, forms and reports 6.   D In the end of 2002 our system has to be acceable for about 50 users.L Has anyone experince with those servers? Considering the high price of these" machines we appricate any respons.     Regards,
 Hamid Bourchi    ------------------------------  % Date: Thu, 11 Apr 2002 16:37:31 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>  Subject: Re: a new server ' Message-ID: <3CB59FAB.5243DA3E@aaa.com>   = I think the 1200 series are mostly great value for the money. ; I'v used a number of AS1200/533, and been very pleased with > them. It's big step up from the 1000A, anyway. The DS20 series; are of course even better, but the step from 1000A->1200 is  bigger then from 1200->DS20.   According to :? http://www.compaq.com/alphaserver/performance/perf_by_name.html % the relative performance values are :    1000A/233         :  350/ 1200/400 2 CPU    : 1940  (5.5 times the 1000A) E DS20E/833         : 3100  (1.6 times the 1200 or 8.8 times the 1000A)   ? You don't note what prices you'v got for the 1200 and the DS20, < but I'd guess that, since the 1200 is a used one, it must be< much cheaper then the DS20. I'd put my monny in an extra 1GB( mem in the AS1200. And maybe extra disk.  = In Sweden I'd probably get 5 used AS1200 for 1 new DS20E/833.   / If you can, look for a used AS1200 533 Mhz box.   F Btw, the 50 "users", are they developers or end (client-based) users ?    Jan-Erik Sderholm.    Hamid Bourchi wrote: >  > Hello, > L > After experincing some serius performance problems during last 2 weeks, we< > are considering to buy a new Alphaserver. Our options are:A > 1 - a second hand Alpha 1200. 2 processor 400 MHZ and 1 GB mem. ? > 2 - a new Alphaserver DS20E 1 processor 833 MHZ and 1 GB mem.  > 9 > Right now we are running with following configurations:  > hardware: ) > Alpha 1000A 233 Mhz processor, 512 mem.  > 
 > sofware:
 > OpenVMS 7.2  > Oracle 8.1.7 as database' > Oracle designer, forms and reports 6.  > F > In the end of 2002 our system has to be acceable for about 50 users.N > Has anyone experince with those servers? Considering the high price of these$ > machines we appricate any respons. > 
 > Regards, > Hamid Bourchi    ------------------------------  % Date: Thu, 11 Apr 2002 11:13:11 -0400 1 From: Michael Austin <maustin@firstdbasource.com>  Subject: Re: a new server 2 Message-ID: <3CB5A807.2A37833C@firstdbasource.com>   Jan-Erik Sderholm wrote:  > ? > I think the 1200 series are mostly great value for the money. = > I'v used a number of AS1200/533, and been very pleased with @ > them. It's big step up from the 1000A, anyway. The DS20 series= > are of course even better, but the step from 1000A->1200 is  > bigger then from 1200->DS20. >  > According to :A > http://www.compaq.com/alphaserver/performance/perf_by_name.html ' > the relative performance values are :  >  > 1000A/233         :  3501 > 1200/400 2 CPU    : 1940  (5.5 times the 1000A) G > DS20E/833         : 3100  (1.6 times the 1200 or 8.8 times the 1000A)  > A > You don't note what prices you'v got for the 1200 and the DS20, > > but I'd guess that, since the 1200 is a used one, it must be> > much cheaper then the DS20. I'd put my monny in an extra 1GB* > mem in the AS1200. And maybe extra disk. > ? > In Sweden I'd probably get 5 used AS1200 for 1 new DS20E/833.  > 1 > If you can, look for a used AS1200 533 Mhz box.  > H > Btw, the 50 "users", are they developers or end (client-based) users ? >  > Jan-Erik Sderholm.  >  > Hamid Bourchi wrote: > > 
 > > Hello, > > N > > After experincing some serius performance problems during last 2 weeks, we> > > are considering to buy a new Alphaserver. Our options are:C > > 1 - a second hand Alpha 1200. 2 processor 400 MHZ and 1 GB mem. A > > 2 - a new Alphaserver DS20E 1 processor 833 MHZ and 1 GB mem.  > > ; > > Right now we are running with following configurations: 
 > > hardware: + > > Alpha 1000A 233 Mhz processor, 512 mem.  > >  > > sofware: > > OpenVMS 7.2  > > Oracle 8.1.7 as database) > > Oracle designer, forms and reports 6.  > > H > > In the end of 2002 our system has to be acceable for about 50 users.P > > Has anyone experince with those servers? Considering the high price of these& > > machines we appricate any respons. > >  > > Regards, > > Hamid Bourchi     D Isn't Oracle charging their license fees by the CPU clock rate?  The DS20 is a very nice machine.  A You didn't say how much disk space, but from a description of you H current system is not that much.  What kind of performance problems?  ItE may not have anything to do with the CPU/Mem, but could be related to  the disk configuration.  --   Regards,  7 Michael Austin            Registered Linux User #261163 7 First DBA Source, Inc.    http://www.firstdbasource.com  Sr. Consultant 704-947-1089 (Office)  704-236-4377 (Mobile)    ------------------------------  % Date: Thu, 11 Apr 2002 17:24:30 +0200 , From: "Hamid Bourchi" <hbourchi@hotmail.com> Subject: Re: a new server . Message-ID: <a949r5$458$1@odysseus.uci.kun.nl>   50 end users and 3 developers.5 we got a indication price of 15.000 uro for Alpha1200 8 exclusive harddisks, and about 30,000 uro for DS20E/833.    5 "Jan-Erik Sderholm" <aaa@aaa.com> schreef in bericht ! news:3CB59FAB.5243DA3E@aaa.com... ? > I think the 1200 series are mostly great value for the money. = > I'v used a number of AS1200/533, and been very pleased with @ > them. It's big step up from the 1000A, anyway. The DS20 series= > are of course even better, but the step from 1000A->1200 is  > bigger then from 1200->DS20. >  > According to :A > http://www.compaq.com/alphaserver/performance/perf_by_name.html ' > the relative performance values are :  >  > 1000A/233         :  3501 > 1200/400 2 CPU    : 1940  (5.5 times the 1000A) G > DS20E/833         : 3100  (1.6 times the 1200 or 8.8 times the 1000A)  > A > You don't note what prices you'v got for the 1200 and the DS20, > > but I'd guess that, since the 1200 is a used one, it must be> > much cheaper then the DS20. I'd put my monny in an extra 1GB* > mem in the AS1200. And maybe extra disk. > ? > In Sweden I'd probably get 5 used AS1200 for 1 new DS20E/833.  > 1 > If you can, look for a used AS1200 533 Mhz box.  > H > Btw, the 50 "users", are they developers or end (client-based) users ? >  > Jan-Erik Sderholm.  >  > Hamid Bourchi wrote: > > 
 > > Hello, > > K > > After experincing some serius performance problems during last 2 weeks,  we> > > are considering to buy a new Alphaserver. Our options are:C > > 1 - a second hand Alpha 1200. 2 processor 400 MHZ and 1 GB mem. A > > 2 - a new Alphaserver DS20E 1 processor 833 MHZ and 1 GB mem.  > > ; > > Right now we are running with following configurations: 
 > > hardware: + > > Alpha 1000A 233 Mhz processor, 512 mem.  > >  > > sofware: > > OpenVMS 7.2  > > Oracle 8.1.7 as database) > > Oracle designer, forms and reports 6.  > > H > > In the end of 2002 our system has to be acceable for about 50 users.J > > Has anyone experince with those servers? Considering the high price of these & > > machines we appricate any respons. > >  > > Regards, > > Hamid Bourchi    ------------------------------  % Date: Thu, 11 Apr 2002 17:31:26 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>  Subject: Re: a new server ' Message-ID: <3CB5AC4E.20E28831@aaa.com>    OK. 2 Hard to say without the specs of each box, but I'd3 definitly thought of a larger difference beteween a 4 AS1200/400 and a DS20E/855. A second hand CPU for a 5 AS 1200/533 is about 2.500 EUR in Sweden, and a whole : AS1200/533, single CPU, as less then 10.000 EUR, as far as I can remember (in Sweden).   A *Personaly*, I think the price of the AS1200/400 is a bit high...    Jan-Erik Sderholm.    Hamid Bourchi wrote: >   > 50 end users and 3 developers.7 > we got a indication price of 15.000 uro for Alpha1200 : > exclusive harddisks, and about 30,000 uro for DS20E/833. >    ------------------------------  % Date: Thu, 11 Apr 2002 11:47:22 -0400 + From: "Main, Kerry" <Kerry.Main@Compaq.com>  Subject: RE: a new server T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF401AB1E64@kaoexc01.americas.cpqcorp.net>  H >>> Isn't Oracle charging their license fees by the CPU clock rate?  <<<  > Actually, the new Oracle model is based on "per cpu" model.=20  F Therefore, from an Oracle license perspective, the smaller number of =G CPU's that will adequately address all performance issues the better. = B Depending on the application, the number of cpu's is the type of =7 decision that can be a huge impact on the overall cost.X   Regardsr  
 Kerry Main Senior Consultant  Compaq Canada Corp.  Professional Services0 Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----; From: Michael Austin [mailto:maustin@firstdbasource.com]=20r Sent: April 11, 2002 11:13 AMr To: Info-VAX@Mvb.Saic.Coml Subject: Re: a new server        Jan-Erik S=F6derholm wrote:e >=20J > I think the 1200 series are mostly great value for the money. I'v used =  J > a number of AS1200/533, and been very pleased with them. It's big step =  J > up from the 1000A, anyway. The DS20 series are of course even better,=20? > but the step from 1000A->1200 is bigger then from 1200->DS20.r >=20 > According to :=20nA > http://www.compaq.com/alphaserver/performance/perf_by_name.htmlt' > the relative performance values are :o >=20 > 1000A/233         :  3501 > 1200/400 2 CPU    : 1940  (5.5 times the 1000A) G > DS20E/833         : 3100  (1.6 times the 1200 or 8.8 times the 1000A)n >=20H > You don't note what prices you'v got for the 1200 and the DS20, but=20J > I'd guess that, since the 1200 is a used one, it must be much cheaper=20J > then the DS20. I'd put my monny in an extra 1GB mem in the AS1200. And =   > maybe extra disk.n >=20? > In Sweden I'd probably get 5 used AS1200 for 1 new DS20E/833.e >=201 > If you can, look for a used AS1200 533 Mhz box.i >=20H > Btw, the 50 "users", are they developers or end (client-based) users ? >=20 > Jan-Erik S=F6derholm.s >=20 > Hamid Bourchi wrote: > >i
 > > Hello, > >OG > > After experincing some serius performance problems during last 2=20eJ > > weeks, we are considering to buy a new Alphaserver. Our options are: =  J > > 1 - a second hand Alpha 1200. 2 processor 400 MHZ and 1 GB mem. 2 -=20= > > a new Alphaserver DS20E 1 processor 833 MHZ and 1 GB mem.  > > ; > > Right now we are running with following configurations: 
 > > hardware:a+ > > Alpha 1000A 233 Mhz processor, 512 mem.I > >m > > sofware: > > OpenVMS 7.2e > > Oracle 8.1.7 as database) > > Oracle designer, forms and reports 6.l > >rJ > > In the end of 2002 our system has to be acceable for about 50 users. =  J > > Has anyone experince with those servers? Considering the high price=20/ > > of these machines we appricate any respons.  > >u > > Regards, > > Hamid Bourchi     F Isn't Oracle charging their license fees by the CPU clock rate?  The = DS20 is a very nice machine.  C You didn't say how much disk space, but from a description of you =eJ current system is not that much.  What kind of performance problems?  It =G may not have anything to do with the CPU/Mem, but could be related to =s the disk configuration.e --=20a Regards,  7 Michael Austin            Registered Linux User #261163n7 First DBA Source, Inc.    http://www.firstdbasource.comt Sr. Consultant 704-947-1089 (Office)y 704-236-4377 (Mobile)o   ------------------------------   Date: 11 Apr 2002 03:35 CSTm' From: carl@gerg.tamu.edu (Carl Perkins) ) Subject: Re: Alphaserver EV7 announcementn- Message-ID: <11APR200203354584@gerg.tamu.edu>M  , David Froble <davef@tsoft-inc.com> writes...J }Two questions come to mind. Is there anything in EV7 that would offer anyI }benefits over EV6 in a single processor system?  Is there any penalty in I }continuing to fab EV6 CPUs in order to continue building DS10s and such?s }  }DaveM  C The on-chip memory controllers for 8 channels of Rambus memory giveB@ the EV7 a LOT more memory bandwidth (and probably lower latency,@ although Rambus memory isn't the best in this area) than any EV61 system has per CPU (especially the low end DS10).M  F A single processor EV7 system would be great. It would just be wastingL the interprocessor communication channels and multiprocessor synchronization stuff.  D They are supposedly going to keep making the EV68 and may do another shrink or two of it.   --- Carl   ------------------------------    Date: 11 Apr 2002 07:45:15 -0500- From: koehler@encompasserve.org (Bob Koehler)t) Subject: Re: Alphaserver EV7 announcemente3 Message-ID: <JuO1uLCfqzzI@eisner.encompasserve.org>l  f In article <3CB44B17.B0A4F9B9@firstdbasource.com>, Michael Austin <maustin@firstdbasource.com> writes: > Bob Koehler wrote: >> . >> In article <rdeininger-0804022122240001@1cust111.tnt2.nashua.nh.da.uu.net>, rdeininger@mindspring.com (Robert Deininger) writes:e3 >> > Compaq has announced the EV7 alphaserver line:  >> >6 >> >    http://www.compaq.com/hps/announce/apr_02.html >> >% >> > A MARVELous picture is included.  >> 0E >>    So is the DS10 still the entry level server?  Looks like it is.o >> HH >>    I've got a formerly multi-project DEC 10000/610 I need to replace.I >>    A DS10 has more CPU power, similar I/O thoughput, and it looks like < >>    I can fit the disks and memory in the same little box. >> a, >>    Now where did I put that sales number? > H > Or if they will support the 1u DS10L, you can get a whole rack of them1 > with Storage Works or SAN.  Neat configuration.s >   F    I only need one.  I'll go for the slightly higher disk storage that     will fit inside the 3U model.   ------------------------------  % Date: Thu, 11 Apr 2002 09:17:18 -0400m5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>o) Subject: Re: Alphaserver EV7 announcementn, Message-ID: <a942o8$bvf5$1@lead.zk3.dec.com>  B Carl Perkins wrote in message <11APR200203354584@gerg.tamu.edu>...- >David Froble <davef@tsoft-inc.com> writes...rK >}Two questions come to mind. Is there anything in EV7 that would offer any J >}benefits over EV6 in a single processor system?  Is there any penalty inJ >}continuing to fab EV6 CPUs in order to continue building DS10s and such? >} >}Dave >nD >The on-chip memory controllers for 8 channels of Rambus memory giveA >the EV7 a LOT more memory bandwidth (and probably lower latency, A >although Rambus memory isn't the best in this area) than any EV6w2 >system has per CPU (especially the low end DS10). > G >A single processor EV7 system would be great. It would just be wasting-= >the interprocessor communication channels and multiprocessorn synchronizationc >stuff.j >:E >They are supposedly going to keep making the EV68 and may do another. >shrink or two of it.h >h  H EV68 is cheaper, and fast.  EV7 is more expensive, and faster.  A lot ofF what makes it expensive will be wasted in a 1p system, you make up the1 expense when you start making MP systems with it.    ------------------------------    Date: 11 Apr 2002 09:09:14 -0500+ From: young_r@encompasserve.org (Rob Young)X) Subject: Re: Alphaserver EV7 announcement03 Message-ID: <x$eBQeOKl7$o@eisner.encompasserve.org>   W In article <11APR200203354584@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:e. > David Froble <davef@tsoft-inc.com> writes...L > }Two questions come to mind. Is there anything in EV7 that would offer anyK > }benefits over EV6 in a single processor system?  Is there any penalty inwK > }continuing to fab EV6 CPUs in order to continue building DS10s and such?e > }  > }Daveh > E > The on-chip memory controllers for 8 channels of Rambus memory give,B > the EV7 a LOT more memory bandwidth (and probably lower latency,B > although Rambus memory isn't the best in this area) than any EV63 > system has per CPU (especially the low end DS10).  >   ; 	I believe 12.6 or 12.8 GByte/second bandwidth.  Latency isyD 	much better.  Single EV7 CPU latency numbers are 75 ns load-to-use.> 	Typical single processor system latencies are on the order of: 	100+ ns best-in-class, I don't know EV6 numbers and don't? 	believe I've seen then.  Just going by recent comp.arch posts:   \ http://groups.google.com/groups?selm=a67n4v%24jg8%241%40ausnews.austin.ibm.com&output=gplain  A "The fastest current uniprocessor systems have actual load-to-usef( memory latency of slightly over 100 ns."   >hH > A single processor EV7 system would be great. It would just be wastingN > the interprocessor communication channels and multiprocessor synchronization > stuff. >   B 	Ah... but still, I believe that would have about twice the memory 	bandwidth of Privateer!   				Robe   ------------------------------  # Date: Thu, 11 Apr 2002 15:09:22 GMTr- From: goathunter@goatley.com (Hunter Goatley)a/ Subject: Re: ANN: HGFTP V3.0-1 is now available 1 Message-ID: <3cb5a0b3.165824863@news.process.com>e  H On Thu, 11 Apr 2002 06:27:17 +0200, Paul Sture <p_sture@decus.ch> wrote:   >Hunter Goatley wrote: >>   >A really good pitch for HGFTP.. >/ Thanks.e  I >But when I try to look at it using Mozilla, it pops up a dialog box with0C >the title "Alert" with the contents "VMS TCPware V5.5-3". The mainA
 >display says  > I ftp.process.com is running the TCPware FTP server, and this behavior fromaH Mozilla was discussed earlier in this thread.  Mozilla does a SYST, getsH the result and displays the Alert box, then does nothing else.  (The FTPI log files confirm this---after the SYST result, it drops the connection.)e  L Changing the SYST output so that it doesn't mention VMS at all does nothing,I which makes me believe that Mozilla's FTP is just severely broken somehownJ (as someone else in this thread suggested).  (Not to mention that NetscapeN V6.2, based on Mozilla, gets an accvio when talking to HGFTP in UNIX-emulation mode.)   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/ 8 goathunter@goatley.com    http://www.goatley.com/hunter/< New Robert R. McCammon site: http://www.RobertRMcCammon.com/   ------------------------------  % Date: Thu, 11 Apr 2002 11:17:52 +0200d< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>7 Subject: Announcing ht://Dig 3.1.6 Release1 for OpenVMSe2 Message-ID: <a93kcb$l3e$1@ID-56200.news.dfncis.de>  F After the zlib buffer overflow bug also hit ht://Dig, it was a logicalD step to jump to the next stable version on that occasion. Please seeA http://www.htdig.org/RELEASE.html for extensions and fixes to the  original version 3.1.6.-  D ht://Dig for OpenVMS has grown up a bit more in that it now comes asF a PCSI kit. There still is a lot to do in the post-installation phase, but we're getting there.  6 The kit is available at (sorry for the wrapping links)  L http://www.pdv-systeme.de/users/martinv/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zi p_axpexe     andyI ftp://ftp.pdv-systeme.de/vms/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zip_axpexee   4099 blocks in size.  9 The sources are in the same directories under the name ofr* htdig-3_1_6-vms-rel1-src.zip (9635 blocks)  > And for those who attend the German DECUS symposium next week:< On Thursday, 14 o'clock, Raum Schumann, I will introduce it.   cu,    Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.dewJ One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Thu, 11 Apr 2002 15:17:22 +0200d< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>; Subject: Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMSz3 Message-ID: <a942dd$3tf5$1@ID-56200.news.dfncis.de>s  A "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> schrieb imi8 Newsbeitrag news:a93kcb$l3e$1@ID-56200.news.dfncis.de...H > After the zlib buffer overflow bug also hit ht://Dig, it was a logicalF > step to jump to the next stable version on that occasion. Please seeC > http://www.htdig.org/RELEASE.html for extensions and fixes to theh > original version 3.1.6.3 >HF > ht://Dig for OpenVMS has grown up a bit more in that it now comes asH > a PCSI kit. There still is a lot to do in the post-installation phase, > but we're getting there. >M8 > The kit is available at (sorry for the wrapping links) >r >aL http://www.pdv-systeme.de/users/martinv/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zi
 > p_axpexe	 >     andaK > ftp://ftp.pdv-systeme.de/vms/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zip_axpexe  >a > 4099 blocks in size. >p; > The sources are in the same directories under the name ofe, > htdig-3_1_6-vms-rel1-src.zip (9635 blocks) >e@ > And for those who attend the German DECUS symposium next week:> > On Thursday, 14 o'clock, Raum Schumann, I will introduce it. >t > cu, 
 >   Martin > --L > One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer9 > One OS to find them           | work: mv@pdv-systeme.de L > One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/@ > And in the Darkness bind them.| home: martin@radiogaga.harz.de >f >e   ------------------------------  % Date: Thu, 11 Apr 2002 15:24:43 +02001< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>; Subject: Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMS53 Message-ID: <a942r8$522c$1@ID-56200.news.dfncis.de>p   Once again:i  ? "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> wrote...a8 > The kit is available at (sorry for the wrapping links) >< >@L http://www.pdv-systeme.de/users/martinv/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zi
 > p_axpexe   The URL should have readL http://www.pdv-systeme.de/users/martinv/htdig/gnu-axpvms-htdig-v0301-6r1-1.p csi-zip_axpexe.                                         ^^^^^! Sorry.  	 >     andnK > ftp://ftp.pdv-systeme.de/vms/gnu-axpvms-htdig-v0301-6r1-1.pcsi-zip_axpexe   > That URL (or direct anonymous FTP access) however should work.   cu,b   Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.de/ One OS to bring them all      |e( http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------   Date: 11 Apr 2002 13:49:12 GMT< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>; Subject: Re: Announcing ht://Dig 3.1.6 Release1 for OpenVMSr3 Message-ID: <a9448o$4g39$1@ID-56200.news.dfncis.de>    ------------------------------  % Date: Thu, 11 Apr 2002 14:43:59 +0100,T From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>9 Subject: Re: another hint of post-merger direction, maybee$ Message-ID: <3CB5931F.70707@sun.com>   Rudolf Wingert wrote:    > Hello, > O > dear Andrew, do you accept the TOP500 result? If yes, why don't we see no Sun2N > under the TOP50? But we do see a few Alphas. If the Suns are so much faster,C > why don't they sold as HPS? This is reality, not a virtual image.o >        Dear Rudolf   E 1.    Yes I am happy with the top500 list, some people arn't so happy E        about the measure used (Linpack) but numbers are available forsA        most HPC platforms and so its an expedient measure to use.e  C 2.    HPC wasn't a traditional Sun market, this has changed and the H        fact that Sun is well represented in the top500 list is testiment        to this..  D My objections to the Alpha performance BS promoted on this group areC when these claims are applied to commercial systems. The HPC market5A is very small compared to the commercial server marketplace, mostsF commercial applications are not well suited to the shared nothing typeF platforms that figure at the top end of the top500 list and those thatF do (web serving for example) are I/O and integer based and not modeled by Linpack.5  H The Alpha presence in the top500 list almost exclusively based on ES40'sB this isn't suprising as it and the ES45 are the only decent Alpha  servers in production.  C Note that the Sun's in the top500 list are mostly larger systems as E are the HP's. Where are the GS boxes ? They aren't there because they.& are too big, to expensive and to slow.  D And this is a characteristic which is also evident in the commercialA server space. The GS series are to big, to expensive and to slow.r@ Anyone who thinks otherwise is deluding themselves, even Rob has@ stopped trying to defend GS performance prefering to try to hype futures.     Regards    Andrew Harrisoni   ------------------------------  # Date: Thu, 11 Apr 2002 14:50:43 GMTd% From: jlsue <jlsuexxxz@insightbb.com>h9 Subject: Re: another hint of post-merger direction, maybem8 Message-ID: <0s7bbuc31rqatolsu0kmr0upmr7sv713ii@4ax.com>  E On Thu, 11 Apr 2002 14:43:59 +0100, Andrew Harrison SUNUK Consultancy'3 <andrew_nospam.remove_this.harrison@sun.com> wrote:t   >h   >rD >Note that the Sun's in the top500 list are mostly larger systems asF >are the HP's. Where are the GS boxes ? They aren't there because they' >are too big, to expensive and to slow.- >O  C See now.  There is NO WAY you would have any information to explaincB why these systems aren't in this list.  You take one little fact -? that they aren't there - and you use it to fabricate this wholee. *opinion* that you then post as if it is fact.  F If you've got any other data to demonstrate how we can be lead to thisE conclusions besides the lack of occurrence of a data point, then feelT free to present it.   A And the whole thing about "too big, to [sic] expensive" is almostnD funny.  In the HPC space, usually speed wins the bids, with size andC expense being a secondary (or lower) concern  - just take a look atn0 the cost of these projects for evidence of that.  E >And this is a characteristic which is also evident in the commercial B >server space. The GS series are to big, to expensive and to slow.A >Anyone who thinks otherwise is deluding themselves, even Rob has A >stopped trying to defend GS performance prefering to try to hypee	 >futures.n >e  @ Well Andrew, you must believe that if you repeat something oftenF enough people will just believe you.  Once again, though, I'm going toE request to see this evidence as proof for your assertions above.  Notz  sure how you'll do that, though.  A I can tell you that I've seen hundreds of $millions of dollars inkC sales of GS-series systems over the last few months.  And get this: E I'm not in sales!!!  So surely what I do read/hear about is only parte
 of the total.   E So, I'm really curious what inforamtion you use to decide that it "is - also evident in the commercial server space."   1 Not speaking for anyone, certainly not DEC/Compaqs- (get rid of the xxxz in my address to e-mail)    ------------------------------  % Date: Thu, 11 Apr 2002 10:33:56 -040005 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 9 Subject: Re: another hint of post-merger direction, maybes, Message-ID: <a94780$de67$1@lead.zk3.dec.com>  2 Andrew Harrison SUNUK Consultancy wrote in message <3CB5931F.70707@sun.com>...e >r >e >Rudolf Wingert wrote: >e	 >> Hello,g >>L >> dear Andrew, do you accept the TOP500 result? If yes, why don't we see no SunaG >> under the TOP50? But we do see a few Alphas. If the Suns are so muchc faster,TD >> why don't they sold as HPS? This is reality, not a virtual image. >> >  >m >d >Dear Rudolf > F >1.    Yes I am happy with the top500 list, some people arn't so happyF >       about the measure used (Linpack) but numbers are available forB >       most HPC platforms and so its an expedient measure to use. > D >2.    HPC wasn't a traditional Sun market, this has changed and theI >       fact that Sun is well represented in the top500 list is testiment3 >       to this. >-E >My objections to the Alpha performance BS promoted on this group arelD >when these claims are applied to commercial systems. The HPC marketB >is very small compared to the commercial server marketplace, mostG >commercial applications are not well suited to the shared nothing typenG >platforms that figure at the top end of the top500 list and those thatsG >do (web serving for example) are I/O and integer based and not modeledr >by Linpack. >dI >The Alpha presence in the top500 list almost exclusively based on ES40'saB >this isn't suprising as it and the ES45 are the only decent Alpha >servers in production., >lD >Note that the Sun's in the top500 list are mostly larger systems asF >are the HP's. Where are the GS boxes ? They aren't there because they' >are too big, to expensive and to slow.s >gE >And this is a characteristic which is also evident in the commercialiB >server space. The GS series are to big, to expensive and to slow.A >Anyone who thinks otherwise is deluding themselves, even Rob has-A >stopped trying to defend GS performance prefering to try to hypee	 >futures.e >.  L Andy, it's actually hard to say.  Maybe nobody has submitted results.  MaybeI nobody really pays attention much to it.  Maybe only your big-ass systemsdI are on it, because nothing else you sell is worth spit.  The GS series isoI optimized towards large aggregate I/O throughput, coupled with reasonably2J high-performance CPUs - optimized for commercial application.  The smallerG systems are are optimized somewhat differently, and can drive their I/OpJ busses at saturation, but don't have as many to drive.  Since they also doL not have to rely on complex NUMA hierarchies to tie together 2-4 CPU's, theyK can be just that much faster.  They are very popular with the set who likeseI to tie them together with quadrics switches and build supercomputer-class ? systems -- the types who are interested in things like LINPACK.l  K Of course, all this will be academic when Marvel ships.  At which point allyL you will be able to do is drool like an idiot.  And while that is a "future"J the future isn't very far away.  So try not to say something that you will" only regret a few months from now.  J PS - I am glad you are happy with your representation on the bottom of theL list.  Let us know when you are happy with being represented on the top.  It4 isn't likely to EVER happen with Sparc and Slowaris.   ------------------------------  % Date: Thu, 11 Apr 2002 12:40:17 -0400s; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>f2 Subject: Re: Best DCL programm you've ever written$ Message-ID: <3cb5bce2$1@news.si.com>  A >When was it, what does it do for you and whe can it be obtained?   " 23-May-1999.  Here's what it does:   $!  QUEUE_CONTROL J $!  This procedure starts or stops all print queues, or the specified type of
 $!  queue. $!3 $!  P1 = "HELP"         prints a short HELP messagey0 $!  P1 = "START"        starts all chosen queues/ $!  P1 = "STOP"         stops all chosen queues  $!* $!  P2 = "" or "ALL"    selects all queuesI $!  P2 = "GENERIC"      selects generic print queues or print queues withu( $!                      processor PRTSMBB $!  P2 = "LAT"          selects print queues with processor LATSYML $!  P2 = "LPR"          selects print queues with processor LPRSMB or LPDSMBB $!  P2 = "LPD"          selects print queues with processor LPDSMBC $!  P2 = "SCRIPTSERVER" selects print queues with processor SSV$SMBdC $!  P2 = "TALARIS"      selects print queues with processor TLAPSMBsC $!  P2 = "TCP"          selects print queues with processor TCPSYMB  $!E $!  P3 = "" or "LOG"    displays the START or STOP command to be usedAL $!  P3 = "NOLOG"        supresses displaying the START or STOP command to be $!                      used  $ It can be obtained by writing to me. --A Brian Tillman                   Internet: tillman_brian at si.com.A Smiths Aerospace                          tillman at swdev.si.comu= 3290 Patterson Ave. SE, MS      Addresses modified to prevent?< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------    Date: 11 Apr 2002 07:40:46 -0500- From: koehler@encompasserve.org (Bob Koehler)a Subject: Re: C issues...3 Message-ID: <mIoVt6SM0e67@eisner.encompasserve.org>a  a In article <a91g17$mfr$1@newslocal.mitre.org>, sssslewis@mazda.mitre.org (Keith A. Lewis) writes:BM > The default mode for logical names created with the DCL "define" command isrJ > supervisor, but the default mode for $CRELNM is user.  User-mode logicalL > names in the process table are deleted on image rundown (when your programD > ends), which is why you can't see it afterwards at the DCL prompt. > + > Try using LNM$JOB instead of LNM$PROCESS.0  K    Try using LIB$SET_LOGICAL instead of $CRELNM.  It can, and defaults to, m    supervisor mode.-   ------------------------------  + Date: Thu, 11 Apr 2002 14:50:28 +0100 (MET)s9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>o* Subject: Re: can VMS use these SCSI disks?; Message-ID: <01KGFVVBXCW29D50GB@sysdev.deutsche-boerse.com>i  E > Be aware that for any SCSI drives, depending on the requirements of	G > your version of VMS, you may need to reprogram some SCSI MODE pages. 2H > Usually this affects the error correction page.  If these bits are setD > wrong, the usual symptom is a "fatal drive error" when MOUNTing or > INITing the drive.  1C > I believe the ST410800N drives are the last of the CDC Wren line.n  > Is Reprogramming necessary for these drives with 7.2x or 7.3x?  B > They are built like tanks and will run forever.  If you have the4 > space and power for them, I highly recommend them.  A I collected them this morning (thus the questions here).  Pretty sA impressive!  I have an old SCSI expansion box with a huge fan....a  D On a related note:  Can one simply put SCSI disks in some enclosure F (other than expansion box, i.e. VMS machine, InfoServer,...), connect C them to the VMS machine which is to use them via the external SCSI  A interface, power it up and then use them?  What about even if theeF enclosure is also doing other things on the SCSI bus (but not with theA disk in question and, of course, assuming no SCSI-ID conflicts). o   ------------------------------  % Date: Thu, 11 Apr 2002 08:40:33 +0200 ( From: "arachnis" <arachnis_@hotmail.com>3 Subject: Re: Can't change resolution on ELSA Gloriaw$ Message-ID: <3cb52e55@news.euriware>  J Peter, you should have a look at the DECW$DEVICE_CONFIG_XX.com file (under
 SYS$MANAGER)..  D Replace XX by the "device" type of your Elsa Gloria. You can find it executing the SH DEV G command.r  - Perhaps here is a solution to your problem...n   regards,   Xav1        < "pbc" <pbc@informchaos.com> a crit dans le message de news:# a92amd$1fs8$1@news.radio-msu.net...l > Hi >sG > I can't change default resolution on my ELSA Gloria on XP1000 running 	 > VMS7.3.aH > I did try DECW$PRIVATE_SERVER_SETUP.COM with DECW$XSIZE_IN_PIXELS etc. with
 > no luck. >v > Thanks >- > PeterD >f >O >V   ------------------------------  % Date: Thu, 11 Apr 2002 12:21:46 -0400 ; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>t! Subject: Re: CLD problem (flight)a$ Message-ID: <3cb5b88a$1@news.si.com>  K >Thanks. Now, does anyone know to whom we should send this "bug" report forx	 >FLIGHT ?n  B I just fetched FLIGHT V3.1 from Compaq's freeware40 site and these+ corrections were already a part of the kit.e --A Brian Tillman                   Internet: tillman_brian at si.comsA Smiths Aerospace                          tillman at swdev.si.comy= 3290 Patterson Ave. SE, MS      Addresses modified to preventi< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------    Date: 11 Apr 2002 08:18:45 -0500- From: koehler@encompasserve.org (Bob Koehler)r5 Subject: determine display server number for SSH X11?n3 Message-ID: <4tdnd4H20MDo@eisner.encompasserve.org>   @    How can I determine the X11 server number for a SSH forwarded    connection?  D    Here's the setup.  I've got Exceed running on W2K, with puTTY forG    it's SSH client.  I SSH login using puTYY with X11 forwarding turned E    on into a Multinet 4.3A SSH service (VMS systems that I don't have     elevated privileges on).w  F    SSH sets up an X11 server on the Multinet end which tunnels back toA    the Exceed X11 server on the PC.  The correct host name for a cJ    "set display" command is always the VMS machine, and the server number I    is usually 1, sometimes 2 (never 0 since that's the VMS workstation's o    local server).   F    The VMS system is running something that attempts to figure out andE    setup the display for me, I don't know what.  It works for non-SSHlG    connections directly from an X11 server to the VMS system.  But on a/G    SSH connection it always comes up with the PC address with a number aJ    appended (resembles UNIX DISPLAY environment variable style), server 0.  D    If I run decw$clock without correcting the display setting, it of    course fails.  G    If I run sys$system:decw$session, it somehow figures out the correcthH    server and brings up the DECwindows (CDE) log in screen on my PC.  IfD    I log in this way and do a "show display" I get the correct info.<    So somehow decw$session can figure out the server number.  I    Currently, we're setting the display to the correct node and guessing nJ    at the server number, then using trial and error.  Not exactly secure, H    we could pop up a window on someone else's SSH tunneled X11 server.  ?    Usually I test using decw$clock as that's not too dangerous.   G    If decw$session can get it right, where does it get it from so I canc    get it right?   ------------------------------  % Date: Thu, 11 Apr 2002 09:28:40 -0400e5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>m' Subject: Re: Evolution of the DEC mousea, Message-ID: <a943dj$av9e$1@lead.zk3.dec.com>   Tom Linden wrote in message ...v7 >the older mice had a circular 7 pin connector; whereass5 >the newer ones have a 6 pin pc compatible connector.w >v: >Does anyone know if it is possible to associate the wires7 >from one with the other?  The objective is to put some 8 >VAXs, DEC 3000/300 and DECstation 5000 on a KVM switch. >s6 >The mice are all made by Logitech, the older ones are6 >Models VSXXX-GA and use only 6 wires.  The newer ones6 >ar M-S38 and use only 5 wires.  What I want to end up6 >with is a cable with the older 7 pin connector at one3 >end and the newer 6 pin at the other, if possible.t >r* >These are both three button mice; however >/  J In the beginning, mice were simple quadrature mechanisms, and the hardware counted pulses and phase shift.-  L The next mouse from DEC was a serial mouse.  It has power and tx/rx lines inE it, and runs at 4800 baud.  That mini-din connector goes right into ap standard serial port.4  K The PS/2 mouse on the other hand is a greasy little serial-like device, but I it can't just be plugged into any-old serial port, it is plugged into thea4 junk I/O for a PC, which has a KB/Mouse countroller.  L The answer is, even if you could get them to be electrical compatable, their; data is completely different.  You can't mix them on a KVM.e  J You should be able to search google for the pin-out for the old DEC serialH mouse.  I no longer have the hardcopy documentation.  The ones that haveD been used on all Alpha's since the DEC 3000 (Flamingo/Sandpiper) are standard PS/2 compatable mice.   ------------------------------  % Date: Thu, 11 Apr 2002 08:11:10 -0700t# From: "Tom Linden" <tom@kednos.com>r' Subject: RE: Evolution of the DEC mouset9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEPBELAA.tom@kednos.com>   < OK, that rules out one possibility.  Next option is to use a? 4 port KVM switch for only DEC style mice, in which I just haveeB figure out how to associate a wire with a pin, makeup some adapter cables, and voila.  > Are LK 201 and 401  keyboards compatible?  Not sure what to do about a LK-501.    > -----Original Message-----< > From: Fred Kleinsorge [mailto:kleinsorge@star.zko.dec.com]( > Sent: Thursday, April 11, 2002 6:29 AM > To: Info-VAX@Mvb.Saic.Come) > Subject: Re: Evolution of the DEC mousee >e >n >o! > Tom Linden wrote in message ...d9 > >the older mice had a circular 7 pin connector; whereas 7 > >the newer ones have a 6 pin pc compatible connector.t > >i< > >Does anyone know if it is possible to associate the wires9 > >from one with the other?  The objective is to put somem: > >VAXs, DEC 3000/300 and DECstation 5000 on a KVM switch. > >T8 > >The mice are all made by Logitech, the older ones are8 > >Models VSXXX-GA and use only 6 wires.  The newer ones8 > >ar M-S38 and use only 5 wires.  What I want to end up8 > >with is a cable with the older 7 pin connector at one5 > >end and the newer 6 pin at the other, if possible.r > >t, > >These are both three button mice; however > >u >eL > In the beginning, mice were simple quadrature mechanisms, and the hardware! > counted pulses and phase shift.r > ? > The next mouse from DEC was a serial mouse.  It has power ands > tx/rx lines inG > it, and runs at 4800 baud.  That mini-din connector goes right into aT > standard serial port.+ >0A > The PS/2 mouse on the other hand is a greasy little serial-liker
 > device, butwK > it can't just be plugged into any-old serial port, it is plugged into the 6 > junk I/O for a PC, which has a KB/Mouse countroller. >e< > The answer is, even if you could get them to be electrical > compatable, theirh= > data is completely different.  You can't mix them on a KVM.h > L > You should be able to search google for the pin-out for the old DEC serialJ > mouse.  I no longer have the hardcopy documentation.  The ones that haveF > been used on all Alpha's since the DEC 3000 (Flamingo/Sandpiper) are  > standard PS/2 compatable mice. >e >h >a >o >c   ------------------------------  % Date: Thu, 11 Apr 2002 11:35:31 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>u' Subject: Re: Evolution of the DEC mouses, Message-ID: <a94are$ddfu$1@lead.zk3.dec.com>  J Do a google search.  I know that I put the pin out into a reply years ago.G It also was in the back of the VCB02 manual, which I gave away to SRI an while back.e     Tom Linden wrote in message ...a= >OK, that rules out one possibility.  Next option is to use ai@ >4 port KVM switch for only DEC style mice, in which I just haveC >figure out how to associate a wire with a pin, makeup some adapterd >cables, and voila.  >t? >Are LK 201 and 401  keyboards compatible?  Not sure what to doo >about a LK-501. >h >> -----Original Message-----t= >> From: Fred Kleinsorge [mailto:kleinsorge@star.zko.dec.com]w) >> Sent: Thursday, April 11, 2002 6:29 AMt >> To: Info-VAX@Mvb.Saic.Com* >> Subject: Re: Evolution of the DEC mouse >> >> >>" >> Tom Linden wrote in message ...: >> >the older mice had a circular 7 pin connector; whereas8 >> >the newer ones have a 6 pin pc compatible connector. >> >= >> >Does anyone know if it is possible to associate the wiresr: >> >from one with the other?  The objective is to put some; >> >VAXs, DEC 3000/300 and DECstation 5000 on a KVM switch.  >> >9 >> >The mice are all made by Logitech, the older ones are39 >> >Models VSXXX-GA and use only 6 wires.  The newer onesc9 >> >ar M-S38 and use only 5 wires.  What I want to end up 9 >> >with is a cable with the older 7 pin connector at oner6 >> >end and the newer 6 pin at the other, if possible. >> >- >> >These are both three button mice; however  >> > >>D >> In the beginning, mice were simple quadrature mechanisms, and the hardware" >> counted pulses and phase shift. >>@ >> The next mouse from DEC was a serial mouse.  It has power and >> tx/rx lines inPH >> it, and runs at 4800 baud.  That mini-din connector goes right into a >> standard serial port. >>B >> The PS/2 mouse on the other hand is a greasy little serial-like >> device, butL >> it can't just be plugged into any-old serial port, it is plugged into the7 >> junk I/O for a PC, which has a KB/Mouse countroller.  >>= >> The answer is, even if you could get them to be electrical  >> compatable, their> >> data is completely different.  You can't mix them on a KVM. >>F >> You should be able to search google for the pin-out for the old DEC serialK >> mouse.  I no longer have the hardcopy documentation.  The ones that havecG >> been used on all Alpha's since the DEC 3000 (Flamingo/Sandpiper) ared! >> standard PS/2 compatable mice.f >> >> >> >> >> >    ------------------------------  % Date: Thu, 11 Apr 2002 12:31:20 +0100r* From: "Rainer Giese" <giese@volkswerft.de>  Subject: EWA-bootdriver messages4 Message-ID: <a93oln$2144$1@ID-138444.news.dfncis.de>  J After changing a DPW500au to boot from a 7.3-server instead of 7.2-1 I gotL dozens repeating EWA-I-BOOTDRIVER messages telling me, what ethernet adapterK found, what the console settings are and what will be done. One time at theaH beginning at 7.2-1 was enough for me, now some important messages vanish! between this "ten screens of ..."n  * Anyone knows how to make it less verbose ?   Regards, Rainer Giese   ------------------------------  % Date: Thu, 11 Apr 2002 07:57:35 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)$ Subject: Re: EWA-bootdriver messagesK Message-ID: <rdeininger-1104020757350001@1cust183.tnt3.nashua.nh.da.uu.net>e  C In article <a93oln$2144$1@ID-138444.news.dfncis.de>, "Rainer Giese"t <giese@volkswerft.de> wrote:  K >After changing a DPW500au to boot from a 7.3-server instead of 7.2-1 I gotrM >dozens repeating EWA-I-BOOTDRIVER messages telling me, what ethernet adaptereL >found, what the console settings are and what will be done. One time at theI >beginning at 7.2-1 was enough for me, now some important messages vanishe" >between this "ten screens of ..." > + >Anyone knows how to make it less verbose ?t  - Strange behaviour.  I haven't seen it before.m  $ Is the system's firmware up to date?   ------------------------------  + Date: Thu, 11 Apr 2002 15:21:47 +0100 (MET)m9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>i) Subject: Re: Fearless VMS Prognosticationi; Message-ID: <01KGFX83ZHTE9D50GB@sysdev.deutsche-boerse.com>i  - > The number itself is virtually meaningless.l > . > ...and by the way, 17% does not impress me.   3 95% of all statistics are made up on the spot.  :-|e   ------------------------------    Date: 11 Apr 2002 09:01:07 -0500+ From: young_r@encompasserve.org (Rob Young)u) Subject: Re: Fearless VMS Prognosticatione3 Message-ID: <MmDmfp6YSfb+@eisner.encompasserve.org>h  \ In article <3CB4F282.A52C339F@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Rob Young wrote:M >>         There are a good many assumptions in what you write.  First ... ifvI >>         VMS is up 17% as relayed form our European friend... that is ad >>         sizable increase. > M > You can still claim 17% growth, but without the context over what period ithM > happened, that 17% figure doesn't mean that much. Also, one doesn't know ifuJ > this is a one time blip (for instance, replacements for WTC machines) or# > something that will be sustained.    	True.  P > Compaq has destroyed the trust customers had. It will take a serious effort toA > rebuild such a trust in the owner's commitment to that product.u  = 	You are wrong.  That is nothing but idle FUD.  Why?  BecauseA< 	Compaq is actually selling product and meeting or exceeding 	first quarter expectations:  > http://biz.yahoo.com/rb/020409/markets_stocks_us_europe_3.html  N "Underpinning confidence was Compaq, after it said it expected to meet or beatN first-quarter earnings forecasts, which came as a relief after other forecasts/ have indicated earnings remain under pressure."e  # 	So spin that one a bit.  Have fun.    				Rob3   ------------------------------  % Date: Thu, 11 Apr 2002 14:46:09 +0100"% From: Alan Greig <a.greig@virgin.net>S) Subject: Re: Fearless VMS Prognostication 8 Message-ID: <rk4bbugdt5nf90aqajndia4cv8u44re9qr@4ax.com>  4 On Wed, 10 Apr 2002 05:45:48 GMT, "Terry C. Shannon" <terryshannon@attbi.com> wrote:   H >Damn the doomsayers. Stay tuned for a letter from Carly F re: continued >support for OpenVMS.   C Doesn't the fact that Carly had  previously not mentioned VMS since @ the merger announcement - but did mention NSK - perhaps point toB something having woken her up since just a bit? I wonder what that could have been?  D As Peter Kastner of the Aberdeen Group said (quoting from memory) "IF fully expect HP to treat VMS no differently to MPE. If you want to see' VMS survive let the HP board know now."    ------------------------------    Date: 11 Apr 2002 09:45:09 -0500+ From: young_r@encompasserve.org (Rob Young)u) Subject: Re: Fearless VMS Prognosticatione3 Message-ID: <CEqG$RvVxP5q@eisner.encompasserve.org>t  ` In article <rk4bbugdt5nf90aqajndia4cv8u44re9qr@4ax.com>, Alan Greig <a.greig@virgin.net> writes:6 > On Wed, 10 Apr 2002 05:45:48 GMT, "Terry C. Shannon"! > <terryshannon@attbi.com> wrote:i > I >>Damn the doomsayers. Stay tuned for a letter from Carly F re: continued- >>support for OpenVMS. > E > Doesn't the fact that Carly had  previously not mentioned VMS sinceHB > the merger announcement - but did mention NSK - perhaps point toD > something having woken her up since just a bit? I wonder what that > could have been? >   	 	Numbers.,  F > As Peter Kastner of the Aberdeen Group said (quoting from memory) "I= > fully expect HP to treat VMS no differently to [than] MPE."e  ? 	He is full of beans.  He doesn't accept numbers.  No more thanmC 	the fellow at Gartner did.  At least the Gartner fellow circed his C 	poison to VMS folks prior to publication and when shown numbers hee@ 	chose to ignore them or minimize them.  That was what I had the= 	most problem with when the Gartner fellow wrote his piece.     A 	Hard to bend your "truth" when annoying numbers get in your way!o   				Robm   ------------------------------    Date: 11 Apr 2002 09:59:36 -0500+ From: young_r@encompasserve.org (Rob Young).) Subject: Re: Fearless VMS Prognosticationt3 Message-ID: <siwHrbR2AR9j@eisner.encompasserve.org>   p In article <srht8.314467$2q2.28654790@bin4.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes: > : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:CEqG$RvVxP5q@eisner.encompasserve.org...h >  > ...( > C >> Hard to bend your "truth" when annoying numbers get in your way!n > M > That's pretty funny, coming from someone who never had a problem dismissing0 > a number he didn't like. >    	And those would be?   				Roba   ------------------------------  # Date: Thu, 11 Apr 2002 14:53:12 GMTe* From: "Bill Todd" <billtodd@metrocast.net>) Subject: Re: Fearless VMS Prognostication.C Message-ID: <srht8.314467$2q2.28654790@bin4.nnrp.aus1.giganews.com>   8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:CEqG$RvVxP5q@eisner.encompasserve.org...    ...   B > Hard to bend your "truth" when annoying numbers get in your way!  K That's pretty funny, coming from someone who never had a problem dismissingm a number he didn't like.   - bill   ------------------------------  # Date: Thu, 11 Apr 2002 16:08:28 GMTX% From: jlsue <jlsuexxxz@insightbb.com>s. Subject: Re: file structure allocation failure8 Message-ID: <80dbbugc6qim6gtlo5g4r610f7785pblnj@4ax.com>  E One other thing to note about backup/image to a disk drive is that itCA will try to init the output drive with the same parameters as thek  input drive (especialy on V5.5).  F The problem that can occur is where the output disk is larger than theD input disk - which means a larger "minimum" cluster size.  You can'tD INITIALIZE the larger disk with the smaller cluster size that was in. effect on the smaller disk during the backups.  E One way to get around this is to INITIALIZE the new disk with all the B appropriate qualifiers and values for your circumstances (and just9 ignore using /CLUSTER, INIT will figure out the minimum).r  ! Then, in the backup command, use -  ( 	$ BACKUP/IMAGE <input> <output>:/NOINIT  F That will avoid this problem on V5.5 (as I learned back when I was the system manager).    C On Wed, 10 Apr 2002 11:17:40 -0700, luongm <minh.luong@navigata.ca>o wrote:  A >Just  wondering if anyone would have an answer to this scenario.c >eG >I have an old VAX4000-60 system running VMS V5.5, which the hard driveYB >crashed recently.  The original hard drive was a Seagate ST31200NF >(1.25G), the replacement one is a Seagate ST32430N (2.14G.)  For someF >reason, I keep having the "File structure allocation failure" messageB >when attempted to perform the image restoration from backup tape.# >Bellow is the exact error message.r >cF >"% BACKUP-F-ALLOCFAIL,  File structure allocation failure for DKA700:, >-   SYSTEM-F-DRVERR,     Fatal drive error" > G >I have tried the low level format (or believed to be low level format)sI >utility from the bootstrap prompt ">>>" the syntax is as described here:r >,8 >test /util 10 (then supplied the scsi ID when prompted) >rG >All test came back successfull.  Yet I still cannot "mount/foreign" or . >"initialize" or "backup/image/verify" at all. >rF >Is it possible for a bad scsi drive to pass all the test? From what ID >know the size is not an issue as this VMS version should be able to >handle more than 4GB. > ( >All your feedbacks are appreciated. :-) >e  1 Not speaking for anyone, certainly not DEC/Compaq - (get rid of the xxxz in my address to e-mail)f   ------------------------------  + Date: Thu, 11 Apr 2002 14:53:28 +0100 (MET)e9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>r: Subject: free speech (was: RE: More positive OpenVMS news); Message-ID: <01KGFW8I2M8I9D50GB@sysdev.deutsche-boerse.com>.  > > > That must be the most stupid thing I'v read anyware, ever.8 > > Please, please, please, keep that out of this list ! >c9 > Ah, a wee bit of a problem with freedom of speech, eh? n  I That's not the point.  The point is that this is a high-volume group and - we should keep things on-topic.    ------------------------------  % Date: Thu, 11 Apr 2002 13:46:39 +0100zT From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>& Subject: Re: Gartners paid off by IBM!& Message-ID: <3CB585AF.3020107@sun.com>   Bob Ceculski wrote:r  A > told you to ignore Gartner ... they are paid off by IBM/Sun ...e > from the inquirer ...s >     > Where in the article does it say that Sun or IBM have paid off Gartner ????  = All it says is that Gartner expect Sun and IBM to take marketT< share of HP and Compaq, this is hardly earthshaking analysis its shared by most analysts.  7 In addition it is also shared by HP and Compaq who havea3 publically stated that they expect a their combinedo4 revenues to decline by I think it was 5%. Well guess3 what that means that their competitors market share; will increase.  5 You seem to think that any idea which challenges your 9 own rather limitted view of the world must have orignatede+ from a "Black Hellicopter" type conspiracy.   6 If you had a bit more analytic nouse you would realise5 that the only conspiracy is the one being perpetrated " on you by your favourite supplier.  5 I like your posting style by the way it reminds me ofn  	 Mr Brass.u   Regardsn Andrew Harrison,    0 > Gartner out of line on Compaq takeover future  > - > VMS, other platforms safe whatever happens y( > By Mike Magee, 10/04/2002 16:35:53 BST > H > THERE'S NOTHING like a high-profile, contentious IT industry merger toD > serve as a catalyst for "gloom and doom" predictions. Here's SKC's9 > take on an analyst assessment issued on March 22, 2002.hA > Printed with permission from Shannon knows Compaq, (c) Terry C.i > Shannon 2002.e > D > The day after Compaq shareholders voted overwhelmingly in favor ofF > Hewlett-Packard's proposed acquisition of Compaq, Stamford, CT-basedH > Gartner Group weighed in with a polemic fraught with dire implicationsF > for the largest and most contentious merger in the history of the IT > industry.o > G > On Friday March 22, Paul McGuckin, Gartner Unix & Midrange StrategiesuE > analyst published a research note entitled "Ten Key Consequences of  > the HP/Compaq Deal." > D > "Whether shareholders approve or veto the proposed merger, GartnerD > believes that it will have profound consequences for HP and Compaq9 > customers and for the industry," Mr. McGurkin asserted.  > D > "Customers can expect that Tru64 UNIX, OpenVMS, Netaction and manyH > storage platforms are likely at risk," the analyst opined, adding that@ > "Whether merged or separate, HP and Compaq will likely use theG > decision to justify radical housecleaning of products." HP and Compaq ; > customers have a unique opportunity to migrate off of thec# > aforementioned "risky" platforms.0 > F > Indeed, the analyst said, "For many customers, the risk will finallyH > outweigh the technological inertia and loyalty to vendors and products> > that have kept them tied to marginal products." What's more,F > enterprises will have a chance to lock in steep, multiyear discountsG > as HP and Compaq-- whether merged or separate--try to win business att< > all costs to demonstrate their market power and stability. > C > Despite Peter Mercury's recent assertions to the contrary, "Fielda? > service and sales will likely undergo substantial turmoil, aseG > merger-driven reassignments or cost-cutting layoffs cause some of thetA > best employees to leave and some geographic territories to have0# > rougher transitions than others."i > A > Gartner went on to claim that the implications for the computertF > industry as a whole include likely gains in market share on the partG > of Dell Computer, IBM and Sun Microsystems, all of whom are likely tohF > exploit disruption and uncertainty at Compaq and HP through at least@ > 3Q02 and perhaps until 1Q03, largely because of the short-termH > instability associated with merging or implementing alternative plans. > F > The IT industry as a whole will experience a new round of price cutsE > as vendors, led by HP and Compaq, try to hold onto market share. AtrD > the same time, Linux and IBM AIX will increase their support amongC > ISVs, likely at the expense of HP/UX and Compaq Tru64 UNIX. In aniG > effort to capitalize on proven Compaq amnd HP strengths, Dell and SunrF > will likely partner with, or acquire, service firms to prepare for a% > greater focus on selling solutions.g > : > And a bit lower on the food chain, Gartner believes thatF > transformation in the PC, server, and output markets will accelerateC > as all vendors scrutinize their manufacturing processes, businessoH > models and research investments. More plausible is the prediction thatG > customer skepticism of Compaq and HP promises will likely increase as,E > the vendors reveal product road maps and modify premerger promises.  > H > Rejection of the merger would create a different kind of skepticism asG > the separate companies try to brighten the dark pictures they paintedoF > of their futures as separate entities and to convince customers thatB > they have viable alternative "go it alone" strategies. (Based onB > discussions with senior Compaq executives, SKC believes that Mr.G > Capellas and his lieutenants have in fact crafted a workable "Plan B"-= > scenario in the unlikely event that such a scenario must belE > implemented. Compaq's 1FQ02 financial results, due out on April 18,f- > should help validate the Compaq game plan).7 >  > Trust But VerifyE > Taken at face value, these predictions do not bode well for The NewlG > HP. While it is true that near-tern disruption and uncertainty shouldvG > help rival vendors persuade incumbent Compaq and HP customers to seekdC > a more stable IT partner, and while it is evident that Tru64 UNIXlA > ultimately will be relegated to HP/UX "organ donor" status, SKC F > believes that Compaq and HP are taking steps to minimize distruptionC > and maximize continuity in the firms' respective Customer Service 
 > franchises.t > E > Tarring OpenVMS and the merged Compaq-HP storage portfolio with theSC > "risky scheme" brush seems a bit short-sighted: Compaq's midrangenF > StrorageWorks product set--which coincidentally occupies an enviableI > position in Gartner's "Magic Quadrant"--should profit from the addition % > of HP's high-end storage portfolio.u > G > As for the OpenVMS operating system, the oft-maligned but financiallyvG > successful bet-your-business enterprise OS generates an extimated $2BiF > USD or more of high-margin revenue each year, the VMS-on-IPF portingG > effort is proceeding at a brisk clip, and the OS's unique status as ae? > certified Defense Information Infrastructure-Common OperatingrD > Evironment (DII-COE)-compliant product serves as a very compelling > life insurance policy. > G > Then too, any decision to decommit from OpenVMS would cost The New HPlG > a hefty revenue stream while ensuring a rapid and wholesale migrationaG > from HP to an alternative vendor such as IBM. Unless Doctor Kevorkian F > is on call in the Compaq-HP "Clean Room," it is highly unlikely that/ > the OS will be targeted for early retirement.s > A > Although neither Compaq nor HP have stepped up to the plate and.H > affirmed a long-term commitment to the operating environment, this tooE > shall change Real Soon Now: relaible word has it that Carly FiorinaeG > and her Compaq counterparts are drafting a comprehensive statement ofoE > support for OpenVMS. SKC advises OpenVMS customers to seek a secondrH > opinion, and to adopt a "trust but verify" stance regarding the claims2 > of industry analysts, high-profile or otherwise. >    ------------------------------    Date: 11 Apr 2002 08:15:32 -0500- From: koehler@encompasserve.org (Bob Koehler)-& Subject: Re: Gartners paid off by IBM!3 Message-ID: <SyUoHjYaL5lD@eisner.encompasserve.org>8  \ In article <3CB4EB0D.361478AA@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > N > Gartner need not be paid by IBM. Their biggest customers are large IBM shopsH > who want Gartner to write stuff to support their decisions to buy IBM.  @    IIRC they were founded by a bunch of former IBM employees and0    expanded by hiring disgruntled DEC employees.  '    Real unbiased opinions there, right?z   ------------------------------  % Date: Thu, 11 Apr 2002 05:53:22 -0500o From: jnuis@excite.fr:! Subject: GOV GRANTS, NEVER REPAY.-; Message-ID: <00004bd66741$00006c83$00003929@ww8.tiki.ne.jp>o   <html><body> <hr width =3D "100%">aK <center><font size =3D "+2" color =3D "#44C300"><b>Government Grants E-Boo=e k 2002 edition</font></b><p> <table><Tr><td> K <li>You Can Receive The <font color =3D "green"><b>Money</b></font> You Ne=h ed...HK <li>Every day <b><font color =3D "green">millions of dollars</font></b> ar=t' e given away to people, just like you!! K <li>Your Government spends <b><font color =3D "green">billions</font></b> =e$ of tax dollars on government grants.D <li>Do you know that private foundations, trust and corporations areH <li>required to give away a portion of theirs assets. It doesn't matter,K <li>where you live (USA ONLY), your employment status, or if you are broke=c	 , retired > <li>or living on a fixed income. There may be a grant for you! <hr width =3D "100%">oK <li><font color =3D "red"><b>ANYONE</b></font> can apply for a Grant from =  18 years old and up!K <li>We will show you HOW & WHERE to get Grants. <font color =3D "red"><b>T= I HIS BOOK IS NEWLY UPDATED WITH THE MOST CURRENT INFORMATION!!!</b></font>f3 <li>Grants from $500.00 to $50,000.00 are possible!2, <li>GRANTS don't have to be paid back, EVER!K <li>Grants can be ideal for people who are or were bankrupt or just have b=g
 ad credit. </td></tr></table>: <br><font size =3D "+1">Please Visit Our Website<p></font>M And Place Your <font color =3D "red"> <b>Order TODAY!</b> </font><a href =3D= K  "http://www.software4you.net/grant/grant/"><b><font size=3D"5">CLICK HERE=i </font></b> </a><p>&nbsp;<p> <font size=3D"1">iC We apologize for any email you may have inadvertently received.<br>oK Please <a href =3D "http://www.software4you.net/remove.htm">CLICK HERE</a>=m/  to be removed from future mailings.</font><br>    </BODY>  </HTML>    ------------------------------  # Date: Thu, 11 Apr 2002 12:36:58 GMT 2 From: Mike Rechtman <michael.rechtman@digital.com>3 Subject: Re: Help! VT100 emulators - some problems.r+ Message-ID: <3CB5ABEE.55DF2D61@digital.com>e   Been there, done that -oF Hebrew has two character types - print and cursive. The standard fontsB on a Hebrew-enabled VT220 were print only (yes Virginia, there are escape S) codes to make a VT220 work right-to-left)rG Back in the pre-historic mists on RSTS/E I wrote a DCL command file to sD load a cursive font in DRCS. Each character had to be bit-mapped and encoded to nD a list of integers... Luvly fun! - But it still works (tried it) on  VMS.   Mike     Thomas Dickey wrote: > / > Paul Williams <flo@uk.thalesgroup.com> wrote:d > > Thomas Dickey wrote: > >>? > >> Brian Tillman <tillman_brian@notnoone.notnohow.com> wrote:  > >> >G > >> > There was a command procedure on a DECUS tape long ago that usedrF > >> > SIXEL programming to change all the characters on the screen ofD > >> > a VT to be backward (or upside down, I can't recall which any
 > >> > more).o > >>H > >> that's probably using DRCS, but it doesn't affect the default font. >  > > Maybe :) > L > > The VT320 (don't know about the VT220) allows you to download a font and > G > now that you mention it, the VT220 docs don't mention an exclusion...e > L > > give it the same name as a built-in character set. And from then on, theK > > SCS sequences will refer to your font. Over the course of a few days, IlK > > used to find the font getting corrupted, but I never determined whether C > > this would happen with normal soft fonts, or just because I wasn* > > perverting the font-mapping mechanism. > G > > And no, I'm not going to give you a patch to emulate this in xterm!x > H > it's been on my list for a while, but low-priority (the only way I canB > see is to make a bitmap for each glyph, which sounds expensive). >  > --? > Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>  > http://dickey.his.com  > ftp://dickey.his.com   -- .E --------------------------------------------------------------------- E Usual disclaimer: All opinions are mine alone, perhaps not even that.d? Mike Rechtman                            *rechtman@tzora.co.il*aF Kibbutz Tzor'a.                          Voice (home): 972-2-9908337  B   "20% of a job takes 80% of the time, the rest takes another 80%"E ---------------------------------------------------------------------n -----BEGIN GEEK CODE BLOCK-----  Version: 3.1: GCM/CS d(-)pu s:+>:- a++ C++ U-- L-- W++ N++ K? w--- V+++$6 PS+ PE-- t 5? X- tv-- b+ DI+ D-- G e++ h--- r+++ y+++@ ------END GEEK CODE BLOCK------p   ------------------------------  % Date: Thu, 11 Apr 2002 09:43:16 -0400y5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>03 Subject: Re: Help! VT100 emulators - some problems. , Message-ID: <a9448u$ce1k$1@lead.zk3.dec.com>  " Thomas Dickey wrote in message .... >Paul Williams <flo@uk.thalesgroup.com> wrote: >> Thomas Dickey wrote:D >>>u> >>> Brian Tillman <tillman_brian@notnoone.notnohow.com> wrote: >>> >hF >>> > There was a command procedure on a DECUS tape long ago that usedE >>> > SIXEL programming to change all the characters on the screen oftC >>> > a VT to be backward (or upside down, I can't recall which anys >>> > more). >>>AG >>> that's probably using DRCS, but it doesn't affect the default font.e >e >> Maybe :)L >tK >> The VT320 (don't know about the VT220) allows you to download a font andr >eF >now that you mention it, the VT220 docs don't mention an exclusion... > K >> give it the same name as a built-in character set. And from then on, thetJ >> SCS sequences will refer to your font. Over the course of a few days, IJ >> used to find the font getting corrupted, but I never determined whetherB >> this would happen with normal soft fonts, or just because I was) >> perverting the font-mapping mechanism.s >wF >> And no, I'm not going to give you a patch to emulate this in xterm! >nG >it's been on my list for a while, but low-priority (the only way I can A >see is to make a bitmap for each glyph, which sounds expensive).> >   G Been there, done that.  The "UISX" product emulated the UIS programmingjL interface from the VWS window system, on top of X11.  It did precisely that,I created it's own client side font glyphs, loaded them as pixmaps, and didsF the "text" drawing by blit operation.  It actually isn't that bad of aL performer.  I needed to be able to do it anyway, since UIS allowed arbitrary2 scaling and rotation of the text and drawing path.  K I would suggest a better solution would be to implement a downloadable fontrL extension.  This would allow you to download a font to the server on-the-flyK in say, BDF format.  This should be fairly simple, since the terminal fontse' themselves are simple monospaced types.E  K DRCS is one area that no emulator that I know of "gets right" because it is J a pain in the ass without the underlying support for defining new fonts on the fly.   ------------------------------  % Date: Thu, 11 Apr 2002 14:58:51 +0100w, From: Paul Williams <flo@uk.thalesgroup.com>3 Subject: Re: Help! VT100 emulators - some problems. 2 Message-ID: <3CB5969B.992990F8@uk.thalesgroup.com>   Mike Rechtman wrote: >  > Been there, done that -iB > Hebrew has two character types - print and cursive. The standard@ > fonts on a Hebrew-enabled VT220 were print only (yes Virginia,< > there are escape codes to make a VT220 work right-to-left)  9 Did a Hebrew-enabled VT220 have a different model number?8  H Is there any chance you could dig out the escape codes that enabled this feature, please?   Regards, Paul   ------------------------------   Date: 11 Apr 2002 15:17:11 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net> 3 Subject: Re: Help! VT100 emulators - some problems.d* Message-ID: <a949dn$me5$1@news1.Radix.Net>  4 Fred Kleinsorge <kleinsorge@star.zko.dec.com> wrote:  M > I would suggest a better solution would be to implement a downloadable fontsN > extension.  This would allow you to download a font to the server on-the-fly  F but then you have to modify the X server (a lot more work for users to build/install).-   -- -= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>  http://dickey.his.com@ ftp://dickey.his.com   ------------------------------  # Date: Thu, 11 Apr 2002 13:57:09 GMTZ# From: "John Smith" <a@nonymous.com>i1 Subject: HP/Compaq - smoking gun or grassy knoll? E Message-ID: <VCgt8.1354$nA1.538@news02.bloor.is.net.cable.rogers.com>t  " HP Chief's Merger Comments Surface Wed Apr 10, 4:33 PM ET& By BRIAN BERGSTEIN, AP Business Writer  L SAN JOSE, Calif. - Two days before Hewlett-Packard Co. shareholders voted onL the contested purchase of Compaq Computer Corp., HP chief Carly Fiorina toldI her top lieutenant she was nervous about the outcome and suggested taking>4 "extraordinary" steps to win over two big investors.  L "If you would take Deutsche Bank, I'll take Northern Trust, get on the phoneJ and see what we can get, but we may have to do something extraordinary forL those two to bring them over the line here," Fiorina said in a voice mail to& chief financial officer Robert Wayman.  K The message was anonymously forwarded to a reporter at the San Jose MercuryfH News, which printed a transcript of it Wednesday and made the audio clip available online..  I The disclosure comes as HP's last-minute moves to win support for the $19fI billion Compaq deal are at the center of a lawsuit against the company by " dissident director Walter Hewlett.  F Hewlett, the son of an HP co-founder and the leader of the proxy fightI against the deal, claims HP improperly got the investment arm of Deutsche2K Bank to switch 17 million votes in favor of the deal by threatening to takeaA future business away from the bank. Deutsche had helped arrange ar< multibillion-dollar line of credit for HP just days earlier.  K Hewlett wants a Delaware judge to throw out Deutsche Bank's votes, which he K believes were enough to give HP what Fiorina called a "slim but sufficient"6' margin of victory in the March 19 vote.g  G Fiorina's voice mail to Wayman apparently was retrieved by someone withtF access to an internal server connected to the company phone system. HPJ confirmed that it was authentic, but claimed that nothing discussed in the  message was improper or illegal.  G Wayman said in a statement that HP executives were constantly assessingaL whether they had effectively pitched the deal to large investors and "did inB fact make extraordinary efforts" in the final days, with dozens of last-minute presentations.  L Still, Hewlett's lawyers believe the voice mail could support their DelawareL lawsuit and expect the message will be turned over in the pretrial discoveryC (news - web sites) phase. The trial is scheduled to begin April 23.-  D Dennis Block, who heads the mergers and acquisitions division at theG Cadwalader, Wickersham & Taft law firm in New York, said the voice mailDK shows that Fiorina and Wayman went all out to lobby investors but is not ind> itself evidence they resorted to improper or coercive tactics.  9 "I do not view the voice mail as a smoking gun," he said.t  F But Charles Elson, director of the corporate governance program at theJ University of Delaware, said the voice mail "certainly does not help" HP's case.t  G "The use of the term `extraordinary' leads one to question what kind ofm% means are extraordinary," Elson said.n  L In her message to Wayman on the night of March 17, Fiorina said she and HP'sH proxy solicitor were nervous that Deutsche Bank and Northern Trust wouldH reject the deal. Deutsche Bank controlled 25 million HP shares; Northern, Trust Global Investors had about 13 million.  L "And so the suggestion is that you call the guy at Deutsche Bank again firstL thing Monday morning," Fiorina said in the voice mail. "And if you don't getK the right answer from him, then you and I need to demand a conference call,lH an audience, etc., to make sure that we get them in the right place. So,G Alan (Miller, HP's proxy solicitor) is feeling like you need a definitenJ answer from the vice chairman, and if it's the wrong one, we have to swing
 into action."I  K Deutsche Bank has refused to comment. Northern Trust Global Investors wouldXL not disclose whether it voted for or against the deal, but said its position( did not change at all in the final days.   ------------------------------  # Date: Thu, 11 Apr 2002 14:50:48 GMT * From: "Bill Todd" <billtodd@metrocast.net>5 Subject: Re: HP/Compaq - smoking gun or grassy knoll?a@ Message-ID: <cpht8.84847$r7.7413452@bin8.nnrp.aus1.giganews.com>  . "John Smith" <a@nonymous.com> wrote in message? news:VCgt8.1354$nA1.538@news02.bloor.is.net.cable.rogers.com...s$ > HP Chief's Merger Comments Surface > Wed Apr 10, 4:33 PM ET( > By BRIAN BERGSTEIN, AP Business Writer > K > SAN JOSE, Calif. - Two days before Hewlett-Packard Co. shareholders votedt onI > the contested purchase of Compaq Computer Corp., HP chief Carly Fiorinai toldK > her top lieutenant she was nervous about the outcome and suggested takinge6 > "extraordinary" steps to win over two big investors.  K Coupled with some of the judge's comments in dismissing HP's attempt to getoK the case thrown out, there does seem to be room for at least some hope thatnF Carly (and the merger) will wind up skewered by her own hand.  But, as' usual, we'll just have to wait and see..   - bill   ------------------------------  % Date: Thu, 11 Apr 2002 13:34:36 +0100CT From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>6 Subject: Re: I was right!  Alpha will live in Itanium!& Message-ID: <3CB582DC.5000506@sun.com>   Bill Todd wrote:  C > "Bob (The Mouth) Ceculski" <bob@instantwhip.com> wrote in messages9 > news:d7791aa1.0204101555.413e3a5a@posting.google.com...  > : >>from the inquirer ... notice the server inflection point- >>Alpha step in write up ... I told you J.F.!e >> > K > You've told people a bunch of things.  Again and again.  But that doesn'te > give them any more weight. > K > And babbling on again today:  first, the Inquirer article (which at leastyN > some people will read) that quotes Terry's article (which otherwise would beL > read by far fewer people) critiquing (and if you read it carefully not allN > that negatively) a Gartner report (the only thing that a whole lot of people1 > will read and many will believe), and now this.e >     9 How long does Bob think it takes to design a new CPU ????s  ; More raving, any server inflection point (whatever that is)g; in 2004 will not have been effected by Intels aquisition ofi% any Alpha designers its way to early.S   Regardst   Andrew Harrisonf    e   ------------------------------  % Date: Thu, 11 Apr 2002 09:48:35 -0400c5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>s6 Subject: Re: I was right!  Alpha will live in Itanium!, Message-ID: <a944iv$70th$1@lead.zk3.dec.com>  2 Andrew Harrison SUNUK Consultancy wrote in message <3CB582DC.5000506@sun.com>...e >t >  >Bill Todd wrote:  >eD >> "Bob (The Mouth) Ceculski" <bob@instantwhip.com> wrote in message: >> news:d7791aa1.0204101555.413e3a5a@posting.google.com... >>; >>>from the inquirer ... notice the server inflection pointp. >>>Alpha step in write up ... I told you J.F.! >>>  >>L >> You've told people a bunch of things.  Again and again.  But that doesn't >> give them any more weight.  >>L >> And babbling on again today:  first, the Inquirer article (which at leastL >> some people will read) that quotes Terry's article (which otherwise would beI >> read by far fewer people) critiquing (and if you read it carefully nots allhH >> that negatively) a Gartner report (the only thing that a whole lot of people2 >> will read and many will believe), and now this. >> >r > : >How long does Bob think it takes to design a new CPU ???? >l< >More raving, any server inflection point (whatever that is)< >in 2004 will not have been effected by Intels aquisition of& >any Alpha designers its way to early. >e  L Yeah, a quote from someone who's company can't design a new CPU chip that is9 even close to the last generation of chips on the market.n  4 I've listened to the Intel NDA.  I'm a happy camper.   ------------------------------  # Date: Thu, 11 Apr 2002 14:46:13 GMTr* From: "Bill Todd" <billtodd@metrocast.net>6 Subject: Re: I was right!  Alpha will live in Itanium!C Message-ID: <Vkht8.314297$2q2.28648921@bin4.nnrp.aus1.giganews.com>$  @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message& news:a944iv$70th$1@lead.zk3.dec.com...   ...#  6 > I've listened to the Intel NDA.  I'm a happy camper.  K You might be a bit less happy if you had had the experience of listening toMF them in 1995.  And 1996.  And 1997.  And 1998.  And 1999.  And 2000...   - bill   ------------------------------    Date: 11 Apr 2002 02:05:10 -04000 From: cstacy@theworld.com (Christopher C. Stacy)  Subject: Re: IA64 is not the VAX( Message-ID: <ur8lm93e1.fsf@theworld.com>  @ >>>>> On 10 Apr 2002 14:13:15 -0700, Eric Smith ("Eric") writes:  9  Eric> cstacy@theworld.com (Christopher C. Stacy) writes:t<  >> Finally, what is the basis for your implied premise that+  >> Lisp is not a general purpose language?n  I  Eric> I reread my posts and don't believe that I ever said anything thatl  Eric> implies that.  J  Eric> But if you build a processor optimized to run Lisp, it's a special-H  Eric> purpose processor.  And if you build a computer around that, it'sG  Eric> a special-purpose computer.  You seem to be the only person thatt  Eric> disagrees.   A Are you aware that the halfword architectecture of the PDP-10 wasd< quite intentionally designed specifically to support Lisp?  9 The Lisp function CAR is known as HLRZ, and CDR is HRRZ. >A I guess the PDP-10 must not have been a general purpose computer!l   ------------------------------    Date: 10 Apr 2002 23:10:26 -07003 From: Eric Smith <eric-no-spam-for-me@brouhaha.com>o  Subject: Re: IA64 is not the VAX0 Message-ID: <qhd6x64vfx.fsf@ruckus.brouhaha.com>  2 cstacy@theworld.com (Christopher C. Stacy) writes:C > Are you aware that the halfword architectecture of the PDP-10 was>> > quite intentionally designed specifically to support Lisp?    D It was intended to support a lot of things.  Lisp was one but didn'tJ get particularly more attention from the designers than Fortran and Algol.   ------------------------------    Date: 11 Apr 2002 02:48:39 -04000 From: cstacy@theworld.com (Christopher C. Stacy)  Subject: Re: IA64 is not the VAX( Message-ID: <uu1qipw6w.fsf@theworld.com>  @ >>>>> On 10 Apr 2002 23:10:26 -0700, Eric Smith ("Eric") writes:  9  Eric> cstacy@theworld.com (Christopher C. Stacy) writes: E  >> Are you aware that the halfword architectecture of the PDP-10 wass@  >> quite intentionally designed specifically to support Lisp?    K  Eric> It was intended to support a lot of things.  Lisp was one but didn'tdQ  Eric> get particularly more attention from the designers than Fortran and Algol.s  = So say you, but Alan Kotok who designed the PDP-6 and PDP-10 l" with Gordon Bell says differently:  D    "Indeed, that's me (without the R.). I was chief architect of theD    PDP-10, many, many years ago. The architecture dated back to 1963?    and the PDP-6, which I worked on with Gordon Bell. And, yes, ;    facilitating a good LISP implementation was an importanta    consideration."   ------------------------------  % Date: Thu, 11 Apr 2002 09:03:22 +0100 ( From: Martyn <mpattKNICKERS@bigfoot.com>. Subject: Re: Insufficient virtual memory error* Message-ID: <3CB5434A.1030600@bigfoot.com>   Witold Waldman wrote:e  A >I have been trying to run a Fortran 90 program that dynamically  A >allocates/deallocates a large array whose size is increased with ? >each call. I need to create arrays around 50-60 MB in size in r) >the application that I am trying to run.  >hJ >I have created a short test program to demonstrate the problem. When the K >program runs (on a Compaq Alphaserver with 256 MB RAM and running OpenVMS  / >7.2), the following error message is produced:e >u >   ZAP$ run testarray >   Allocating 10 MB array.a >   Deallocating 10 MB array.o >   Allocating 15 MB array. 4 >   %FOR-F-INSVIRMEM, insufficient virtual memory!/ ' >     unit !SL  file !AS!/  user PC !XL:3 >   %TRACE-F-TRACEBACK, symbolic stack dump follows-< >    image    module    routine             line      rel PC
 >   abs PCB >    DEC$FORRTL                                 0 000000000005BB78 >   000000007C647B78B >    DEC$FORRTL                                 0 000000000005BDC4 >   000000007C647DC4B >    TESTARRAY  TESTARRAY  TESTARRAY           15 00000000000001B8 >   00000000000301B8B >                                               0 FFFFFFFF84EDD3D4 >   FFFFFFFF84EDD3D4 >3H >When this error occurs, my program is trying to dynamically allocate anK >array that is 15 MB in size. I have asked the current system administratorpM >of the Alphaserver to increase my paging file, which he has done (to 900000 rJ >blocks). I don't know how to check the settings from my account, though.  >eL >The Fortran 90 code of the test program is as follows. It compiles cleanly J >using the command line "fortran testarray.f90" on our system. The COMPAQ 9 >Fortran compiler that we have supports F77, F90 and F95.l >n >        program testarray >r >          implicit none > 2 >          real,allocatable,dimension(:)     :: A1 >          integer::im >  >          i=1024*1024/4 >o/ >          write(*,*) 'Allocating 10 MB array.'p >          allocate(A1(i*10)) 1 >          write(*,*) 'Deallocating 10 MB array.'  >          deallocate(A1) / >          write(*,*) 'Allocating 15 MB array.'V >          allocate(A1(i*15))01 >          write(*,*) 'Deallocating 15 MB array.'  >          deallocate(A1)o/ >          write(*,*) 'Allocating 20 MB array.'t >          allocate(A1(i*20))a1 >          write(*,*) 'Deallocating 20 MB array.'w >          deallocate(A1)r/ >          write(*,*) 'Allocating 40 MB array.'  >          allocate(A1(i*40))e1 >          write(*,*) 'Deallocating 40 MB array.'o >          deallocate(A1)c/ >          write(*,*) 'Allocating 50 MB array.'i >          allocate(A1(i*50))i1 >          write(*,*) 'Deallocating 50 MB array.'n >          deallocate(A1)  >h >          stopr >        end program testarray >oG >Can anyone offer any suggestions that might fix this problem? I think e; >that it is an OpenVMS setup problem related to my account.j > * >Thanks for any help that can be provided. >1 >Witold Waldmane& >witold.waldman@XE1dsto.defence.gov.auB >Remove capital letters and digits to make a valid e-mail address. >tI Depends which VMS you're running but the SYSGEN parameter virtualpagecnt lI could be the problem, get the system manager to check it (if it's there).    -- d, Remove KNICKERS before replying by Email ;-)   ------------------------------  % Date: Thu, 11 Apr 2002 09:33:25 -0400l1 From: "Mark D. Jilson" <jilly@clarityconnect.com> . Subject: Re: Insufficient virtual memory error2 Message-ID: <3CB590A5.1D32B857@clarityconnect.com>  G It will be your pagefile quota that is at issue here.  You will need its> set to allow you to describe the 50MB array plus other programF overhead.  After running the program do a SHOW PROCESS/ACC and look atB your peak usage.  This will be for the successful case of the 10MBG array.  Adjust accordingly for the 50MB array.  If the problem was withb0 VIRTUALPAGECNT you would receive a VASFUL error.  
 Martyn wrote:R >  > Witold Waldman wrote:e > B > >I have been trying to run a Fortran 90 program that dynamicallyC > >allocates/deallocates a large array whose size is increased withV@ > >each call. I need to create arrays around 50-60 MB in size in+ > >the application that I am trying to run.f > > K > >I have created a short test program to demonstrate the problem. When the L > >program runs (on a Compaq Alphaserver with 256 MB RAM and running OpenVMS1 > >7.2), the following error message is produced:s > >e > >   ZAP$ run testarray > >   Allocating 10 MB array.e > >   Deallocating 10 MB array.g > >   Allocating 15 MB array..5 > >   %FOR-F-INSVIRMEM, insufficient virtual memory!/r) > >     unit !SL  file !AS!/  user PC !XL 5 > >   %TRACE-F-TRACEBACK, symbolic stack dump follows.> > >    image    module    routine             line      rel PC > >   abs PCD > >    DEC$FORRTL                                 0 000000000005BB78 > >   000000007C647B78D > >    DEC$FORRTL                                 0 000000000005BDC4 > >   000000007C647DC4D > >    TESTARRAY  TESTARRAY  TESTARRAY           15 00000000000001B8 > >   00000000000301B8D > >                                               0 FFFFFFFF84EDD3D4 > >   FFFFFFFF84EDD3D4 > >8J > >When this error occurs, my program is trying to dynamically allocate anM > >array that is 15 MB in size. I have asked the current system administrator N > >of the Alphaserver to increase my paging file, which he has done (to 900000K > >blocks). I don't know how to check the settings from my account, though.  > >oM > >The Fortran 90 code of the test program is as follows. It compiles cleanlyiK > >using the command line "fortran testarray.f90" on our system. The COMPAQk; > >Fortran compiler that we have supports F77, F90 and F95.  > >e > >        program testarray > >e > >          implicit none > >i4 > >          real,allocatable,dimension(:)     :: A1 > >          integer::if > >  > >          i=1024*1024/4 > > 1 > >          write(*,*) 'Allocating 10 MB array.'  > >          allocate(A1(i*10))r3 > >          write(*,*) 'Deallocating 10 MB array.'i > >          deallocate(A1)u1 > >          write(*,*) 'Allocating 15 MB array.'o > >          allocate(A1(i*15))r3 > >          write(*,*) 'Deallocating 15 MB array.'  > >          deallocate(A1)l1 > >          write(*,*) 'Allocating 20 MB array.'e > >          allocate(A1(i*20))s3 > >          write(*,*) 'Deallocating 20 MB array.'r > >          deallocate(A1)h1 > >          write(*,*) 'Allocating 40 MB array.'n > >          allocate(A1(i*40))f3 > >          write(*,*) 'Deallocating 40 MB array.'l > >          deallocate(A1)r1 > >          write(*,*) 'Allocating 50 MB array.'j > >          allocate(A1(i*50)) 3 > >          write(*,*) 'Deallocating 50 MB array.'t > >          deallocate(A1)u > >e > >          stopt  > >        end program testarray > > H > >Can anyone offer any suggestions that might fix this problem? I think= > >that it is an OpenVMS setup problem related to my account.o > >e, > >Thanks for any help that can be provided. > >i > >Witold Waldmans( > >witold.waldman@XE1dsto.defence.gov.auD > >Remove capital letters and digits to make a valid e-mail address. > >yJ > Depends which VMS you're running but the SYSGEN parameter virtualpagecntK > could be the problem, get the system manager to check it (if it's there).a >  > --. > Remove KNICKERS before replying by Email ;-)   -- uC Jilly	- Working from Home in the Chemung River Valley - Waverly, NYv0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  % Date: Thu, 11 Apr 2002 11:25:40 -0400o1 From: Michael Austin <maustin@firstdbasource.com>y. Subject: Re: Insufficient virtual memory error2 Message-ID: <3CB5AAF4.B11D54D8@firstdbasource.com>   "Mark D. Jilson" wrote:  > I > It will be your pagefile quota that is at issue here.  You will need it @ > set to allow you to describe the 50MB array plus other programH > overhead.  After running the program do a SHOW PROCESS/ACC and look atD > your peak usage.  This will be for the successful case of the 10MBI > array.  Adjust accordingly for the 50MB array.  If the problem was withf2 > VIRTUALPAGECNT you would receive a VASFUL error. >  > Martyn wrote:e > >e > > Witold Waldman wrote:u > >eD > > >I have been trying to run a Fortran 90 program that dynamicallyE > > >allocates/deallocates a large array whose size is increased withhB > > >each call. I need to create arrays around 50-60 MB in size in- > > >the application that I am trying to run.  > > >RM > > >I have created a short test program to demonstrate the problem. When the N > > >program runs (on a Compaq Alphaserver with 256 MB RAM and running OpenVMS3 > > >7.2), the following error message is produced:M > > >u > > >   ZAP$ run testarray > > >   Allocating 10 MB array.P! > > >   Deallocating 10 MB array.b > > >   Allocating 15 MB array.a7 > > >   %FOR-F-INSVIRMEM, insufficient virtual memory!/ + > > >     unit !SL  file !AS!/  user PC !XL 7 > > >   %TRACE-F-TRACEBACK, symbolic stack dump followsD@ > > >    image    module    routine             line      rel PC > > >   abs PCF > > >    DEC$FORRTL                                 0 000000000005BB78 > > >   000000007C647B78F > > >    DEC$FORRTL                                 0 000000000005BDC4 > > >   000000007C647DC4F > > >    TESTARRAY  TESTARRAY  TESTARRAY           15 00000000000001B8 > > >   00000000000301B8F > > >                                               0 FFFFFFFF84EDD3D4 > > >   FFFFFFFF84EDD3D4 > > >dL > > >When this error occurs, my program is trying to dynamically allocate anO > > >array that is 15 MB in size. I have asked the current system administratormP > > >of the Alphaserver to increase my paging file, which he has done (to 900000M > > >blocks). I don't know how to check the settings from my account, though.  > > >rO > > >The Fortran 90 code of the test program is as follows. It compiles cleanlysM > > >using the command line "fortran testarray.f90" on our system. The COMPAQ = > > >Fortran compiler that we have supports F77, F90 and F95.t > > >1 > > >        program testarray > > >n > > >          implicit none > > >t6 > > >          real,allocatable,dimension(:)     :: A1 > > >          integer::i1 > > >  > > >          i=1024*1024/4 > > >M3 > > >          write(*,*) 'Allocating 10 MB array.'i! > > >          allocate(A1(i*10)) 5 > > >          write(*,*) 'Deallocating 10 MB array.'D > > >          deallocate(A1) 3 > > >          write(*,*) 'Allocating 15 MB array.'t! > > >          allocate(A1(i*15))c5 > > >          write(*,*) 'Deallocating 15 MB array.'M > > >          deallocate(A1)t3 > > >          write(*,*) 'Allocating 20 MB array.'.! > > >          allocate(A1(i*20)) 5 > > >          write(*,*) 'Deallocating 20 MB array.'  > > >          deallocate(A1) 3 > > >          write(*,*) 'Allocating 40 MB array.'n! > > >          allocate(A1(i*40))m5 > > >          write(*,*) 'Deallocating 40 MB array.'S > > >          deallocate(A1)c3 > > >          write(*,*) 'Allocating 50 MB array.'.! > > >          allocate(A1(i*50))d5 > > >          write(*,*) 'Deallocating 50 MB array.'  > > >          deallocate(A1)d > > >  > > >          stop " > > >        end program testarray > > >tJ > > >Can anyone offer any suggestions that might fix this problem? I think? > > >that it is an OpenVMS setup problem related to my account.) > > >-. > > >Thanks for any help that can be provided. > > >: > > >Witold WaldmanM* > > >witold.waldman@XE1dsto.defence.gov.auF > > >Remove capital letters and digits to make a valid e-mail address. > > >cL > > Depends which VMS you're running but the SYSGEN parameter virtualpagecntM > > could be the problem, get the system manager to check it (if it's there).m > >r > > --0 > > Remove KNICKERS before replying by Email ;-) >  > --G > Jilly   - Working from Home in the Chemung River Valley - Waverly, NY.L >         - jilly@clarityconnect.com                      - Brett Bodine fanL >         - Mark.Jilson@Compaq.com                        - since 1975 or so5 >         - http://www.jilly.baka.com               -      Look at pgflquo.H I would hope that there will be only a few users running this at a time.F That is a lot of memory.  Are you certain there is not a better way to achieve the goal? Just curious.r     --   Regards,  7 Michael Austin            Registered Linux User #261163W7 First DBA Source, Inc.    http://www.firstdbasource.com5 Sr. Consultant 704-947-1089 (Office)t 704-236-4377 (Mobile)a   ------------------------------    Date: 11 Apr 2002 13:15:31 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: Itanium troubles - Message-ID: <87ofgqlsss.fsf@prep.synonet.com>m    david20@alpha2.mdx.ac.uk writes:  E > In article <bruce-3AF942.10303611042002@copper.ipg.tsnz.net>, Bruce - > Hoult <bruce@hoult.org> writes: >In articlee8 > <2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>,+ > > davidsen@tmr.com (bill davidsen) wrote:   D > >> If I can replace three racks of Xeon mail servers with one rack> > >> of Itanium, given that the systems and o/s have reached aC > >> reliable state and the hardware cost is right, why would I not 
 > >> replace?   D > >Dumb question, but why on earth would you require a whole rack --' > >let alone three -- for mail servers?   A > I'd guess because he has tens of thousands of mail users and isiF > using Exchange.  Later versions of Exchange may have improved but itD > certainly was the case that Exchange didn't scale anything like as# > well as Unix or VMS mail systems.e  A Then drop PMDF on  a pair of DS20Es with a few Km of FDDI between 4 them. Cheaper, faster, smaller, and WAY less power.   @ I think the place for profit in Si(lly) Valley will be in aircon upgrades...r   -- i< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.i@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 11 Apr 2002 06:24:15 GMT) From: leslie@clio.rice.edu (Jerry Leslie)  Subject: Re: Itanium troubles5' Message-ID: <a93a6f$k8d$1@joe.rice.edu>1  ) David Froble (davef@tsoft-inc.com) wrote:5. : Ok, then, keeping in the mode of asking why: : J : Why on earth would anyone with any significant workload use Exchange --  : for mail servers?* : D The only drop-in replacement for Exchange is H-P's OpenMail product,A which enjoyed the same stealth marketing as VMS. It was EOLed andi then turned over to Samsung.  A OpenMail even supports calendaring. So the desktop user could use E Outlook and be unaware that there was a unix/OpenMail server, instead5 of an Exchange server.  F Oracle deep-sixed their Oracle InterOffice system after problems with # mail delivery inside their network:r  ?    http://www.zdnet.com/zdnn/stories/news/0,4586,389501,00.htmlf+    News: Oracle yanks its own e-mail systeme  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Thu, 11 Apr 2002 08:32:48 +0200.3 From: Terje Mathisen <terje.mathisen@hda.hydro.com>t Subject: Re: Itanium troublese- Message-ID: <3CB52E10.D695AA63@hda.hydro.com>e   bill davidsen wrote:I >   And I'm not sure if the 68k was really big endian in the usual sense, H > where bytes 1234 is little, 4321 is big, and I have the odd memory forG > some reason that the original 68k (16bit bus) was 2143. Since my last G > 68k was a "Unix-PC" (aka ATT 7300 and 3B1, modulo some minor cosmetic.F > changes), I'm unable to test that, and unwilling to do any research.$ > Someone will know and pontificate.  3 I believe only DEC ever made 'middle-endian' boxen.m   Terje  -- n  - <Terje.Mathisen@hda.hydro.com>@ "almost all programming can be viewed as an exercise in caching"   ------------------------------  % Date: Thu, 11 Apr 2002 08:38:29 +0200e3 From: Terje Mathisen <terje.mathisen@hda.hydro.com>  Subject: Re: Itanium troubles - Message-ID: <3CB52F65.A980CC6B@hda.hydro.com>v   aaron spink wrote: > 7 > "David Froble" <davef@tsoft-inc.com> wrote in messages& > news:3CB50289.30403@tsoft-inc.com...0 > > Ok, then, keeping in the mode of asking why: > >MK > > Why on earth would anyone with any significant workload use Exchange --U
 > for mail > > servers? > > K > Because it is/was the thing to do.  I remember that when IBM bought LotusrK > the edict came down that the company would switch to Notes.  My memory is L > kinda hazy( maybe Del will pipe in here), but all of Rochester was runningN > mail off of a fairly old single AS/400.  They ended up having to clear out aN > whole machine room and stuff it with x86 boxen just to replace what was in 2 > racks.  F We're running Notes on our IBM aix cluster, specifically to get rid of some racks of NT/x86 servers.i  = Since Notes is an IBM product, this should scale well, right?i   Terjee   -- o  - <Terje.Mathisen@hda.hydro.com>@ "almost all programming can be viewed as an exercise in caching"   ------------------------------  % Date: Thu, 11 Apr 2002 18:42:22 +1200t# From: Bruce Hoult <bruce@hoult.org>  Subject: Re: Itanium troubles-= Message-ID: <bruce-E681AC.18422211042002@copper.ipg.tsnz.net>   - In article <3CB52E10.D695AA63@hda.hydro.com>,j5  Terje Mathisen <terje.mathisen@hda.hydro.com> wrote:s   > bill davidsen wrote:K > >   And I'm not sure if the 68k was really big endian in the usual sense,mJ > > where bytes 1234 is little, 4321 is big, and I have the odd memory forI > > some reason that the original 68k (16bit bus) was 2143. Since my lastRI > > 68k was a "Unix-PC" (aka ATT 7300 and 3B1, modulo some minor cosmetictH > > changes), I'm unable to test that, and unwilling to do any research.& > > Someone will know and pontificate. > 5 > I believe only DEC ever made 'middle-endian' boxen.e  E And that was only the VAX FP format, for data compatability with the t PDP-11.o   -- Bruce   ------------------------------   Date: 11 Apr 2002 07:31:36 GMT+ From: mailer-daemon@bof.de (Patrick Schaaf)o Subject: Re: Itanium troublesl5 Message-ID: <3cb53bd8$0$165$9b622d9e@news.freenet.de>d  % Bruce Hoult <bruce@hoult.org> writes:t  B >In article <2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>,) > davidsen@tmr.com (bill davidsen) wrote:r  E >> If I can replace three racks of Xeon mail servers with one rack ofjH >> Itanium, given that the systems and o/s have reached a reliable state; >> and the hardware cost is right, why would I not replace?c  G >Dumb question, but why on earth would you require a whole rack -- let 4! >alone three -- for mail servers?   B As an example, we are running about 20 racks full of mail servers.2 That's about 4 million active POP3/IMAP mailboxen.  C We use bread-and-butter x86 servers for the task, and neither blade D servers, nor "bigger server boxes", would be more cost-effective for" such a trivially partitioned task.  N I don't know how this equation looks for people running MS Exchange, though...   best regards	   Patrickp   ------------------------------  % Date: Thu, 11 Apr 2002 19:56:17 +1200G# From: Bruce Hoult <bruce@hoult.org>w Subject: Re: Itanium troubles = Message-ID: <bruce-3B45B4.19561711042002@copper.ipg.tsnz.net>o  5 In article <3cb53bd8$0$165$9b622d9e@news.freenet.de>,,-  mailer-daemon@bof.de (Patrick Schaaf) wrote:g  ' > Bruce Hoult <bruce@hoult.org> writes:t > D > >In article <2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>,+ > > davidsen@tmr.com (bill davidsen) wrote:- > G > >> If I can replace three racks of Xeon mail servers with one rack of J > >> Itanium, given that the systems and o/s have reached a reliable state= > >> and the hardware cost is right, why would I not replace?r > I > >Dumb question, but why on earth would you require a whole rack -- let  # > >alone three -- for mail servers?o > D > As an example, we are running about 20 racks full of mail servers.4 > That's about 4 million active POP3/IMAP mailboxen. > E > We use bread-and-butter x86 servers for the task, and neither blade F > servers, nor "bigger server boxes", would be more cost-effective for$ > such a trivially partitioned task. > P > I don't know how this equation looks for people running MS Exchange, though...  H So you're using *BSD or Linux?  How much bandwidth have you got feeding  into those boxes?w   -- Bruce   ------------------------------  % Date: Thu, 11 Apr 2002 20:01:20 +1200a+ From: "G Swaine" <g.swaine@the.gluepot.com>t Subject: Re: Itanium troubles - Message-ID: <a93gv5$1l8v$1@news.comnet.co.nz>i  6 "Skipper Smith" <skipper@no-spam.calweb.com> wrote ...) > bill davidsen <davidsen@tmr.com> wrote:l# > > Dan Pop <Dan.Pop@ifh.de> wrote: 1 > >| "Bill Todd" <billtodd@metrocast.net> writes:eF > >| >POWER4 and I think SPARC and maybe MIPS) will run either-endian.	 Linux wase@ > >| >developed primarily on little-endian machines (x86s), runs little-endian onF > >| >Alpha (though it might run big-endian somewhere - I don't know), andt > >|< > >| The first Linux port was to m68k, which is a big-endian
 processor. > >0/ > >  The first Linux port was the i386 version.r > >.C > >  And I'm not sure if the 68k was really big endian in the usual- sense, >3B > The 68K was and still is absolutely big endian in every sense of byte > ordering.z >dE > >where bytes 1234 is little, 4321 is big, and I have the odd memory# for C > >some reason that the original 68k (16bit bus) was 2143. Since mya last? > >68k was a "Unix-PC" (aka ATT 7300 and 3B1, modulo some minor  cosmetic= > >changes), I'm unable to test that, and unwilling to do any 	 research.-% > >Someone will know and pontificate.0  1 I would test for endianness in the following way:hF (Assuming "byte = 8 bits", and fetches will "zero extend" if necessary to fill a CPU register)   7     fetch two bytes from memory address into a register (     if value in register is 258  (01 02)         processor is big endianw,     else if value in register is 513 (02 01)"         processor is little endian     else7         processor may be malfunctioning (or is unknown)r	     endifi  6     fetch four bytes from memory address into register:     if the value in the register is 16909060 (01 02 03 04)         processor is big endianl+     else if value is 67305985 (04 03 02 01)i"         processor is little endian     else'         processor endianness is unknowno	     endift       memory address:e/         define four byte sequence:  01 02 03 04i   > >e > >--l< > >bill davidsen <davidsen@tmr.com> CTO, TMR Associates, IncF > >  Programming without software engineering is like sculpting with a chainpF > >saw. The very talented can produce a work of art, the mediocre wind up withmE > >a misshapen lump in a pile of rubble, and in neither case does then endeF > >result have more than a passing resemblance to the original intent. >o > --D > Skipper Smith                         Helpful Knowledge ConsultingD > Worldwide         Microprocessor       Architecture       TrainingD > PowerPC, ColdFire, 68K, CPU32                Hardware and Software > D > /* Remove no-spam. from the reply address to send mail directly */   --   Quote for the month-  @ "There are two kinds of computers: obsolete, and obsolete-ready"! (Tim Shoppa, comp.arch newsgroup)r   --  B g.swaine@the.gluepot is a bogus address. I can be contacted via an> intermediary: gem at gem win co nz. I apologise to the genuine( respondents that this may inconvenience.   ------------------------------   Date: 11 Apr 2002 08:19:46 GMT+ From: mailer-daemon@bof.de (Patrick Schaaf)u Subject: Re: Itanium troublesi6 Message-ID: <3cb54722$0$4052$9b622d9e@news.freenet.de>  % Bruce Hoult <bruce@hoult.org> writes:t  E >> As an example, we are running about 20 racks full of mail servers.F5 >> That's about 4 million active POP3/IMAP mailboxen.s >>  F >> We use bread-and-butter x86 servers for the task, and neither bladeG >> servers, nor "bigger server boxes", would be more cost-effective forn% >> such a trivially partitioned task.e   >So you're using *BSD or Linux?h  @ Linux, exim as MTA, homegrown IMAP/POP3 servers to talk to MUAs,E and OpenLDAP providing the glue. The system has four "scaling groups"-   	- LDAP- 	- incoming MX 	- outgoing SMTP queue 	- mailbox machinest  F Each layer is scaled independant of the others, and uses a specializedL disk drive arrangement. There's about 40 mailbox machines, steadily growing.  : >How much bandwidth have you got feeding into those boxes?  C 2xOC48.  We're a data center that happens to also be part of a ring C in an IP core network. Aggregate data center traffic is 1-2 gbit/s,lD with the mail stuff being a (significant, but not overwhelming) part of that.  F I'd have to do some work to figure out the actual bandwidth use of theE mail system, or other system details, and I'm reluctant to post them.s. If you are curious, please contact me by mail.   best regards	   Patrickt   ------------------------------   Date: 11 Apr 2002 08:22:14 GMT From: Dan.Pop@ifh.de (Dan Pop) Subject: Re: Itanium troubles * Message-ID: <a93h3m$9qc$1@sunnews.cern.ch>  a In <Tm2t8.7279$bt3.358627480@newssvr17.news.prodigy.com> davidsen@tmr.com (bill davidsen) writes:m  K >In article <a8ugii$lr7$1@sunnews.cern.ch>, Dan Pop <Dan.Pop@ifh.de> wrote:Kg >| In <NKus8.66823$w7.5656854@bin6.nnrp.aus1.giganews.com> "Bill Todd" <billtodd@metrocast.net> writes:  >| C >|  O >| >POWER4 and I think SPARC and maybe MIPS) will run either-endian.  Linux was0O >| >developed primarily on little-endian machines (x86s), runs little-endian on4H >| >Alpha (though it might run big-endian somewhere - I don't know), and >| 1E >| The first Linux port was to m68k, which is a big-endian processor.  > - >  The first Linux port was the i386 version.   1 Then, what was the original development platform?r  H >  And I'm not sure if the 68k was really big endian in the usual sense,G >where bytes 1234 is little, 4321 is big, and I have the odd memory forh8 >some reason that the original 68k (16bit bus) was 2143.  2 You're confusing the original 68k with the PDP-11.   >Since my lastF >68k was a "Unix-PC" (aka ATT 7300 and 3B1, modulo some minor cosmeticE >changes), I'm unable to test that, and unwilling to do any research.t  J You don't have to do any research: if the original 68k were middle-endian,H all the other members of the family would have used the same byte order,G for binary compatibility reasons.  This is not an architectural featurei9 that gets changed during the lifetime of an architecture.   # >Someone will know and pontificate.   D This is an implicit assumption that I didn't know what I was talking about in my previous post.   Dan* -- Dan Popt DESY Zeuthen, RZ group Email: Dan.Pop@ifh.de1   ------------------------------  % Date: Thu, 11 Apr 2002 20:58:54 +1200l# From: Bruce Hoult <bruce@hoult.org>e Subject: Re: Itanium troubles.= Message-ID: <bruce-2CB466.20585411042002@copper.ipg.tsnz.net>A  6 In article <3cb54722$0$4052$9b622d9e@news.freenet.de>,-  mailer-daemon@bof.de (Patrick Schaaf) wrote:M  ' > Bruce Hoult <bruce@hoult.org> writes:  > G > >> As an example, we are running about 20 racks full of mail servers.(7 > >> That's about 4 million active POP3/IMAP mailboxen.  > >> SH > >> We use bread-and-butter x86 servers for the task, and neither bladeI > >> servers, nor "bigger server boxes", would be more cost-effective for ' > >> such a trivially partitioned task.  > ! > >So you're using *BSD or Linux?  > B > Linux, exim as MTA, homegrown IMAP/POP3 servers to talk to MUAs,G > and OpenLDAP providing the glue. The system has four "scaling groups"  > 	 > 	- LDAPp > 	- incoming MX > 	- outgoing SMTP queue > 	- mailbox machinest > H > Each layer is scaled independant of the others, and uses a specializedN > disk drive arrangement. There's about 40 mailbox machines, steadily growing. > < > >How much bandwidth have you got feeding into those boxes? > E > 2xOC48.  We're a data center that happens to also be part of a ringgE > in an IP core network. Aggregate data center traffic is 1-2 gbit/s,oF > with the mail stuff being a (significant, but not overwhelming) part
 > of that. > H > I'd have to do some work to figure out the actual bandwidth use of theG > mail system, or other system details, and I'm reluctant to post them. 0 > If you are curious, please contact me by mail.   That OK ... it's BIG.   F If you need 40 boxes then any business with, say, a mere T3 (45 Mbps) # should do fine with a single box...i  * Unless you're running Exchange, of course.   -- Bruce   ------------------------------  % Date: Thu, 11 Apr 2002 10:54:08 +0200t( From: Bernd Paysan <bernd.paysan@gmx.de> Subject: Re: Itanium troublest" Message-ID: <hvi39a.f08.ln@miriam>   JF Mezei wrote:0  E > Because some manager decided to standardize on microsoft because it D > is easy to find microsoft weenies to operate your servers, whereasB > if you buy some proprietary system, it is harder to find people.  A Strange. Who redefined the word "proprietary" recently? Ask your c@ manager which system is proprietary, and which are open source: D Exchange, sendmail, or qmail. And - after the initial configuration C is done - which of them needs handholding all day long, which just wC require security patches once in a while, and which runs for years   without touching.t   -- g Bernd Paysan7 "If you want it done right, you have to do it yourself"h http://www.jwdt.com/~paysan/   ------------------------------   Date: 11 Apr 2002 09:15:02 GMT+ From: mailer-daemon@bof.de (Patrick Schaaf)e Subject: Re: Itanium troubles 5 Message-ID: <3cb55416$0$153$9b622d9e@news.freenet.de>e  % Bruce Hoult <bruce@hoult.org> writes:i  G >If you need 40 boxes then any business with, say, a mere T3 (45 Mbps) h$ >should do fine with a single box...  K Most likely, yes. It could probably also route and firewall the T3 as well,a if you were suicidal.   D For an internal mailserver, I'd still suggest at least seperate SMTP: and mailbox systems, with LDAP running replicated on both.  D Again, if you were suicidal, they could also be failover routers and firewalls for your LAN :-)   best regards	   Patrickn   ------------------------------  + Date: Thu, 11 Apr 2002 09:40:18 +0000 (UTC)e From: david20@alpha2.mdx.ac.uk Subject: Re: Itanium troubles + Message-ID: <a93lm2$f2g$1@aquila.mdx.ac.uk>   U In article <3CB50289.30403@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:e  >david20@alpha2.mdx.ac.uk wrote: >af >> In article <bruce-3AF942.10303611042002@copper.ipg.tsnz.net>, Bruce Hoult <bruce@hoult.org> writes: >>  D >>>In article <2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>,* >>>davidsen@tmr.com (bill davidsen) wrote: >>R >> I'd guess because he has tens of thousands of mail users and is using Exchange.M >> Later versions of Exchange may have improved but it certainly was the casenP >> that Exchange didn't scale anything like as well as Unix or VMS mail systems. >p >e- >Ok, then, keeping in the mode of asking why:  > R >Why on earth would anyone with any significant workload use Exchange -- for mail 	 >servers?r >i >Davee  ) Political decision by senior management ?t  M (and of course -- "We must have it because it includes inbuilt calendering.")f  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  + Date: Thu, 11 Apr 2002 10:02:56 +0000 (UTC), From: david20@alpha2.mdx.ac.uk Subject: Re: Itanium troubless+ Message-ID: <a93n0g$fsh$1@aquila.mdx.ac.uk>b  \ In article <87ofgqlsss.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:! >david20@alpha2.mdx.ac.uk writes:  >tF >> In article <bruce-3AF942.10303611042002@copper.ipg.tsnz.net>, Bruce. >> Hoult <bruce@hoult.org> writes: >In article9 >> <2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>,U, >> > davidsen@tmr.com (bill davidsen) wrote: >cE >> >> If I can replace three racks of Xeon mail servers with one rackr? >> >> of Itanium, given that the systems and o/s have reached aDD >> >> reliable state and the hardware cost is right, why would I not >> >> replace? >aE >> >Dumb question, but why on earth would you require a whole rack ---( >> >let alone three -- for mail servers? > B >> I'd guess because he has tens of thousands of mail users and isG >> using Exchange.  Later versions of Exchange may have improved but itaE >> certainly was the case that Exchange didn't scale anything like asn$ >> well as Unix or VMS mail systems. >.B >Then drop PMDF on  a pair of DS20Es with a few Km of FDDI between5 >them. Cheaper, faster, smaller, and WAY less power. i >aA >I think the place for profit in Si(lly) Valley will be in aircona >upgrades... >   K I agree that PMDF is a realy great product (I use it on my systems) howevertL almost any VMS or Unix based mail solution eg Sendmail, Exim, Cyrus IMAP etc) will probably scale better than Exchange.r  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Thu, 11 Apr 2002 12:05:30 +0100o& From: Ken Green <Ken.Green@kgcc.co.uk> Subject: Re: Itanium troublesc* Message-ID: <3CB56DFA.CB89428B@kgcc.co.uk>   david20@alpha2.mdx.ac.uk wrote:e   >aM > I agree that PMDF is a realy great product (I use it on my systems) howeveriN > almost any VMS or Unix based mail solution eg Sendmail, Exim, Cyrus IMAP etc+ > will probably scale better than Exchange.e >h > David Webb > VMS and Unix team leader > CCSS > Middlesex University  H Whilst almost anything might scale better than Exchange, the normal UnixH mbox mail form doesn't scale well for mail clients. It's not to bad withD something like POP where typically the client empties the file every time they connect.  K When it's used to store mail messages though it's horrid. Firstly, there is2G no sharing of storage, each recipient gets a private copy of a message. B For corperate email systems this is a real downer since managementK types like to send out huge great MS documents to large distribution lists.e  K The biggest issue though is with the fact all your messages are in a singlelO file. Currently my in box has over 500 messages in it, every time I do anythingeG in here the whole file would need to be processed. Say delete a messagetI from near the start any the nearly the whole file needs to be re-written.0I Mail box files of over 200MB aren't uncommon, processing this lot is slowIL even for single user systems. When there are thousands of actively connected  users this doesn't work to well.  G Storing email on the server is a good idea in many cases (even when the E users keep local copies on their PC/laptop) since it's easy to backupa? the server, and impractical to backup everyones individual PCs.t  F Of course Outlook doesn't help here by lumping all the local mail intoE a huge great file, the file keeps changing so the whole file needs tor be re-backed up.   Cheers   Ken-   ------------------------------  % Date: Thu, 11 Apr 2002 14:03:22 +0200 * From: Alexis Cousein <al@brussels.sgi.com> Subject: Re: Itanium troublesA/ Message-ID: <3CB57B8A.6020103@brussels.sgi.com>r   bill davidsen wrote:  H >   I never knew UNICOS ran on anything but Cray (we had Cray2 and XMP).    B He did say Unicos/MK, and the T3Es *were* Cray machines with Alpha> processors (and a very unusual memory hierarchy architecture).   ------------------------------    Date: 11 Apr 2002 07:48:21 -0500- From: koehler@encompasserve.org (Bob Koehler)h Subject: Re: Itanium troublesm3 Message-ID: <3dVQJlpU3MVl@eisner.encompasserve.org>c  j In article <lf2t8.7278$TY3.358055008@newssvr17.news.prodigy.com>, davidsen@tmr.com (bill davidsen) writes: > I >   Many, if not all, of the MIPS would run either endian, I had some old 1 > DEC Ultrix machines which were configured big. m  E    IIRC Ultirx was never ported from little-endian dependencies builtc!    into the original VAX version.w   ------------------------------    Date: 11 Apr 2002 07:49:54 -0500- From: koehler@encompasserve.org (Bob Koehler)o Subject: Re: Itanium troublesp3 Message-ID: <ecueSAk6KuP4@eisner.encompasserve.org>y  h In article <slrnab9rf0.1tb1.skipper@web1.calweb.com>, skipper@no-spam.calweb.com (Skipper Smith) writes: > G > The 68K was and still is absolutely big endian in every sense of byteA > ordering.  >   F    Of course the follow-on PowerPC is bi-endian like most current RISC    machines.   ------------------------------   Date: 11 Apr 2002 13:04:05 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troublesp0 Message-ID: <a941k5$ntv$1@pegasus.csx.cam.ac.uk>  c In article <3dVQJlpU3MVl@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:-m |> In article <lf2t8.7278$TY3.358055008@newssvr17.news.prodigy.com>, davidsen@tmr.com (bill davidsen) writes:t |> > cL |> >   Many, if not all, of the MIPS would run either endian, I had some old4 |> > DEC Ultrix machines which were configured big.  |> lH |>    IIRC Ultirx was never ported from little-endian dependencies built$ |>    into the original VAX version.  ; MIPS Ultrix was mixed-endian, though I now forget the exactw, ordering.  It was quite a useful testbed :-)     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679-   ------------------------------  % Date: Thu, 11 Apr 2002 16:59:18 +0200 ( From: Bernd Paysan <bernd.paysan@gmx.de> Subject: Re: Itanium troubles-" Message-ID: <6c849a.rmk.ln@miriam>   Ken Green wrote:  E > Whilst almost anything might scale better than Exchange, the normallF > Unix mbox mail form doesn't scale well for mail clients. It's not toD > bad with something like POP where typically the client empties the > file every time they connect.   E It's not that bad on the client side, either. All reasonable clients nC index the mbox file. Only once in a while the deleted messages are 0 cleaned up.e  D mbox folder with one file per mail would be "nice to have", though, D and having a mail server that links all identical mails to the same B file, too. Some years ago, when Hans Reiser said that he wants to A have the file system be the database, I proposed to add a way to  C access directories as if they were structured files - in this case lB you would have a plugin that makes the mbox directory look like a D file to clients that don't know about this format improvement. mbox 2 was one of the prominent examples, back then, too.  D > When it's used to store mail messages though it's horrid. Firstly,D > there is no sharing of storage, each recipient gets a private copyA > of a message. For corperate email systems this is a real downeryD > since management types like to send out huge great MS documents to > large distribution lists.i  B But that's just abusing the mail system. Large distribution lists E (broadcast) should be send by NNTP (news). And large files should be nA put on a web-page, i.e. the news should just have an URL. And M$ sD documents should be converted to a printable format before sending. B The boss's motivation letter, written in Word (two paragraphs, no C formating, 200MB) contains all the other letters the boss wrote in g+ his entire carreer in the undo history ;-).s   --   Bernd Paysan7 "If you want it done right, you have to do it yourself"  http://www.jwdt.com/~paysan/   ------------------------------  % Date: Thu, 11 Apr 2002 11:58:04 -0400-( From: David Froble <davef@tsoft-inc.com> Subject: Re: Itanium troublesa, Message-ID: <3CB5B28C.4010009@tsoft-inc.com>   JF Mezei wrote:.   > David Froble wrote:o > R >>Why on earth would anyone with any significant workload use Exchange -- for mail
 >>servers? >> > O > Because they made a commitment to evil Bill to do so, cannabalising their own14 > more profitable product just to please Bill Gates.    K So, you're implying that the majority of today's computer users are stupid?2    O > Because some manager decided that the company should standardize on MicrosofteP > because that was sure to survive while other products don't have such a bright
 > future ?    6 Well, 'some manager' is doing his part to insure that.    P > Because some manager decided to standardize on microsoft because it is easy toI > find microsoft weenies to operate your servers, whereas if you buy somei2 > proprietary system, it is harder to find people.    L People can learn whatever the job requires.  There aint no such thing as an K already trained employee, regardless of what managers wish for.  There are  P skills, but the employee still must learn the business of the company, or, he's  not helping the company.  P A few years back people were putting their resumes on Monster.COM, or wherever, Q and playing musical jobs, like we played musical chairs as young children.  It'd pM be interesting to see which segment, such as windoz, of the job seekers were aB most mobile.  What good is finding someone that you cannot retain?  K Finding people wouldn't be such an issue if you weren't running such labor a intensive systems.    H > The list of reasons can be long. The fact that Microsoft maintains itsP > monopoly is an indication that product quality isn't so important as following0 > the trend to make sure you're not left behind. >   N Don't think.  Just follow the crowd.  Hey, that's how the Lemmings do it, the N whole bunch follow each other off the clift into the ocean, if I remember the  story correctly.  , I think my original question is still valid.   Dave   ------------------------------  % Date: Thu, 11 Apr 2002 12:04:06 -04008( From: David Froble <davef@tsoft-inc.com> Subject: Re: Itanium troubleso* Message-ID: <3CB5B3F6.90209@tsoft-inc.com>   Paul Repacholi wrote:e  " > david20@alpha2.mdx.ac.uk writes: >  > E >>In article <bruce-3AF942.10303611042002@copper.ipg.tsnz.net>, Brucei- >>Hoult <bruce@hoult.org> writes: >In articlea8 >><2W2t8.7282$9X3.358686772@newssvr17.news.prodigy.com>, >>* >>>davidsen@tmr.com (bill davidsen) wrote: >>>  > C >>>>If I can replace three racks of Xeon mail servers with one racko= >>>>of Itanium, given that the systems and o/s have reached akB >>>>reliable state and the hardware cost is right, why would I not >>>>replace? >>>> > D >>>Dumb question, but why on earth would you require a whole rack --' >>>let alone three -- for mail servers?i >>>8 > A >>I'd guess because he has tens of thousands of mail users and is0F >>using Exchange.  Later versions of Exchange may have improved but itD >>certainly was the case that Exchange didn't scale anything like as# >>well as Unix or VMS mail systems.e >> > C > Then drop PMDF on  a pair of DS20Es with a few Km of FDDI betweene6 > them. Cheaper, faster, smaller, and WAY less power.     K That reminds me of a conversation with a friend a year or so back.  He was  O working on some software to run on one or two, cannot remember, DS20s for mail iK handling.  He was pretty confident of the capabilities.  His statement was tN something like "This thing is going to be so fast that when it's fired up the M whole Internet will hear a big sucking sound from the mail being pulled in." gE Not sure what kind of communications were going to be available.  :-)4   Dave   ------------------------------  % Date: Thu, 11 Apr 2002 14:20:16 +0100,% From: Alan Greig <a.greig@virgin.net>)0 Subject: Re: Just another linux nightmare day #28 Message-ID: <r61bbucopmpo2jvtctm09j4icscqeic3lc@4ax.com>  F On Tue, 09 Apr 2002 12:52:17 +0100, Nic Clews <sendspamhere@127.0.0.1> wrote:   >Patrick Young wrote:- >-D >> I'm sure Bob is well meaning in his efforts, even if they may not# >> be in the "best possible taste".c >-= >...recollections of Kenny Everett and his TV show, smiles...   $ How could we forget "Cupid Stunt" !.    But I'm telling you the plot :-)  F Non-UK residents (or followers of Brittish comedies) will have no idea what we're talking about.    ------------------------------  % Date: Thu, 11 Apr 2002 13:45:05 +0200m9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>e< Subject: Mail system scalability (Was: Re: Itanium troubles)' Message-ID: <3CB57741.F994CC87@aaa.com>n  6 The best mail system I know in terms of scalability is3 MEMO. Jointly delveloped by Ericsson and Vovlo, buta7 not used very much today. It run on IBM mainframe under.; MVS. One single system supported all MEMO users at Ericssoni5 worlwide, 50.000+ users. User interface was 3270. AndtB since it was one single system, any message sent to a distribution= list was only stored as one copy, even if the receptients was = spread over the whole world. Distinution time was zero (sincec" nothing realy got "sent" anyware).  = Central backup, nothing on the "clients", central management.e   Jan-Erik Sderholm.d   ------------------------------  % Date: Thu, 11 Apr 2002 12:57:15 +0100.4 From: John Laird <john@laird-towers.freeserve.co.uk> Subject: Re: Memory Corruption8 Message-ID: <n0uabusm7siqknacr06m9vdkge99n40e98@4ax.com>  J On Thu, 11 Apr 2002 15:08:27 +0010, paddy.o'brien@zzz.tg.nsw.gov.au wrote:  
 >JLSue wrote:n >rA >>I had this problem once-upon-a-time when I declared a character F >>variable in Fortran to some size (say 30 characters), and then triedH >>to stuff in 31 (or more) characters of data.  At first it worked fine,D >>but one day it started doing strange things after a recompile.  ItE >>seems that the variables were mapped differently in memory, and thenF >>last byte got stuffed into an integer variable that just happened toE >>get mapped directly after the character variable.  And the order ofwC >>declaration did not determine which variable mapped where (again,aG >>iirc).  If I had been luckier, the extra data would have been stuffedwD >>into nowrite pages (i.e., code) and I'd have gotten an ACCVIO, butF >>instead I spent days trying to figure it via debug (kept seeing thisB >>variable's value change when I was deep down into a an area that >>didn't use it).  >lN >If you obtained this result from a straight assignment, then there was a bug  >in the compiler.p >h >CHARACTER A*30,B*31 >... >A = B > M >should only move 30 bytes.  It should not try to "stuff" the 31st byte of B  O >anywhere.  Possibly C, via strcpy, could do anything with it, but not Fortran.s  	 Yes, but:    CHARACTER A*30 ...t4 N = (some expression that at run-time equates to 31)
 A(1:N) = B  I will silently walk off the end of A if bounds-checking is not enabled.  IrK don't recall off-hand if the compiler will always emit an error if N can be2H seen to be 31 at compile-time, or again whether this is dependent on the bounds checking. -- D
 John Laird Yezerski Roper Ltd   ------------------------------  % Date: Thu, 11 Apr 2002 14:26:51 +0100i% From: Alan Greig <a.greig@virgin.net>c Subject: Re: Memory Corruption8 Message-ID: <mk3bbuklt8utnhcmffne4ein8u152le1ng@4ax.com>  A On Tue, 09 Apr 2002 16:56:10 GMT, jlsue <jlsuexxxz@insightbb.com>6 wrote:   >1D >Here's another thing to try.  When compiling optimized, turn on allD >checks.  For Fortran, this was /check=all iirc.  At the very least, >put in /checks=bounds.r  E When talking to CA high level support about a problem in one of theiraA MANMAN routines (Fortran) I said I'd tracked it down by compilingyD /check=bounds. The guy, who supposedly was one of their main coders,# responded with "what does that do?"n   Eeek!o  E >What you will then get is an accvio if/when your code tries to stuffi >too much into a variable. > @ >I had this problem once-upon-a-time when I declared a characterE >variable in Fortran to some size (say 30 characters), and then triednG >to stuff in 31 (or more) characters of data.  At first it worked fine,2C >but one day it started doing strange things after a recompile.  ItmD >seems that the variables were mapped differently in memory, and theE >last byte got stuffed into an integer variable that just happened tooD >get mapped directly after the character variable.  And the order ofB >declaration did not determine which variable mapped where (again,F >iirc).  If I had been luckier, the extra data would have been stuffedC >into nowrite pages (i.e., code) and I'd have gotten an ACCVIO, butcE >instead I spent days trying to figure it via debug (kept seeing thisrA >variable's value change when I was deep down into a an area that  >didn't use it). >aA >Compiling with the checks can help detect this situation easier.  >>2 >Not speaking for anyone, certainly not DEC/Compaq. >(get rid of the xxxz in my address to e-mail)   ------------------------------  # Date: Thu, 11 Apr 2002 15:04:45 GMT % From: jlsue <jlsuexxxz@insightbb.com>o Subject: Re: Memory Corruption8 Message-ID: <q19bbu4be0girlva6evt94kgiak50k4bmv@4ax.com>  C On Thu, 11 Apr 2002 15:08:27 +0010, paddy.o'brien@zzz.tg.nsw.gov.au  wrote:   > N >If you obtained this result from a straight assignment, then there was a bug  >in the compiler.. >  >CHARACTER A*30,B*31 >... >A = B >(M >should only move 30 bytes.  It should not try to "stuff" the 31st byte of B yO >anywhere.  Possibly C, via strcpy, could do anything with it, but not Fortran.i >o  F Hmm... well, it's been about 10 years now, so I can't recall the exactA nature of the code.  But it was Fortran, I can say that for sure. C It's possible that it was more something like declaring the string,2B and using it in a call to a non-Fortran VMS system routine of some sort.i  F It had never occurred to me at the time that VMS would allow something to overrun the bounds.  1 Not speaking for anyone, certainly not DEC/Compaq - (get rid of the xxxz in my address to e-mail)    ------------------------------  % Date: Thu, 11 Apr 2002 17:23:31 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>n Subject: Re: Memory Corruption' Message-ID: <3CB5AA73.244BD437@aaa.com>m  5 Well, it isn't *VMS* that allows something to overruni: the bounds, as long as it stays in the writable, data area of the actual process.  : It's the specific language RTL that should check this kindC of bounds, isn't it ? And it often possible to set te checks on/off < (off to gain some performance) via switches to the compiler.   Jan-Erik     jlsue wrote: > H > It had never occurred to me at the time that VMS would allow something > to overrun the bounds.   ------------------------------  # Date: Thu, 11 Apr 2002 10:41:23 GMT 5 From: peter.watkinson1@ntlworld.com (Peter Watkinson) @ Subject: Merger problems was Re: WIreless networking + VMS/Tru641 Message-ID: <3cb56843.2955171@news.cable.ntl.com>r  D On Thu, 11 Apr 2002 06:58:06 +0000 (UTC), mustang@ucc.asn.au.invalid wrote:  + >Main, Kerry <Kerry.Main@compaq.com> wrote:v	 >: Peter,e > J >: I suspect what you want is to access VMS applications via WAP devices - >: is this correct?  >iF >I suspect Peter wants to participate in a wireless LAN with an Alpha.G >Peter, I recommend that you either check out a Cisco Workgroup bridge,pC >or one of these: http://www.enterasys.com/products/items/CSISA-AX/m >c >Cheersb >/dave >-- & >" I don't get mad.... I get stabby. " >- William "Fat Tony" Williams.t    > Quite simple really. I want to connect my Cable modem internetE connection which is downstairs to a VMS/Tru64 workstation (when I get-@ it) which will be upstairs without having to drill a hole in the ceiling. :-)  B This leads me onto another thought. Around Christmas time I poppedC into the local PC World computer store and while browsing I spotted > the wirless networking equipment - this equipment seemed quiteF expensive at the time (this could have hust been PC World prices whichD arent renowned to be cheap here in the UK) anyway I thought about itF again so I had a look at http://www.dabs.com (who use VMS BTW and wereF on that Alpha-IA64 assurance program video that someone posted a whileF back) and the prices seemed much cheaper. This led me to think that ifE wireless networking starts to take off (bit slow here in the UK) thenmD pretty soon maybe people will begin to become less attached to thereF PC's in the home and will start to use wireless devices - laptops PDAsD etc and this could slow down the market for PC's even more. Also allE this wireless activity will require the need for some pretty powerful-F servers the sort that the Alpha processor is ideally suited. Maybe SunF and IBM can already see this happening and they probably can't beleiveA there luck at the alphacide. Maybe Compaq shareholders need to be0C aware of this and vote against the merger. Then compaq can do a SGImF have a management clearout bring back the Alpha and concentrate on HPC" and mobile devices such as IPaqs. F  If you see my post in comp.arch I suspect that because of the lack ofE games that IA64 will have no chance against Hammer on the PC desktop. D No wonder SGI has to flog there Itaniums on Ebay there is no use for them except boat anchors ;-)   cheers,$    a Peter Watkinson> peter.watkinson1@ntlworld.coms   ------------------------------  # Date: Thu, 11 Apr 2002 10:58:58 GMT 5 From: peter.watkinson1@ntlworld.com (Peter Watkinson)pD Subject: Re: Merger problems was Re: WIreless networking + VMS/Tru641 Message-ID: <3cb56b0e.3670218@news.cable.ntl.com>i  F On Thu, 11 Apr 2002 10:41:23 GMT, peter.watkinson1@ntlworld.com (Peter Watkinson) wrote:t  E >On Thu, 11 Apr 2002 06:58:06 +0000 (UTC), mustang@ucc.asn.au.invalide >wrote:  > , >>Main, Kerry <Kerry.Main@compaq.com> wrote:
 >>: Peter, >>K >>: I suspect what you want is to access VMS applications via WAP devices -M >>: is this correct? >>G >>I suspect Peter wants to participate in a wireless LAN with an Alpha.CH >>Peter, I recommend that you either check out a Cisco Workgroup bridge,D >>or one of these: http://www.enterasys.com/products/items/CSISA-AX/ >> >>Cheers >>/dave  >>-- e' >>" I don't get mad.... I get stabby. ":  >>- William "Fat Tony" Williams. >6    F I might add that I beleive that this merger is based on pure greed. NoC one in their right mind is going to rely on 64bit windows for there A high reliable systems. Pretty soon when The New HP starts posting C large losses on their PC's loses Large contracts for Servers to IBMnD and Sun and shrivels to a smaller size then there management will be given the chop :-)   cheers,.         Peter Watkinson  peter.watkinson1@ntlworld.com    ------------------------------  % Date: Thu, 11 Apr 2002 08:20:11 -0400 # From: "Tom Cole" <tom.cole@sas.com>  Subject: Re: mice infestation / Message-ID: <a93v3m$b9l$1@license1.unx.sas.com>d  . "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJIGENHELAA.tom@kednos.com...i7 > I have too many mice and Keyboards at my work area...    <BANDWIDTH WASTE="100%">  C KVM's didn't work for us either, we got a number of Cable-Augmentedu; Terminators, which were helpful in getting rid of the mice.a   </BANDWIDTH> -- Tom Cole, tom.cole@sas.com/ "Linux is only free if your time is worthless."    ------------------------------  % Date: Thu, 11 Apr 2002 09:53:28 -0400m5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>e Subject: Re: mice infestationk, Message-ID: <a944s1$bv44$1@lead.zk3.dec.com>  C No.  You can't mix the VS4000 & DEC 3000 mice 78 keyboards with therF PS2-style mice and keyboard on the newer systems.  They are completely* different electrically, and protocol wise.     Tom Linden wrote in message ... 8 >I have too many mice and Keyboards at my work area, and8 >KVM switches are for PC's I have figured out how to get4 >RGB output into VGA, but the mice and keyboards are; >different, well the older ones, VAX 4000/90, and 3000/M300r; >two of each.  is there an adapter to use a PC sttyle mouse = >with  these.  The keyboards have RJ style connectors, in thed2 >chassis on the VAXs and in a cable on the alphas. >e >Appreciate input.   ------------------------------  % Date: Thu, 11 Apr 2002 08:14:26 -0700i# From: "Tom Linden" <tom@kednos.com>0 Subject: RE: mice infestationi9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEPBELAA.tom@kednos.com>d  @ I wasn't planning to mix them, I was going to "adapt the cables"A so as to use the KVM as a "bent pipe" for presumably a compatible  set of mice and keyboards.   > -----Original Message-----< > From: Fred Kleinsorge [mailto:kleinsorge@star.zko.dec.com]( > Sent: Thursday, April 11, 2002 6:53 AM > To: Info-VAX@Mvb.Saic.Comd > Subject: Re: mice infestationa >  > E > No.  You can't mix the VS4000 & DEC 3000 mice 78 keyboards with the H > PS2-style mice and keyboard on the newer systems.  They are completely, > different electrically, and protocol wise. >  > ! > Tom Linden wrote in message ...c: > >I have too many mice and Keyboards at my work area, and: > >KVM switches are for PC's I have figured out how to get6 > >RGB output into VGA, but the mice and keyboards are= > >different, well the older ones, VAX 4000/90, and 3000/M300r= > >two of each.  is there an adapter to use a PC sttyle mousee? > >with  these.  The keyboards have RJ style connectors, in the 4 > >chassis on the VAXs and in a cable on the alphas. > >  > >Appreciate input. >  >    ------------------------------  # Date: Thu, 11 Apr 2002 11:53:17 GMT # From: "John Smith" <a@nonymous.com>e' Subject: Re: More positive OpenVMS newseD Message-ID: <NOet8.327$nA1.286@news02.bloor.is.net.cable.rogers.com>  > Nobody can scan it because it's written with disappearing ink.    < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3CB4E232.21A48695@fsi.net...e > Phillip Helbig wrote:c > >tK > > > well, I received today a letter from Carly ... it is available from QnJ > > > for customers like us who want to purchase alpha/vms systems and areH > > > unsure what will happen to vms after the merger ... she states herJ > > > support for vms in the letter ... now do you think she would want toJ > > > tell them one thing then pull the carpet out from under them ... notL > > > very good p.r. move! I was put on to this letter by someone high up in Q4F > > > ... it is for real ... now either Carly is lying to customers in writinge# > > > or she isn't ... which is it?- > >oL > > I hope to hear something interesting at the DECUS symposium in Bonn next	 > > week.< > >-A > > Can you scan the letter and put the resulting jpg on the web?h >tH > ...or scan it and run some OCR software against the scan, clean it up, > then post the results? >mJ > ...or does someone have the text in machine-readable (non-graphic) form? >n > -- > David J. Dachterao > dba DJE Systemsi > http://www.djesys.com/ >h* > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/n   ------------------------------  # Date: Thu, 11 Apr 2002 11:55:12 GMTr# From: "John Smith" <a@nonymous.com>t' Subject: Re: More positive OpenVMS news'D Message-ID: <AQet8.339$nA1.158@news02.bloor.is.net.cable.rogers.com>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3CB4E3AC.D2DE6A7E@fsi.net...h > *TOTALLY DISGUSTED SIGH* > F > See, that's the part I don't get. Here is a piece of news that couldJ > alone could be used to double VMS's market penetration "overnight" (it'sI > quotes, o.k.? No flames, please). So what do they do? They make it NDA.i > G > WTF, man???!!! Did *ANY* of these people stay awake in their businessw' > classes, assuming they took any???!!!d >g > > Will provide info when ithL > > goes public. Perhaps someone already has a public document and can share it > > in this forum. >rI > I think the cat's outta the bag, y'all. Time for Carly, Curly & Crew tog
 > come clean.     F IBM could learn a thing or two from Carly & Curly about spreading FUD. They're masters at it.   ------------------------------    Date: 11 Apr 2002 07:53:16 -0500- From: koehler@encompasserve.org (Bob Koehler)t' Subject: Re: More positive OpenVMS newsr3 Message-ID: <Dz72uM4TPEhc@eisner.encompasserve.org>   l In article <a91rnj$8da$1@fizban.fizban.pprd.abbott.com>, "Dave Gudewicz" <david.gudewicz@abbott.com> writes:H > What's a VERY PUBLIC commitment?  Something on the Goodyear blimp that@ > travels around the country, both here and abroad?  Billboards?. > TV/Radio/Newspaper/Time magazine, etc stuff?      Could be a good start.e  M > And if John Q. Public saw any of this VERY PUBLIC stuff, what exactly wouldlJ > it mean to her/him?  My guess, absolutely nothing.  And lots more wasted > money.  H    At first, yes.  But there's a need to get the message to John Q. CIO.   ------------------------------    Date: 11 Apr 2002 07:55:05 -0500- From: koehler@encompasserve.org (Bob Koehler)m' Subject: Re: More positive OpenVMS newss3 Message-ID: <qULBvaSAbmuf@eisner.encompasserve.org>s  f In article <3CB49EAD.334730E3@firstdbasource.com>, Michael Austin <maustin@firstdbasource.com> writes: > F > I have it on good authority that they would bury 15-20 PDP11/73's inD > concentric rings around a nuclear war head and it was "amazing theG > amount of data that could be gathered before they melted down" during 7 > the time when they were doing underground testing....( >       Talk about fault tolerant!a   ------------------------------    Date: 11 Apr 2002 08:03:21 -0500- From: koehler@encompasserve.org (Bob Koehler)-' Subject: Re: More positive OpenVMS news73 Message-ID: <TWEUFEh45dNq@eisner.encompasserve.org>e  n In article <Lq0t8.19378$%s3.6545440@typhoon.ne.ipsvc.net>, "Terry C. Shannon" <terryshannon@attbi.com> writes: > 5 > "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messager   >>= >> That must be the most stupid thing I'v read anyware, ever.n7 >> Please, please, please, keep that out of this list !i >> > I > Ah, a wee bit of a problem with freedom of speech, eh? Quel dommage. GotK > ahead and killfile me if I've offended your sensibilities. But be advisedfN > that I am an American citizen who saw 3K of his compadres killed by fanaticsI > on September 11. I also have a wee bit of a problem with the Butcher ofp
 > Baghdad. >   @    While I am please to see the damage done to Al-Qeda and theirD    Taliban cronies, and I too have lots of problems with the fellow K    in Baghdad, I do not have so many problems with the people of Baghdad.  o?    I would hate to see them become victims of Sadam's misdeeds.>  F    But Sadam may be capable of the kinds of misdeeds that I would feel:    justified the use of nukes.  We all just hope he won't.  H    And I'm not about to pull my punches, politics, or free speech on the%    forum that comp.os.vms has become.    ------------------------------  % Date: Thu, 11 Apr 2002 14:31:13 +01000% From: Alan Greig <a.greig@virgin.net>h' Subject: Re: More positive OpenVMS newsC8 Message-ID: <u04bbu8lvsrs9v7doeh2ec7j1jjbauaqf9@4ax.com>  A On 10 Apr 2002 06:15:26 -0700, bob@instantwhip.com (Bob Ceculski)r wrote:  b >JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3CB3866E.B19DA1BF@videotron.ca>... >> Bob Ceculski wrote:J >> > well, I received today a letter from Carly ... it is available from QI >> > for customers like us who want to purchase alpha/vms systems and are-O >> > unsure what will happen to vms after the merger ... she states her support, >> > for vms in the letter ... >> l& >> What is the wording used by Carly ? >> MJ >> Is it something like "support our existing customers", or "maximise the( >> potential of this operating system" ? >sH >it is a lot more than that ... I promised this high up vms contact thatE >I would not release all the details as he stated this was being heldkD >back only because of the merger proceedings, but soon you will see!G >Maybe Terry have privilidge to this letter and he can report on it ...i  > If someone want's to send me a copy anonymously I'm sure I can reliably leak it...p   ------------------------------  % Date: Thu, 11 Apr 2002 10:04:17 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>n' Subject: Re: More positive OpenVMS newso, Message-ID: <a945gb$ct2j$1@lead.zk3.dec.com>  A David J. Dachtera wrote in message <3CB4E3AC.D2DE6A7E@fsi.net>...z >"Terry C. Shannon" wrote: >>I >> "Phillip Helbig" <HELBPHI@sysdev.deutsche-boerse.com> wrote in messagei8 >> news:01KGEEGBNA809EEGPR@sysdev.deutsche-boerse.com...D >> > > Damn the doomsayers. Stay tuned for a letter from Carly F re:	 continuedd >> > > support for OpenVMS.i >> > >> > Full text or URL, please. >>4 >> As soon as I get same (it's still NDA RIGHT NOW). >  >*TOTALLY DISGUSTED SIGH*i >dE >See, that's the part I don't get. Here is a piece of news that could I >alone could be used to double VMS's market penetration "overnight" (it'siH >quotes, o.k.? No flames, please). So what do they do? They make it NDA. >tF >WTF, man???!!! Did *ANY* of these people stay awake in their business& >classes, assuming they took any???!!! >' >> Will provide info when itK >> goes public. Perhaps someone already has a public document and can sharen it >> in this forum./ >pH >I think the cat's outta the bag, y'all. Time for Carly, Curly & Crew to >come clean. >e  F While I can neither confirm or deny the existance of said letter, I amJ reliably told that public issuance of such a statement generally cannot beI made by the head of HP, until she is also the head of Compaq.  Should youwK require additional assurances beyond those that are public from the currentnI head of Compaq - to allow you to make informed purchasing decisions - you.+ should contact your account representative.   J So, I would not expect any public statement until after the merger vote isB certified, and the lawsuit is heard by the court later this month.   ------------------------------  # Date: Thu, 11 Apr 2002 14:41:10 GMTh# From: "John Smith" <a@nonymous.com>y' Subject: Re: More positive OpenVMS newsaF Message-ID: <aght8.1592$nA1.1479@news02.bloor.is.net.cable.rogers.com>  J So Fred, don't you suppose that the identical content could be made public7 TODAY by a CEO of a company that's still called Compaq?t  ( Or doesn't he know how to say 'OpenVMS'?  K If HP and Compaq are still 'competitors' in the marketplace, up to the time-H a merger is legally ratified, wouldn't it be to Compaq's benefit to make such a pronouncement TODAY?h  J Or is this alleged letter just more 'bovine scatology', as Stormin' Norman
 once said?          @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message& news:a945gb$ct2j$1@lead.zk3.dec.com...C > David J. Dachtera wrote in message <3CB4E3AC.D2DE6A7E@fsi.net>...  > >"Terry C. Shannon" wrote: > >>K > >> "Phillip Helbig" <HELBPHI@sysdev.deutsche-boerse.com> wrote in messagee: > >> news:01KGEEGBNA809EEGPR@sysdev.deutsche-boerse.com...F > >> > > Damn the doomsayers. Stay tuned for a letter from Carly F re: > continuedi > >> > > support for OpenVMS.  > >> >  > >> > Full text or URL, please. > >>6 > >> As soon as I get same (it's still NDA RIGHT NOW). > >P > >*TOTALLY DISGUSTED SIGH*r > >rG > >See, that's the part I don't get. Here is a piece of news that could5K > >alone could be used to double VMS's market penetration "overnight" (it's@J > >quotes, o.k.? No flames, please). So what do they do? They make it NDA. > >tH > >WTF, man???!!! Did *ANY* of these people stay awake in their business( > >classes, assuming they took any???!!! > >t > >> Will provide info when itG > >> goes public. Perhaps someone already has a public document and canu sharec > it > >> in this forum.n > >tJ > >I think the cat's outta the bag, y'all. Time for Carly, Curly & Crew to > >come clean. > >g >mH > While I can neither confirm or deny the existance of said letter, I amL > reliably told that public issuance of such a statement generally cannot beK > made by the head of HP, until she is also the head of Compaq.  Should youoE > require additional assurances beyond those that are public from theo currentgK > head of Compaq - to allow you to make informed purchasing decisions - youd- > should contact your account representative.i >eL > So, I would not expect any public statement until after the merger vote isD > certified, and the lawsuit is heard by the court later this month. >U >a >a >    ------------------------------    Date: 11 Apr 2002 15:06:06 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]>t' Subject: Re: More positive OpenVMS news 5 Message-ID: <20020411150606.4149.qmail@gacracker.org>f  ; On Thu, 11 Apr 2002, Alan Greig <a.greig@virgin.net> wrote:UB >On 10 Apr 2002 06:15:26 -0700, bob@instantwhip.com (Bob Ceculski) >wrote:e   <snip>  I >>it is a lot more than that ... I promised this high up vms contact thatcF >>I would not release all the details as he stated this was being heldE >>back only because of the merger proceedings, but soon you will see!tH >>Maybe Terry have privilidge to this letter and he can report on it ... >e? >If someone want's to send me a copy anonymously I'm sure I canh >reliably leak it...  $ That would be more my department ;-)  I Might I suggest anyone with the letter check out the following website, I C can assure you that use of the provided interface will ensure it isc( absolutely impossible to trace the leak.    0 http://xenophon.r0x.net/cgi-bin/mixnews-user.cgi     Doc. -- l6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                              http://vmsbox.cjb.net    ------------------------------  % Date: Thu, 11 Apr 2002 11:48:48 -0400o( From: David Froble <davef@tsoft-inc.com>' Subject: Re: More positive OpenVMS news , Message-ID: <3CB5B060.1090308@tsoft-inc.com>   Paul Repacholi wrote:r  5 > "Terry C. Shannon" <terryshannon@attbi.com> writes:s >  > B >> I also have a wee bit of a problem with the Butcher of Baghdad. >> >  > And the Butcher of Beiruit?   7 I'd be nice if this topic had never come up, ... Terry.0  M At least we're quickly decreasing the violence level, lets see, first it was oC W-88s, then the gas man, and now we're down to plinking with F-16s.c  H Maybe it's genetic, and when we find the gene that promotes thoughts of 1 indiscriminate violence, it should be eradicated.t   ------------------------------  % Date: Thu, 11 Apr 2002 17:12:19 +0100o% From: Alan Greig <a.greig@virgin.net>w' Subject: Re: More positive OpenVMS news 8 Message-ID: <lpcbbu8npccsf563ebaaovlhq1rbqverf1@4ax.com>  ) On 11 Apr 2002 15:06:06 -0000, Doc.Cypherm3 <Use-Author-Supplied-Address-Header@[127.1]> wrote:e    @ >>If someone want's to send me a copy anonymously I'm sure I can >>reliably leak it...h > % >That would be more my department ;-)e  J >Might I suggest anyone with the letter check out the following website, ID >can assure you that use of the provided interface will ensure it is) >absolutely impossible to trace the leak.   F Hmm, I think I should maybe check my work email in case I have a copy.D Not that I can send you it if I have it officially :-) UnfortunatelyC been down with some nasty flu type virus the last week or so. Which 8 also means I'm missing the London VMS days this week :-(  F Apparently a temp of 101.7F (38.7C) is a good sign my doctor tells me!   >@1 >http://xenophon.r0x.net/cgi-bin/mixnews-user.cgis >  >w >Doc.a   ------------------------------  % Date: Thu, 11 Apr 2002 16:40:14 +0529 3 From: "padmavati_1@vsnl.net" <padmavati_1@vsnl.net> @ Subject: NEW Import Department - for DEPB  DFRC  SIL & DUTY FREE8 Message-ID: <20020411111118.06A3B1216C@mmb4.vsnl.net.in>   <html>   <head>J <meta http-equiv="Content-Type" content="text/html; charset=windows-1252">9 <meta name="GENERATOR" content="Microsoft FrontPage 4.0">:8 <meta name="ProgId" content="FrontPage.Editor.Document">( <title>Duty Free Import Licences</title> </head>m   <body bgcolor="#FFCCFF">   <table border="1" width="100%">n   <tr>     <td>   <h1 align="center" style="text-align:center"><font size="7" color="#000080"><span style="background-color: #FFCCFF">Padmavati&nbsp; S Enterprise<span style="mso-bidi-font-size: 12.0pt; background-color: #FFCCFF"><o:p>o &nbsp; </span></span></font></h1>	     </td>    </tr>t </table> <p class="MsoBodyText"><b><span style="font-size:13.0pt;mso-bidi-font-size:12.0pt"><font color="#0000FF">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ? </font><font color="#808000"> </font><font color="#660033">Duty 6 Free Import Licences / D.E.P.B / Drawback/DFRC / <o:p> </o:p> </font></span></b></p> <p class="MsoBodyText"><font color="#660033"><b><span style="font-size:13.0pt;mso-bidi-font-size:12.0pt">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; . Clearing<span style="mso-spacerun: yes">&nbsp;1 </span>&amp; Forwarding<span style="mso-spacerun: V yes">&nbsp; </span>Agents.</span></b><span style="font-size:13.0pt;mso-bidi-font-size:
 12.0pt"><o:p>4 </o:p>$ </span></font><font color="#0000FF">P <span style="mso-ignore:vglayout;position:relative;z-index:0;left:-25px;top:6px; width:735px;height:9px"><img src="../../../../My%20Documents/My%20Webs/NEW%20im2.gif" v:shapes="_x0000_s1026" width="735" height="3"></span>&nbsp;<o:p>e </o:p> </font></p>  <p class="MsoNormal"><font color="#FFFFFF"><span style="font-size: 11.0pt; mso-bidi-font-size: 12.0pt; background-color: black; background-repeat: repeat; background-attachment: scroll; mso-highlight: black; background-position: 0%"><b>1,O Humera Arcade, Ground floor, 2/4, Israel Mohalla, Samuel Street, Masjid Bunder,px Mumbai  400009</b></span></font><span style="font-size: 11.0pt; mso-bidi-font-size: 12.0pt"><font color="#FFFFFF"><o:p>  </font><font color="#0000FF"> </o:p> </font>  </span></p>V <h6 align="left" style="text-align:left"><span style="mso-spacerun: yes; font-size: 10.0pt; mso-bidi-font-size: 12.0pt"><font color="#0000FF">&nbsp;</font></span><font color="#0000FF"><span style="font-size:10.0pt;J mso-bidi-font-size:12.0pt">Tel. Office: 91-22-3710825 / 3781915 / 3702102.S TeleFax. : 91-22-3710825/2080404<span style="mso-spacerun: yes">&nbsp; </span>Res.: N 91-22-4142643/4144112<span style="mso-spacerun: yes">&nbsp;&nbsp; </span><o:p> </o:p> </span></font></h6> ] <h6 align="left" style="text-align:left"><font color="#0000FF"><span style="font-size:10.0pt; P mso-bidi-font-size:12.0pt"><span style="mso-spacerun: yes">&nbsp;</span>Mobile : 9820292949 / 9820049216<span style="mso-spacerun: yes">&nbsp; </span></span><span style="font-size:11.0pt;mso-bidi-font-size:12.0pt"><span style="mso-spacerun:  yes">&nbsp;&nbsp;</span>Email: <a href="mailto:girishblodaya@hotmail.com">girishblodaya@hotmail.com</a><span style="mso-spacerun: yes">&nbsp; L </span>/ <a href="mailto:padmavati_1@vsnl.net">padmavati_1@vsnl.net</a><o:p> </o:p> </span></font></h6>gl <p class="MsoNormal"><span style="font-size:14.0pt;mso-bidi-font-size:12.0pt"><font color="#0000FF">To,<o:p> </o:p> </font></span></p>R <h4><font color="#0000FF">The Director (Import / Purchase Dept.)&nbsp;</font></h4>K <p class="MsoNormal"><font color="#0000FF">Dear Sir/Madam,&nbsp;</font></p> O <p class="MsoNormal"><font color="#0000FF">&nbsp;Sub.&nbsp;<strong><u>Duty freehG pass book (DEPB) Special Import Licence Duty Free Advance Licences<o:p>s </o:p> </u></strong></font></p>M <p class="MsoNormal"><strong><font color="#0000FF">&nbsp;</font></strong></p>rL <p class="MsoNormal"><font color="#0000FF">Please contact us for:</font></p>W <p class="MsoNormal"><strong><u><font color="#0000FF">D E P B :</font></u></strong></p>lM <p class="MsoNormal"><font color="#0000FF">We offer DEPB licences at the best M rates for DEBIT against your various imports. As per the current customs duty!P structure the <strong>Nett Savings is 4 to 5 % </strong>on the Nett payable dutyA if&nbsp; debited against duty free pass book (D.E.P.B)</font></p>iX <h2><font color="#0000FF">Special Import Licence / Duty Free Advance Licence</font></h2>F <p class="MsoNormal"><font color="#0000FF">We also offer S.I.L at mostN reasonable premiums and also can offer specific licences for duty free imports$ (Against Advance Licence)</font></p>b <p class="MsoNormal"><font color="#0000FF">&nbsp;<strong><u>Other Services</u></strong></font></p>Q <p class="MsoNormal"><font color="#0000FF">&nbsp;We also provide <strong>Clearing4J Services </strong>for Imports Cargo at Nava Sheva, Mumbai &amp; Mumbai Air Cargo.</font></p>a_ <p class="MsoNormal"><font color="#0000FF">&nbsp;<strong><u>Liason Work</u></strong></font></p>nZ <p class="MsoNormal" style="mso-layout-grid-align:none"><font color="#0000FF">&nbsp;Liason< work for all D.G.F.T and Customs Related Matters.</font></p>v <p class="MsoNormal" style="mso-layout-grid-align:none"><font color="#0000FF">&nbsp;<span style="text-decoration:none;text-underline:none">Thanking you,<span style="mso-tab-count:5">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</span></span></font></p>.Y <h2 style="mso-layout-grid-align:none"><font color="#0000FF">Yours Sincerely,</font></h2> I <h6><b><font size="3" color="#0000FF">Girish Lodaya&nbsp;</font></b></h6>t<h6><span style="mso-tab-count:5">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</span></h6><h6><span style="mso-tab-count:5">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span>&nbsp;</h6>   </body>B   </html>e   ------------------------------  % Date: Thu, 11 Apr 2002 18:03:46 +0200a+ From: Knoll <knoll@infotech.tu-chemnitz.de>b( Subject: northernlight.com response time5 Message-ID: <3CB5B3E2.E6A5A6@infotech.tu-chemnitz.de>n  E It is known that Northern Light no longer provides free Web search tom the general public. D The Special Collection continues to be available for free searching, though.I  G However, access to http://www.northernlight.com has become so slow thatm the D practical value of northernlight has decreased dramatically for me. A I do not know whether this is only true for requests comming fromt Europe.   E Anyway, it is sad to see an excellent search engine gradually passinge away.    Thomas --* http://www.infotech.tu-chemnitz.de/~knoll/   ------------------------------    Date: 11 Apr 2002 04:29:32 -0700- From: adrian_ogden@hotmail.com (Scumbag Adie)b+ Subject: Polycenter Agents for Windows 20001= Message-ID: <91445c94.0204110329.3356213d@posting.google.com>    Hi,   @ Is it possible/what do I need to be able to monitor Windows 2000< machines from a Vaxstation running SEVMS 6.1 with PolycenterD Datacollector 2.2, Polycenter Performance Advisor 2.2 and Polycenter  Network Manager 400 Version 2.4?  ( I know, what a daft thing to want to do!   Ta   A    ------------------------------  % Date: Thu, 11 Apr 2002 14:49:52 +0100fT From: Andrew Harrison SUNUK Consultancy <andrew_nospam.remove_this.harrison@sun.com>9 Subject: Re: Poor Ebay ... should have got a VMS cluster!i& Message-ID: <3CB59480.9050907@sun.com>   Bob Ceculski wrote:l  m > Michael Austin <maustin@firstdbasource.com> wrote in message news:<3CB44A7C.A4C44F5D@firstdbasource.com>...  >  >>jlsue wrote: >>C >>>On 9 Apr 2002 18:01:40 -0700, bob@instantwhip.com (Bob Ceculski)M	 >>>wrote:h >>>r >>>d7 >>>>this is were unix gets you!  from computerworld ...o >>>>( >>>>EBay has three outages in three days >>>> >>>>By BRIAN SULLIVAN  >>>>(April 09, 2002)F >>>>Online auction giant eBay has experienced three outages of varyingE >>>>severity since Sunday, a spokesman for the San Jose-based companyS	 >>>>said.s >>>>	 >>>>[...]t >>>>D >>>Um, be careful here.  The front-end servers are Microsoft serversI >>>(according to Netcraft) and could be the source of the problem.  I didPF >>>not note enough info in this article to draw a conclusion as to theG >>>source of their data corruption.  It could be that some file on a MSdG >>>server was corrupted and it was a replicated file that was then sent  >>>to other front servers. >>>sE >>>I applaud your desire to pursue the shortcomings of other OSs, butAG >>>let's not stoop to the level of some of our competitors in doing so.e >>> 4 >>>Not speaking for anyone, certainly not DEC/Compaq0 >>>(get rid of the xxxz in my address to e-mail) >>>NG >>Okay... this is where NT and Sun will get you... should have bought a?# >>True Cluster. (or TruCluster :) )a >> > M > I am simply pointing out that ebay had a chance to go w/alpha vms, and theywH > didn't ... this would have never happened on a pure vms web server ...G > and I know they have nt frond end, but this sounds like a backend db  & > problem or a hack, and that is unix! >     C Really the last big eBay incident sounded like a Sun server failure D if you were an OpenVMS choir member, turned out to be an application+ failure and an unpatched 3rd party product.n  A Could we conclude that Yahoos recent outages were caused by Alphaa@ server failure. No we cannot anymore than you can attribute this< failure to DBMS/HW/UNIX or anything. If you are going to try3 throwing FUD then you need to make it less obvious.    Regards  Andrew Harrisonb   ------------------------------  # Date: Thu, 11 Apr 2002 16:08:36 GMTi% From: jlsue <jlsuexxxz@insightbb.com> 9 Subject: Re: Poor Ebay ... should have got a VMS cluster!18 Message-ID: <28dbbu0qobooudlmp066fojg69f6ic55bj@4ax.com>  E On Thu, 11 Apr 2002 14:49:52 +0100, Andrew Harrison SUNUK Consultancy03 <andrew_nospam.remove_this.harrison@sun.com> wrote:u   >oD >Really the last big eBay incident sounded like a Sun server failureE >if you were an OpenVMS choir member, turned out to be an application , >failure and an unpatched 3rd party product. >tB >Could we conclude that Yahoos recent outages were caused by AlphaA >server failure. No we cannot anymore than you can attribute thist= >failure to DBMS/HW/UNIX or anything. If you are going to tryg4 >throwing FUD then you need to make it less obvious.    F Funny how you're capable of such restraint when in the defensive mode.: T'would be nice to see such analysis the rest of the time.  1 Not speaking for anyone, certainly not DEC/Compaq - (get rid of the xxxz in my address to e-mail)s   ------------------------------    Date: 11 Apr 2002 00:44:40 -0700* From: Dowe Keller <dowe@krikkit.127.0.0.1>2 Subject: Re: Predictions - just for the hell of it. Message-ID: <m33cy23cie.fsf@krikkit.127.0.0.1>  / koehler@encompasserve.org (Bob Koehler) writes:   J >    Ejecting by s/w control is a good idea on Mac because the OS tends toL >    cache read and write data.  Ejecting by hardware is good enough on PCs 6 >    because the OS does not cache writes to floppies.  ( Which OS, I know of quite a few PC OSes:  B Minix, Linux, Zenix, Hurd, FreeBSD, NetBSD, OpenBSD, CP/M, M$-DOS,K Windows, FreeDOS, BSDi, Solaris, OS/2, BeOS... and I probably forgot a few.i   --  4 dowe@sierratel.com			<http://www.sierratel.com/dowe>E "MSDOS didn't get as bad as it is overnight -- it took over ten years  of careful development." (By dmeggins@aix1.uottawa.ca)m   ------------------------------    Date: 11 Apr 2002 08:10:27 -0500- From: koehler@encompasserve.org (Bob Koehler)l2 Subject: Re: Predictions - just for the hell of it3 Message-ID: <l4WzR8CNwKO4@eisner.encompasserve.org>o  \ In article <3CB4AFD6.93BEF206@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > O > Considering that personal computers were able to format diskettes from aroundoO > 1984 when the 3.5" format became popular, and were able to format the true 5"iP > floppies , and prior to that, word processors and bigger machines were able toM > do 8" floppies, I find it very hard to believe that they could not reliablytC > format a diskette when they designed the digital diskette drives.   G    Interesting.  I never had to format an 8 inch floppy.  They were alloA    VAX console media, or VT78 media.  They came from DEC properly:!    formatted and stayed that way.g  D    I think this was the model DEC was following when they didn't seeG    a need to provide formatting capability on the Pro series.  Then the ;    IBM PC took off and everybody was formatting everything.t   ------------------------------    Date: 11 Apr 2002 08:13:02 -0500- From: koehler@encompasserve.org (Bob Koehler)t2 Subject: Re: Predictions - just for the hell of it3 Message-ID: <bRrIG0C1dISd@eisner.encompasserve.org>   [ In article <m33cy23cie.fsf@krikkit.127.0.0.1>, Dowe Keller <dowe@krikkit.127.0.0.1> writes:d1 > koehler@encompasserve.org (Bob Koehler) writes:e > K >>    Ejecting by s/w control is a good idea on Mac because the OS tends tohM >>    cache read and write data.  Ejecting by hardware is good enough on PCs o7 >>    because the OS does not cache writes to floppies.  > * > Which OS, I know of quite a few PC OSes:  G    Yeah, but the stupid things were designed to support MS-DOS.  If you A    use a real OS you just have to be carefull about those things.    ------------------------------    Date: 11 Apr 2002 10:05:51 -0400. From: Joel Gallun <root@localhost.localdomain>2 Subject: Re: Predictions - just for the hell of it3 Message-ID: <ygabscqba9s.fsf@severn.office.aol.com>s  . Paul Repacholi <prep@prep.synonet.com> writes:  C > > I've seen claims that DEC was one of the first, if not the verya) > > first, commercial entity on the net. a >a2 > DEC and BBN I think. Possibly Honeywell as well.  $ BBN was before DEC -- think about it  ' > Who where the early left coast .coms?s   SRI was one of them    Joel   ------------------------------   Date: 11 Apr 2002 07:28:33 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) ; Subject: Re: Problems writing blank tape in TZ86 tape drivee0 Message-ID: <a93dv1$242$1@n.ruf.uni-freiburg.de>  j In article <3cb49361$0$23948$7402020d@newsfeed.sunrise.ch>, "Markus Mller" <mamueller@dplanet.ch> writes:I >I am trying to initialize a blank tape (totally blank / factory new DLT3fL >cartrige) using the INITIALIZE command, but I always keep getting a "PARITY >ERROR" error message. >oL >Analysing the hardware log I found the following message (only SCSI details >shown below): >h >----------o  >       SCSI CMD        00000008  >                           0050, >                                       READ  >       SCSI STATUS           027 >                                       CHECK CONDITION  >c > EXTENDED SENSE DATAt >k  >       EXTENDED SENSE  000300F0  >                       BB500000  >                       00000000  >                       00000080  >                         3700004 >                                       MEDIUM ERROR: >                                       UNKNOWN ERROR TYPE >----------s >gG >I also tried INITIALIZE/ERASE, but this results in the command to hangeL >indefinitively and to become unkillable while the tape does not perform any >operation.: >1A >What is going wrong? Is the tape drive broken? Do I need specialj: >pre-formatted tapes? Or special tape formatting software?  E You shouldn't need special formatting software and you shouldn't needhM pre-formatted tapes either. There is obviously something wrong with the drive>L or the tape. Thus, do you have other tapes (even used ones) to test? Did youK use the drive in the past or is it the first-time that you are going to usei% it? You know about cleaning tapes?...F   Regards,    Christoph Gartmann"  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, Germany                                           |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Thu, 11 Apr 2002 11:40:31 +0200b2 From: "Ren Schelbaum" <rene.schelbaum@datakom.at>; Subject: Re: Problems writing blank tape in TZ86 tape drive G Message-ID: <3cb559fc$0$13718$5039e797@newsreader01.highway.telekom.at>>  = "Markus Mller" <mamueller@dplanet.ch> schrieb im Newsbeitragn5 news:3cb49361$0$23948$7402020d@newsfeed.sunrise.ch...bJ > I am trying to initialize a blank tape (totally blank / factory new DLT3E > cartrige) using the INITIALIZE command, but I always keep getting am "PARITYu > ERROR" error message.m >FE > Analysing the hardware log I found the following message (only SCSI< detailss > shown below):3 >t > ----------! >        SCSI CMD        00000008p! >                            0050s- >                                        READe! >        SCSI STATUS           02t8 >                                        CHECK CONDITION >0 >  EXTENDED SENSE DATA >s! >        EXTENDED SENSE  000300F0"! >                        BB500000o! >                        00000000t! >                        00000080s! >                          370000.5 >                                        MEDIUM ERROR;; >                                        UNKNOWN ERROR TYPEu > ---------- ><H > I also tried INITIALIZE/ERASE, but this results in the command to hangI > indefinitively and to become unkillable while the tape does not performp anyo > operation. >1B > What is going wrong? Is the tape drive broken? Do I need special; > pre-formatted tapes? Or special tape formatting software?a >n > Configuration:% > VAX 3100-95 running unter VMS 5.5-2  > Tape Drive TZ86 on SCSI busn >o > Markus Mllers >/ >t   Hello!  I I am not sure if that isn't normal operation. I remember having found out"J that the drive tries to read before it starts writing, and that causes the parity errors on a blank tape:D (At least thats the way it was on reel tapes). There used to be someB switches to the INIT-Command that prevented it from reading before
 initializing."   Ren   ------------------------------    Date: 11 Apr 2002 05:29:21 -0700- From: dima_bessonov@mail.ru (Dmitry Bessonov)p8 Subject: Q: BROTHER Laser via DCPS+LAT - Any Experience?< Message-ID: <f6293a9.0204110429.423ae442@posting.google.com>  E Did anyone try BROTHER laser printers via DCPS and LAT? I need to addnD several relatively inexpensive (~<$1000 each)  network lasers to ourF production system. I don't want to use TCP/IP because LAT is generallyE more robust and a lot easier to install and support. Brother seems toeC have a couple of sub-$1000 printers that list support of VMS, DCPS,sB and LAT: HL-1670N and the newer HL-1870N but they're not listed in? DCPS SPD, however. Does anyone have any experience, positive orFE negative, of working with these or earlier Brother models? I couldn'tsD find any other laser currently in production that would fit my bill.   Dmitry   ------------------------------    Date: 11 Apr 2002 06:37:36 -0700$ From: issinoho@slayme.com (issinoho)( Subject: Reboot after adding quorum disk= Message-ID: <d0141774.0204110537.675893c5@posting.google.com>l  3 Here's a dead easy one for all you cluster gurus...N  F I'm adding a HSD-served quorum disk to my dual AlphaServer cluster; inE short, do I HAVE to reboot the cluster after doing the CLUSTER_CONFIGgE tasks and updating modparams and AUTOGENning, or will the quorum diske TAKE immediately.e  A Answer appreciated is this is a 24/7 cluster that could really dodF without a reboot. The Cluster manual doesn't make it 100% clear either way.  4 More details on systems can be supplied if required.  	 Issinoho.t UK.i   ------------------------------    Date: 11 Apr 2002 09:11:49 -0700 From: hchen3@uwo.ca (Sammy)s1 Subject: Redirect SYS$OUTPUT for detached processI= Message-ID: <2c0966c2.0204110811.5a263775@posting.google.com>e   Hi:r  C   I am writing a C++ program which will utilize sys$creprc to spawnNF off a detached process that in turn, will execute either DCL script orB just regular Alpha image. (My system spec: Compaq C++ V6.3-020 for OpenVMS Alpha V7.2-1)r  D   The program will not be able to use lib$spawn since i want all theF jobs to keep executing in the detached process even if i terminate theC main process. In addition, user will be able to redirect sys$outputNC and sys$error of the detached process by providing the file name in; the configuration file.F  ' Here is the psuedo-code for my program:   .   open(newSysOutput.log) // i use C RTL open()     // use c RTL write()8   write timestamp and the image name (or script name) to newSysOutput.log     open(newSysError.err)t8   write timestamp and the image name (or script name) to newSysError.err;  6   sys$crembx() // make an input mailbox for sys$creprc  E   sys$qiow(write "define sys$output newSysOutput.log" to the mailbox);>   sys$qiow(write "open/append sys$out newSysOutput.log" to the mailbox)  E   sys$qiow(write "define sys$error newSysError.err" to the mailbox)  s?   sys$qiow(write "open/append sys$error newSysError.err" to then mailbox)  =   sys$qiow(write "write sys$output "stdout" " to the mailbox)3<   sys$qiow(write "write sys$error "stderr" " to the mailbox)  2   sys$qiow(write "close sys$error" to the mailbox)6   sys$qiow(write "deassign sys$output" to the mailbox))   sys$qiow(write "logout" to the mailbox);  6   sys$getdviw(translate the name of the mailbox......)  !   sys$creprc(image = loginout.exe;              input = mailbox               output = report.log              error = report.logn"              run as detached.....)  1   wait(2 seconds so stuff in mailbox can be read)n     exit();     .   after i execute the program.  I got 4 files.  3   Report.log;1 -> it shows login stuff and CPU times$                   to execute the job  D   newSysError.err;1 -> it shows the time stamp and the line "stderr"  2   newSysOutput.log;2 -> it shows the line "stdout"  0   newSysOutput.log;1 -> it shows the time stamp.  F   The result is what i want. However, i was expecting to have only oneC newSysOutput.log file which will contain BOTH the timestamp and thek line "stdout".  D   I try the entire procedure in a regular DCL terminal session againA and i notice that when i redefine sys$error, it will point to ther@ existing newSysError.err file while sys$output will create a newD version of newSysOutput.log. I would gladly provide more descriptionD or post the actualy code (about 200-300 lines if i simpifly it) if iC didn't describe the situation cleraly. Thanks for the all the help.s     Sammya   ------------------------------    Date: 11 Apr 2002 02:11:52 -0700 From: mrotgeri@web.de (ROM)y Subject: Re: remote mailboxn= Message-ID: <b3ad37dc.0204110111.6958edc5@posting.google.com>/  1 I think you don't understand my (simple) problem.-  D I want to write a C-program with one mailbox on node "B" and fill it up with data from node "A"  1 I create a mailbox and assign it to a net device:g" $DESCRIPTOR( logname, "MailBox" );" $DESCRIPTOR( net_device, "_NET:");         status = sys$crembx(	 		prmflg,o             	&netcmd_chan,	 		maxmsg,s 	       	bufquo, 	       	promsk, 	       	acmode, 	       	&logname );           status = sys$assign( 		&net_device,                	&netdcl_chan,                 	0,                	&logname);  / Now on "B" I can read from this mailbox with  :    	status = sys$qio( 		0, 		netcmd_chan, 		IO$_READVBLK,  		&iosb,	 		NULL,0,_ 		&szResult, 		sizeof(szResult),: 		0, 		NULL,s 		0,0);e No problem so far, but ...  D To write data to the mailbox from node "A" I need a channel (on node8 "A") to this mailbox to use the sys$qio() function (with IO$_WRITEBLK).> How can I get it, what are the things to do to get it working?   On node "A" I tried it with:+ 	$DESCRIPTOR( logname, "NODE_B::MailBox" );d. or	$DESCRIPTOR( logname, "NODE_B::MailBox:" ); in the sys$assign() function   	status = sys$assign(a 		0, 		&netdcl_chan2, 		0,
 		&logname );0& but I always get a SS$_IVDEVNAM error.  $ What's wrong in my code or thinking?E Is it necessary to build up two mailboxes on "A" and "B", assign bothS$ mailboxes to the net device or what?( Is there any example available (C-Code).   ROMe   ------------------------------   Date: 11 Apr 2002 04:58 CSTo' From: carl@gerg.tamu.edu (Carl Perkins)  Subject: Re: remote mailboxo- Message-ID: <11APR200204581372@gerg.tamu.edu>r   mrotgeri@web.de (ROM) writes... % }What's wrong in my code or thinking?pF }Is it necessary to build up two mailboxes on "A" and "B", assign both% }mailboxes to the net device or what?e) }Is there any example available (C-Code).u }  }ROM  C For a somewhat more complex example than you probably want of using H non-transparent DECnet, see SYS$EXAMPLES:DB_REQUESTER.C and DB_SERVER.C.   --- Carl   ------------------------------  % Date: Thu, 11 Apr 2002 08:05:50 +0200.# From: Paul Sture <p_sture@decus.ch>r= Subject: Re: Spring Forward, Fall Back - unfortunately not :- ( Message-ID: <3CB543DE.4948EDEF@decus.ch>  * Alan Winston - SSRL Admin Cmptg Mgr wrote: > c > In article <Gd0xDrrJLTUi@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:  > >In article <00A0C386.B74ADEFC@SSRL04.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes: > P > >> Hey, let's have a flame war over where Daylight Savings Time was originally > >> a good idea.e > >>O > >> I take the position that it was a good idea before reliable electric lighttO > >> was invented.  That extra hour of daylight in the evening didn't just mean T > >> that you got to play catch for an hour longer; it meant that you could function > >> for an hour longer. > >>L > >> I don't see why we should be doing it now.  Most places that don't haveO > >> electric lighting they tend to function more by the sun than by clock timee. > >> anyway, so DST doesn't do them much good. > >> > >nI > >       One of the better reasons is it is light earlier in the Winter,eJ > >       hence school kids are walking around in the morning in daylight. > L > Except that DST only runs in summer.  (Well, late spring to mid-fall.)  ItM > certainly isn't operating when days are shorter.  That doesn't seem like anc > argument for DST.c > G Hmm. Just to complicate things, in the UK, the clock change does have ac benefit in _winter_.  A Back in my schooldays (1970 ish ???) the UK experimented with noteE changing the clocks one winter. It was pretty miserable both going to H school and coming home in the dark. Traffic accident rate increases wereD quoted, and the building industry campaigned vigourously that it wasE costing them money (in the the latter case I failed to understand whyaG they didn't simply start and finish an hour later), so the clock change  was reintroduced.m   __
 Paul Sture Switzerland.   ------------------------------  % Date: Thu, 11 Apr 2002 15:47:43 +010035 From: "Robert A.M. van Lopik" <lopik@mail.telepac.pt>.( Subject: Start Windows program from VMS?" Message-ID: <3cb5a30f@news.wau.nl>  D I would appreciate your suggestions as to how to solve the following problem:  D I have a (binary) Windows program (say 'prog') that takes as input aF filename, processes it and produces an outpult file. Calling sequence: >prog infile outfile -switches  > I need to use this program within a VMS command procedure like $@doprog P1 P2 P3t  E Somehow I have to transfer P1 (the infile) to Windows(2000), initiateMJ execution of 'prog', and get the resulting file back in P2. And I also may; want to transmit the switches in P3 to the Windows program.-  J Before I dive too deep into this, I was wondering what the combined wisdom( of the VMS community would come up with.   Thanks in advanceo7 rob van lopik (DouroVision Consultoria Informtica Lda)a   ------------------------------  # Date: Thu, 11 Apr 2002 11:26:53 GMTm' From: "Hans Vlems" <hvlems@zfree.co.nz> 3 Subject: Re: SW Command Console for Raid array 3000a$ Message-ID: <3cb572f9$1@zfree.co.nz>  7 Yup, you're right it needs the HSZ22 specific software. < Do you have a pointer where I can find and download the kit?  
 Hans Vlems  : Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> wrote:E >Do you have the special version of SWCC that supports RA3000/HSZ22 ?d: >My EXE is called CcHsz22.EXE and when started first opens= >a dialog called "HSZ22 Management" to select between Serial,m8 >SCSI and TCP/IP. Serial is what to select in your case.6 >It wasn't until I found the original CD that was sent@ >with the controller, and installed from that, that I managed toB >get a working connection to the RA3000/HSZ22. Remember, the HSZ22/ >is *not* like the HSZ40/50/70/80 controllers ! : >If you run with the "standard" SWCC, you can't connect to
 >the HSZ22...o >e
 >Jan-Erik.       http://www.zfree.co.nz   ------------------------------    Date: 11 Apr 2002 02:03:33 -0700+ From: paul_hallam@hotmail.com (Paul Hallam)3R Subject: Re: TCPIP V5.0A VERSUS V5.1 coding problem with "network protocol  error"= Message-ID: <5ed44bd3.0204110103.3bc502a0@posting.google.com>i  C Many thanks for the efforts but having added the TCPIP$C_AF_INET to ? the P1 parameter I'm afraid I am still getting the same result.n   Paul   ------------------------------  + Date: Thu, 11 Apr 2002 06:58:06 +0000 (UTC)o  From: mustang@ucc.asn.au.invalid, Subject: Re: WIreless networking + VMS/Tru64* Message-ID: <a93c5u$9om$1@enyo.uwa.edu.au>  * Main, Kerry <Kerry.Main@compaq.com> wrote: : Peter,  I : I suspect what you want is to access VMS applications via WAP devices -r : is this correct?  E I suspect Peter wants to participate in a wireless LAN with an Alpha.yF Peter, I recommend that you either check out a Cisco Workgroup bridge,B or one of these: http://www.enterasys.com/products/items/CSISA-AX/   Cheers /daver --  % " I don't get mad.... I get stabby. "a - William "Fat Tony" Williams.   ------------------------------  # Date: Thu, 11 Apr 2002 10:41:01 GMT 5 From: peter.watkinson1@ntlworld.com (Peter Watkinson)e, Subject: Re: WIreless networking + VMS/Tru641 Message-ID: <3cb56477.1983265@news.cable.ntl.com>m  D On Thu, 11 Apr 2002 06:58:06 +0000 (UTC), mustang@ucc.asn.au.invalid wrote:  + >Main, Kerry <Kerry.Main@compaq.com> wrote:u	 >: Peter,u >dJ >: I suspect what you want is to access VMS applications via WAP devices - >: is this correct?n >cF >I suspect Peter wants to participate in a wireless LAN with an Alpha.G >Peter, I recommend that you either check out a Cisco Workgroup bridge,pC >or one of these: http://www.enterasys.com/products/items/CSISA-AX/r >o >Cheers  >/dave >-- & >" I don't get mad.... I get stabby. " >- William "Fat Tony" Williams.I    > Quite simple really. I want to connect my Cable modem internetE connection which is downstairs to a VMS/Tru64 workstation (when I get @ it) which will be upstairs without having to drill a hole in the ceiling. :-)  B This leads me onto another thought. Around Christmas time I poppedC into the local PC World computer store and while browsing I spotted0> the wirless networking equipment - this equipment seemed quiteF expensive at the time (this could have hust been PC World prices whichD arent renowned to be cheap here in the UK) anyway I thought about itF again so I had a look at http://www.dabs.com (who use VMS BTW and wereF on that Alpha-IA64 assurance program video that someone posted a whileF back) and the prices seemed much cheaper. This led me to think that ifE wireless networking starts to take off (bit slow here in the UK) thenlD pretty soon maybe people will begin to become less attached to thereF PC's in the home and will start to use wireless devices - laptops PDAsD etc and this could slow down the market for PC's even more. Also allE this wireless activity will require the need for some pretty powerfultF servers the sort that the Alpha processor is ideally suited. Maybe SunF and IBM can already see this happening and they probably can't beleiveA there luck at the alphacide. Maybe Compaq shareholders need to beFC aware of this and vote against the merger. Then compaq can do a SGI F have a management clearout bring back the Alpha and concentrate on HPC" and mobile devices such as IPaqs. F  If you see my post in comp.arch I suspect that because of the lack ofE games that IA64 will have no chance against Hammer on the PC desktop.-D No wonder SGI has to flog there Itaniums on Ebay there is no use for them except boat anchors ;-)   cheers,>    j Peter Watkinson  peter.watkinson1@ntlworld.com    ------------------------------  " Date: Thu, 11 Apr 02 16:15:04 +100 From: rok@nuk.uni-lj.sir, Subject: Re: WIreless networking + VMS/Tru64& Message-ID: <3cb5a87b.0@NUK.Uni-Lj.Si>  0 In Article <3cb56477.1983265@news.cable.ntl.com>7 peter.watkinson1@ntlworld.com (Peter Watkinson) writes:s > ? >Quite simple really. I want to connect my Cable modem internet F >connection which is downstairs to a VMS/Tru64 workstation (when I getA >it) which will be upstairs without having to drill a hole in the-
 >ceiling. :-)S  @  Did you consider PLC (Power Line Communications)? The wires are, allready there and you can get up to 24Mbps.   Regards,  D Rok Vidmar                       Internet:  rok.vidmar@nuk.uni-lj.si; National and University Library  Phone:     +386 1 421 5461 ; Turjaska 1, SI-1000 Ljubljana    Fax:       +386 1 421 5464  Slovenia   ------------------------------  # Date: Thu, 11 Apr 2002 16:29:51 GMT 5 From: peter.watkinson1@ntlworld.com (Peter Watkinson)U, Subject: Re: WIreless networking + VMS/Tru641 Message-ID: <3cb5b9e6.1183843@news.cable.ntl.com>o  9 On Thu, 11 Apr 02 16:15:04 +100, rok@nuk.uni-lj.si wrote:a  1 >In Article <3cb56477.1983265@news.cable.ntl.com>h8 >peter.watkinson1@ntlworld.com (Peter Watkinson) writes: >>@ >>Quite simple really. I want to connect my Cable modem internetG >>connection which is downstairs to a VMS/Tru64 workstation (when I get B >>it) which will be upstairs without having to drill a hole in the >>ceiling. :-) >lA > Did you consider PLC (Power Line Communications)? The wires arei- >allready there and you can get up to 24Mbps.e >i	 >Regards,b >rE >Rok Vidmar                       Internet:  rok.vidmar@nuk.uni-lj.si:< >National and University Library  Phone:     +386 1 421 5461< >Turjaska 1, SI-1000 Ljubljana    Fax:       +386 1 421 5464	 >Sloveniao       Hi Rok,g  B Do you mean internet via the mains electricity wires? I'm not sure that's available in the UK.o     regards,     Peter Watkinsonc peter.watkinson1@ntlworld.comT   ------------------------------   End of INFO-VAX 2002.200 ************************