1 INFO-VAX	Fri, 29 Jun 2001	Volume 2001 : Issue 357       Contents:1 Re: 3 Reasons why VMS is alive and probably well+  Alpha -> Itanium  Re: Alpha/IA64 - I had to do it.3 An actual arch question was Re: IA64 Rocks My World  Re: BACKUP listing []  Re: Changing platforms.  Re: Changing platforms.  Re: Changing platforms.  Re: Compaq switches to IA-64 Re: Compaq switches to IA-64 Re: Compaq switches to IA-64- Re: Compaq Transfers Alpha to Intel after EV7 < Re: Compaq Transfers Alpha to Intel after EV7 (was: Alpha de Re: DECnet over IP? @ Email Marketing Tools - Bullet-Proof Hosting! Sending! Software!D Re: Email Marketing Tools - Bullet-Proof Hosting! Sending! Software!  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium. Re: IA64 Rocks My World  Re: IA64 Rocks My World  Inform comedy.... ) Re: lsedit or whatever xwindow on my pc ? $ Re: My take on the Compaq-Intel deal6 Origin of iVMS? (was Re: Full port of VMS to Itanium.): Re: Origin of iVMS? (was Re: Full port of VMS to Itanium.)6 Re: Question to Charlie Matco - reply to Terry ShannonP Re: Question to Charlie Matco - reply to Terry Shannon (andsomegeneral comments)) Re: Sending command to programs with pipe ) Re: Sending command to programs with pipe ) Re: Sending command to programs with pipe ) Re: Sending command to programs with pipe # Thanks Compaq for the new business! ' Re: Thanks Compaq for the new business! ' RE: Thanks Compaq for the new business! ' Re: Thanks Compaq for the new business! 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated  Re: VAX-11/780 boot disk needed < Re: Wailing and moaning... (was: Question to Charlie Matco.)$ Re: Where is Samsung in this story ?$ Re: Where is Samsung in this story ?+ Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either % Re: Windows Images Running Under iVMS % Re: Windows Images Running Under iVMS % Re: Windows Images Running Under iVMS  www.vms-support.com for sale  F ----------------------------------------------------------------------  # Date: Fri, 29 Jun 2001 00:08:22 GMT . From: "Alphaman" <alphaman64@nixspam-home.com>: Subject: Re: 3 Reasons why VMS is alive and probably well+; Message-ID: <WFP_6.15772$P5.5148062@news1.rdc1.tn.home.com>   5 Arne Vajhj <arne.vajhoej@gtech.com> wrote in message # news:3B3B237B.73B7CBEA@gtech.com... - > I am sure that VMS will be ported to IA-64.  > * > [I am much more doubtfull about Tru64 !]  K Actually, Tru64 is a little ahead of OpenVMS -- expect developer's kits for K Tru64 at the end of 2002; for OpenVMS, the beginning of 2003.  (Think of it & like OpenVMS Alpha v1.0/1.5, I guess.)  ? > But more customers will drop VMS and more ISV's wil drop VMS. * > Which will start a bad spiral downwards.  L Not necessarily.  Yeah, there's a lot of cruft out today, but consider this:  L OpenVMS will not have the white box/blue box fiasco of Alphas.  OpenVMS willK run on a standard Itanium from Compaq initially, but it will be able to run J on other platforms (not necessarily supported by Q at first, until they'veK got time to qualify 3rd party hardware as necessary).  Then, wouldn't it be J sweet if any old ISV who bought an Itanium-II system could also buy a copyH of OpenVMS and dual-boot their development workstation?  Oh, and if theyJ bought a Compaq system, they could run Galaxy and screw the "dual-boot" --" just run both OS's simultaneously.  C Perhaps, just perhaps, software will be MORE available for OpenVMS?   J Don't forget, CSA membership and _site-wide_ SDK is MUCH cheaper than just  ONE single-user MSDN membership.   Aaron  --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)    ------------------------------  % Date: Thu, 28 Jun 2001 22:08:23 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca>  Subject: Alpha -> Itanium 9 Message-ID: <ZmR_6.6370$lp3.438632@news20.bellglobal.com>   L It was my understanding that Intel didn't want anything to do with the AlphaK chip or related FAB (if you replay the webcast, you actually hear this from E the lips of one of the IBM execs). However, they do want the complier K technology (especially multi-threading tools) and the people to support it. K Also, I believe they want some assurance from Compaq that Alpha will not be I developed any further after EV7 which then means a ready market place for L IA-64 down the road. By taking some of all of the Alpha microprocessor team,1 that should put the final nail in Alpha's coffin.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/   
 **********  9 "Chuck Taylor" <Chuck.Taylor@Vishay.com> wrote in message 1 news:005701c0fe97$9f884a80$b7081eac@vishay.com... G Does anyone have any insigt as to what this "transfer of technology" to J Intel actually means?  Will a follow on microprocessor from Intel actuallyG be an Alpha in Intel clothing? or will there be some sort of merging of - technologies or will there be something else?   L As I understand the way things are right now: the Alpha has some 15 years orJ so of actually being used and proven in the marketplace and the Itanium isG just beginning to ship in very small numbers (I am not sure that it has J actually shipped - Does anyone out there know of anyone who has one?).  ItI seems to me - granted it may be a simplistic approach - but the people at H Intel could do a lot for their tarnished reputations by just repackagingL and/or relabeling the existing Alpha as their 64-bit solution.  Could anyoneL comment on the differences between the current Itanium and the current AlphaE (instruction set differences, performance, architecutral differences, J whatever, but please try to keep it technical and to the point - there areK too many religious based arguements already - I want some facts - data that @ I can use to make a reasonable decision about what it happening.  H This looks to me that Compaq is getting a lot closer to what they wantedL from DEC in the first place - the services organizations.  Now all they needL to do is get someone to buy the operating systems and the remainind software8 groups and they will get to what they originally wanted.  H It could be a good thing and it could be a bad thing.  I just don't haveG enough technical data to know what the announcement actually means yet. K They were probably vague on purpose but it would still be nice to know what  is intended.  * 17 years of VMS (started with version 3.4)   Chuck Taylor2 Senior Infrastructure Developer Vishay - Siliconix chuck.taylor@vishay.com    ------------------------------  # Date: Thu, 28 Jun 2001 23:43:25 GMT . From: "Alphaman" <alphaman64@nixspam-home.com>) Subject: Re: Alpha/IA64 - I had to do it. ; Message-ID: <xiP_6.15522$P5.5124943@news1.rdc1.tn.home.com>   < Jay E. Morris <morris@thorin.brooks.af.mil> wrote in message& news:9hffoh$kku$1@leo.brooks.af.mil...0 > http://www.epsilon3.com/images/alphainside.gif >    Don't forget my updated one:  2  http://members.home.com/sakovich/AllYourAlpha.jpg   -- Aaron  --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)    ------------------------------  # Date: Fri, 29 Jun 2001 03:24:55 GMT / From: andrew@gurney.reilly.home (Andrew Reilly) < Subject: An actual arch question was Re: IA64 Rocks My World6 Message-ID: <slrn9jnt88.4cd.andrew@gurney.reilly.home>  N > The i960CA was actually the first single-chip superscalar processor in 1990. > J > I'm not completely familiar with the ARM family, but a blurb in May said8 > they are working on a superscalar core called Cheetah.  % I wonder how sensible that really is.   < ARM does pretty nicely already, as the simplest of in-order,< single(*),short-pipe architectures.  Xscale cranks the clock< up to 1GHz.  Performance then depends almost entirely on the: latencies of the memory hierarchy.  Integer (control) code? doesn't gain much in the way of ILP besides loads and branches, ( perhaps, thanks to logical dependancies.  < * Not quite single: the integer multiply is decoupled to the: extent that you can issue non-mulitply instructions in the@ four-cycle shadow of the multiply unit.  I think that intel have? promised to decouple the load unit in a similar fashion.  Maybe / that's what you mean by the term "superscalar".   ; Yippee: Itanium does nicely on floating point code.  But so ; do vector processors, provided that they have enough memory ? bandwidth.  SSE(2), Altivec and EmotionEngine seem to have that < base covered, although the auto-vectorising compilers aren't@ there yet: fast code is all hand written.  Didn't the old vector0 supercomputers already have this wrinkle sorted?  " So here are the questions for c.a:  : Was Alpha really giving us anything that an ARM/MIPS/SPARC= with a vector floating point unit bolted on the side couldn't  provide?  : Does Itanium have anything really going for it besides two7 floating point MAC units (instead of one) and the cache  bandwidth to keep them happy?    --   Andrew   ------------------------------  % Date: Fri, 29 Jun 2001 00:32:09 -0400   From: John Santos <JOHN@egh.com> Subject: Re: BACKUP listing []6 Message-ID: <1010628234408.61768B-100000@Ives.egh.com>  & On 28 Jun 2001, Alan E. Feldman wrote:  d > Ingemar Olson <IOLSON@dairyworld.com> wrote in message news:<01K585U7CDES8ZE80I@dairyworld.com>...D > > I had occasion to actually *look* at one of our backup listings.L > > At the end there were a large number of files which were listed as being" > > in directory "[]" (ie: blank).C > > When I check the disk I find they *are* in an actual directory.  > > H > > This is happening for 2 of our disks, the system disk and one other.Q > > The files from the system disk that show up on the listing as in "[]" appear  R > > to be those in the aliased common directories. Although this is a bit hard to K > > say definitively since I haven't memorized all the files in that tree.  * > > And there are more than 10000 of them.E > > The directory VMS$COMMON does not appear on the listing, however.    [snip about fixing system disk]    > > P > > On the "other" disk we do not have any aliased directories, nor files, that Q > > I'm aware of. In fact I listed out the file-id of every file on the disk and  K > > the sample file's (from the backup listing) file-id appeared only once.  > > P > > It would appear that the backup *is* doing all the files ok, but the listing( > > is confusing / misleading at least.  > E > I suspect that a restoration of this backup will not restore the [] F > files in their proper directories because BACKUP walks the directoryB > structure and checks off files it backs up in a copy it makes of? > INDEXF.SYS. Then any files that are not checked off after the B > directory walk-thru are backed up in []. If it had the directoryH > information for those [] files, I'd think it would have listed them as> > such. (This is based on an old post [c. 1993] explaining theD > difference between /IMAGE and /FAST backups -- I can dig it up andE > re-post it if anyone desires. The explanation there is quite good.)   F I think /NOALIAS is more recent than that and modifies this behaviour.  G What changes is that during the directory walk-through, BACKUP compares E the file's directory back-link and filename (both in the file header) C to the directory entry BACKUP followed to get to the file.  If they C don't match, /NOALIAS causes the file to be skipped.  If it matches C (which is only possible once), it is regarded as the primary entry, D and gets backed up.  Then at the end, after the directory walk-thru,A the any remaining files are backed up as you said, with [] as the A directory.  (These are regarded as "lost" files, even though they ; may have live alias entries, and thus are not really lost.)   F So if a file has an incorrect directory back-link, it will get skipped1 in the directory scan, but caught in the [] pass.   C You can determine what the directory back-link and file header name C are by using $ dump/header.  For example, if dka100:[foo.dat]a.file % is the problem file, then you need to   2 $ dump/header/block=count=0 dka100:[foo.dat]a.file   and look for the:   1 Back link file identification:        (28837,3,0)    and    Identification area 2     File name:                            A.FILE;1  4 lines.  Then look at dka100:[foo]dat.dir;1's file id  ! $ dir/file_id dka100:[foo]dat.dir    Directory DKA100:[FOO]   DAT.DIR;1       (28837,3,0)   > If the file id's (the three numbers in the parentheses) match,@ but the file name in the header doesn't match, then this is your: problem.  I think you can fix this by renaming the file to> what it says in the header and then renaming it back.  (If the@ header says its name is B.FILE;2 and there already is a B.FILE;2@ in the directory, temporarily rename B.FILE;2 to something else,> then rename A.FILE;1 to B.FILE;2 and then back to A.FILE;1 and$ then restore the original B.FILE;2.)   This should fix the problem.  = If the file id's (the three numbers in the parentheses) don't ; match, but the names do, then you need to fix the backlink.   ? I think ANA/DISK will complain if the backlink doesn't point to A a valid directory file, but not if it points to a valid directory = file that doesn't happen to point to the file.  Rename can be > used to move a file from one directory to another, but it will: only change the backlink if it thinks you are renaming the? primary entry.  So there should be a directory somewhere on the ? disk with file id (28837,3,0), so do a directory listing of the > entire disk to find it.  If there is a file with the same name< (but a different file - verify this with dir/file_id) in theD directory, move that file out of harms way, and do a cross-directory= rename and back to fix it.  If both the name and backlink are < wrong, you'll need to fix both at the same time (i.e. header= says it is B.FILE;2, in a directory you find to be [FOO.XXX], < rename [FOO.DAT]A.FILE;1 to [FOO.XXX]B.FILE;2 and then back.@ If you only fix the backlink but not the name, RENAME will think/ it is dealing with an alias and leave it alone.   @ If ANA/DISK does complain about the backlink (because it doesn't@ point to a directory), then I think ANA/DISK/REPAIR will fix it.@ However, it might make the backlink point to [SYSLOST], in whichA case you will have to apply the "wrong backlink, right name" fix.   > If both the backlink and name are correct and ANA/DISK doesn't2 complain, then maybe you've hit a real backup bug.  : What version of VMS, and BACKUP ECO's applied, etc., etc.?  G > Therefore, I think there is an overflow/bug problem like you say in a  > later post in this thread. >   H > Have you tried restoring this save set to a scratch disk? If so, or ifE > you do, what are the results of where the [] files are restored to?   A I think in all these cases, backup/image followed by restore will ! faithfully preserve the problems.   ? The file will get restored as a lost file, but it will keep its A original file ID, and still will be recorded (as an alias) in any > directories that point to it.  Even though the files won't getA backed up under their alias entries, the directories *containing* < the aliases will get backed up, and restored, and will still> contain the same aliases, which will still point to the "lost" file.    > [...ending stuff omitted...] >  > Disclaimer: JMHO > Alan E. Feldman  > afeldman&gfigroup.com   @ P.S.  As John Macallister noted in another post, some lost files= are perfectly normal on an active disk, either temporary work ? files that some program created and is still using, or possibly ? temp files left by a program that didn't clean up after itself.u  9 Also, printing from MAIL and writing to a spooled printers4 create these files.  ANALYZE/DISK will find them and/ ANALYZE/DISK/REPAIR will put them in [SYSLOST].   B But I think from other posts, the files that aren't getting backedD up in the proper directories are normal-looking, non-temporary files8 that are in fact cataloged in some directory or another.   -- i John Santos4 Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Fri, 29 Jun 2001 02:38:42 GMTe. From: "Duane Sand" <duane.sand@mindspring.com>  Subject: Re: Changing platforms.@ Message-ID: <SSR_6.138848$%i7.92850551@news1.rdc1.sfba.home.com>   "Bill Todd" wrotetH > ... But that's sure as hell where they devoted most of their corporate effort,lL > as contrasted with the Alpha end of things (which *was* quite profitable).  J Really?  How long ago was it, that Alpha was profitable?  I don't remember a time.    ------------------------------  % Date: Fri, 29 Jun 2001 00:17:05 -0400d' From: "Bill Todd" <billtodd@foo.mv.com>y  Subject: Re: Changing platforms.( Message-ID: <9hgv9f$8u1$1@pyrite.mv.net>  9 "Duane Sand" <duane.sand@mindspring.com> wrote in messageI: news:SSR_6.138848$%i7.92850551@news1.rdc1.sfba.home.com... >n > "Bill Todd" wroterJ > > ... But that's sure as hell where they devoted most of their corporate	 > effort,oA > > as contrasted with the Alpha end of things (which *was* quitee profitable). >pL > Really?  How long ago was it, that Alpha was profitable?  I don't remember	 > a time.   L Well, VMS's profit a couple of years ago was about $800 million.  Even usingG Winkler's $300 million annual Alpha development figure (which certainlyfJ seems likely to be the absolute limit it could be), and even assuming thatK somehow none of that cost was included in figuring out VMS's profit (though:H while I'm no accountant my suspicion would be that at least a percentageI equal to VMS's share of total Alpha sales was deducted from gross revenue J before determining VMS's profit), VMS's profit handily exceeds the *total*J Alpha development cost, hence by definition Alpha was profitable that year? (unless you think that Compaq sold one hell of a lot of VAXen).   K One can question how important Alpha was to the sales vs. how important VMSgH was (and even possibly arrive at some tentative conclusions by examiningI Unix and Linux Alpha sales), but in the end you can't separate Alpha from-H VMS's profits because you can't have VMS without Alpha (VAXen excepted).  F Don't bother pointing out that this connection will be broken in a fewH years' time now:  that's irrelevant to the statement in question.  If byJ 2004 or so VMS is thriving on IA64, however, I'll happily eat all my wordsJ about the ill-advised nature of recent and not-so-recent Compaq decisions,H even if other changes (e.g., in VMS marketing activity) have occurred asI well; conversely, if VMS is dead or dying (or Compaq is dead) before thatn time, I trust you'll eat yours.n   - bill   ------------------------------  # Date: Fri, 29 Jun 2001 05:45:15 GMTV. From: "Duane Sand" <duane.sand@mindspring.com>  Subject: Re: Changing platforms.@ Message-ID: <LBU_6.139045$%i7.93197619@news1.rdc1.sfba.home.com>   > > "Bill Todd" wrotenB > > > ... But that's sure as hell where they devoted most of theirB > > > corporate effort, as contrasted with the Alpha end of things% > > > (which *was* quite profitable).l > >p > Duane Sand asked:uE > > Really?  How long ago was it, that Alpha was profitable?  I don'te remember > > a time.e >f Bill Todd replied:% > ... (details about VMS profits) ...  >oI > One can question how important Alpha was to the sales vs. how importanty VMSgJ > was (and even possibly arrive at some tentative conclusions by examiningK > Unix and Linux Alpha sales), but in the end you can't separate Alpha fromEJ > VMS's profits because you can't have VMS without Alpha (VAXen excepted).  I I don't challenge that VMS makes nice earnings from offering good systemsa> with good software features to a very desirable customer base.E But somehow the net total of everything associated with Alpha, ie thedG entire division formerly known as Digital, is consistently losing money-F every quarter, and those losses are only barely covered by the profitsI from the division formerly known as Tandem.  I believe it's been that waygG since before both were acquired, and maybe long before, given Digital'sa@ long decline before acquisition.  The fixation on home-grown cpu@ chips named Alpha seems to be a big part of the expense problem.  @ The new plan is to continue offering similarly good systems withA the same good software to the same (and perhaps growing) customerp< base but using a cheaper-to-develop but still fast cpu core. I hope that works for us all.b   ------------------------------  # Date: Fri, 29 Jun 2001 02:07:49 GMT . From: "Duane Sand" <duane.sand@mindspring.com>% Subject: Re: Compaq switches to IA-64t@ Message-ID: <VpR_6.138813$%i7.92798126@news1.rdc1.sfba.home.com>   "David Elliot" askedH > Ah, but what about support for Vax-Float.  Will have to be emulated or will" > IA-64 introduce a new data type?  E No, Intel did not bother implementing any other company's proprietarysB obsolete FP formats, but it did implement all sizes needed for x86 execution/emulation.  / The chances of Intel adding Vax format is zero.lG And you don't want systems built out of special-order customized chips.oA You want systems built from the same high-volume pool of chips asn everyone else that year.0 So Vax fp will need to be done by some software.  D If ranges and mantissa size fit within a supported IEEE format, then= that emulation software would likely convert to and from IEEE B format for each Vax FP operation. Optimizing that out for repeated5 ops would be nice, but requires a smarter translator.a  > On NSK, the proprietary format loses one bit of exponent range> or one bit of mantissa precision on such conversions, plus the; original NSK microcode implementations had some funky rulesn; that Prof IEEE FP would hate.  We wound up implementing NSKA: FP totally in software following the old rules compatibly,* with no use of host machine's IEEE FP hdw.9 The poor performance didn't matter to our customers untilr6 two years ago.  Our telecom customers now want fast FP6 for government-mandated spying on cell phone locations via signal triangulation.s   ------------------------------  % Date: Fri, 29 Jun 2001 00:23:31 -0400D2 From: rdeininger@mindspring.com (Robert Deininger)% Subject: Re: Compaq switches to IA-64eL Message-ID: <rdeininger-2906010023320001@user-2ivebg3.dialup.mindspring.com>  G In article <VpR_6.138813$%i7.92798126@news1.rdc1.sfba.home.com>, "Duanel( Sand" <duane.sand@mindspring.com> wrote:  1 > The chances of Intel adding Vax format is zero.lI > And you don't want systems built out of special-order customized chips.iC > You want systems built from the same high-volume pool of chips aso > everyone else that year.2 > So Vax fp will need to be done by some software. > F > If ranges and mantissa size fit within a supported IEEE format, then? > that emulation software would likely convert to and from IEEE>D > format for each Vax FP operation. Optimizing that out for repeated7 > ops would be nice, but requires a smarter translator.r  F The old alpha architecture documentation actually says the VAX formatsF might be subsetted out of the alpha in a future incarnation.  It seemsH clear VAX floats went into alpha as a (necessary) porting aid, and folks' should be planning a migration to IEEE.p  I At some point (maybe already past?) a bit-perfect emulation of VAX floats C in software on a fast CPU will be faster than the best VAX hardwareaJ implementation.  At that point, a customer who REALLY needs VAX floats can; migrate to a new platform without taking a performance hit.y   -- A Robert Deininger rdeininger@mindspring.comi   ------------------------------  % Date: Fri, 29 Jun 2001 00:55:46 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>s% Subject: Re: Compaq switches to IA-641, Message-ID: <3B3C0A44.3D9436CF@videotron.ca>   Robert Deininger wrote:iK > At some point (maybe already past?) a bit-perfect emulation of VAX floatsdE > in software on a fast CPU will be faster than the best VAX hardwaretL > implementation.  At that point, a customer who REALLY needs VAX floats can= > migrate to a new platform without taking a performance hit.   D What types of applications would make use of vax floating point in a; persistant (written to file or exchanged via telecom) way ?   L If you're exchanging data with any other system, wouldn't be either in ascii text or IEEE format already ?t  K Is it correct to state that it would be a very small number of applicationsmL for whom the internal bit representation of a float or double would matter ?   ------------------------------  # Date: Fri, 29 Jun 2001 01:22:10 GMT"2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)6 Subject: Re: Compaq Transfers Alpha to Intel after EV71 Message-ID: <6LQ_6.272$rc5.7519@news.cpqcorp.net>n  e In article <cRP_6.6337$lp3.396310@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:e  I :...and was suprised to hear a statement by an Intel exec that didn't getaK :very much coverage in this news group or on other web pages. The statementhJ :is that "Intel isn't buying Alpha hardware, and won't be making any Alpha :chips."  C   So?  (I am not certain what assumptions you are operating under.)a  E   While Intel fabs were a source of Alpha microprocessors as part of nD   the DIGITAL sale of the Hudson fab line, IBM serves as foundry for=   the EV7-generation Alpha microprocessors.  For information:0  7     http://www.chips.ibm.com/news/2000/0525_compaq.htmli  H :This means that there is no future for Alpha if Compaq decides to kill  :it in a few years.   F   Work on EV7 (and shrinking it) proceeds at Compaq, as is work on theI   AlphaServer systems that will be based on EV7.  The Alpha semiconductorkG   technology and tools will be transfered to Intel, and (over the next rJ   few years) various the Alpha microprocessor and compiler engineers will I   be offered jobs at Intel -- either now, or as the current projects are  I   completed.  Work on IPF-based (IA-64) systems -- and on porting OpenVMS J   and Tru64 UNIX to IPF -- is underway at Compaq.  For information on some"   of the Compaq IPF hardware, see:  @     http://www.compaq.com/products/servers/proliant64/index.html  K :Remember that the IA-64 project was started in 1995 and was to be ready by I :1997. Delay after delay has transformed Itanium into nothing more than ah :protyping chip for McKinley.b  I   My understanding of the terminology is that Itanium is the name of the aJ   processor family, and not the name of a specific processor.  The Merced H   and McKinley projects are implementations of Itanium.  All that said, G   yes, it has taken a number of years to get IPF (IA-64) systems -- theaK   microprocessor, hardware, and software -- into production.  It will take iI   substational time and effort to port OpenVMS to IPF, and Microsoft has uC   been occupied porting "Windows 2002" and into production as well.s  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Thu, 28 Jun 2001 16:57:23 -0400G) From: "Neil Rieck" <n.rieck@sympatico.ca> E Subject: Re: Compaq Transfers Alpha to Intel after EV7 (was: Alpha de 9 Message-ID: <cRP_6.6337$lp3.396310@news20.bellglobal.com>i  E I finally got around to viewing the Monday morning announcement at...A9 http://www.compaq.com/newsroom/presspaq/062501/index.html H ...and was suprised to hear a statement by an Intel exec that didn't getJ very much coverage in this news group or on other web pages. The statementI is that "Intel isn't buying Alpha hardware, and won't be making any Alphaa chips."   L This means that there is no future for Alpha if Compaq decides to kill it inF a few years. Alpha could live if Itanium (a.k.a. un-obtainium) (a.k.a.F Titanic-tanium) flops and there's enough people remaining in the AlphaI microprocessor division left to sell the whole package to another companycK (IBM? AMD?). Of course, the conspiracy theorists already know that Intel is G only taking on Alpha division staff so that Alpha doesn't rise from thel dead.r  F On a related note, today ASUS (not a server manufacturer) and FIC bothH announced that they will postpone building an Itanium motherboard for atL least a year. The reason? Almost no OS support and no demand translates into
 no market.' http://www.theinquirer.net/28060108.htm J Remember that the IA-64 project was started in 1995 and was to be ready byH 1997. Delay after delay has transformed Itanium into nothing more than a protyping chip for McKinley.  
 Neil Rieck Kitchener/Waterloo/Cambridge,h Ontario, Canada.! http://www3.sympatico.ca/n.rieck/i   ------------------------------  % Date: Thu, 28 Jun 2001 17:59:32 -0700o0 From: Mark Berryman <Mark.Berryman@Mvb.Saic.Com> Subject: Re: DECnet over IP?, Message-ID: <3B3B7084.1108A995@Mvb.Saic.Com>   Didier Morandi wrote:a > C > I want to demonstrate to someone that DECnet Phase V task to taski9 > communication "over IP" works as well as DECnet IV did.  > F > To do that, I need to disable DECnet routing to force the flow to goA > through the IP circuit (I am not very familiar with DECnet OSI)o  F Urk (as I choke on a late lunch).  No.  You do not disable anything to use DECnet over IP.p  D First, test your setup to make sure you can even do DECnet over IP. H What is the IP address of the remote system with which you are trying toE communicate?  Assume, for this example, that it is 192.168.0.1.  Thent type the command:n  ) $ dir ip$192.168.0.1"username password"::e  D where username and password are a valid username and password on the remote node.  H If this works, DECnet is configured properly, the PWIP driver is loaded,? etc.  You can do DECnet over IP.  If not, you haven't installedlF something correctly and I need more info to go on (e.g., the error you get when you try this).o  	 Next try:    $ dir ip$192.168.0.1::  D If this fails, you have no proxies set up on the remote node and youA will need to continue to specify the username and password in the  command line until you do.    C Now, what is the fully qualified domain name of the remote system?  7 Assume, for this example, that it is: test.mydomain.come   Issue the command:  D $ dir test.mydomain.com:: (with username and password, if necessary)  G If this works, great.  If not, you do not have a functioning DNS serverxE or the DNS configuration on your local node is not done correctly, or @ you have not configured you DECnet naming search path to include domains.  G Fixing your DNS, if that is the problem, is more than I want to go intoyF here but, to verify your DECnet naming search path, issue the command:  1 $ mcr ncl show session control naming search pathi  E You should see multiple Directory Service entries displayed.  Here is. what mine looks like:d    '     Naming Search Path                =f        {           [h%           Directory Service = Local ,s           Template = "*"
           ] ,e           [p%           Directory Service = Local ,f           Template = "local:*"
           ] ,e           [ %           Directory Service = Local ,g           Template = "LOCAL:.*"e
           ] ,l           [ &           Directory Service = Domain ,           Template = "*"           ]         }  F You must have at least one entry for Directory Service = Domain or youH will not be able to use domain names in your node reference; you will be? stuck with only the IP$x.x.x.x syntax.  In my example the entry B (Template = "*") lets me specify fully qualified domain names as aD DECnet host (e.g., dir www.compaq.com::).  I can also specify just aF node name and, since my DNS is configured to add a default domain to aH simple node specification, the lookup still works.  In other words, whenG I type $dir node::, the DNS looks up node and fails, it then applies myaF default domain (.SAIC.COM) and looks up node.saic.com which succeeds. D Most IP stacks will let you specify multiple default domains but the! method is specific to each stack.e  ' The search path is defined in the file:p& SYS$MANAGER:NET$SEARCHPATH_STARTUP.NCL   Here is what mine looks like:c   !%  NET$CONFIGURE.COM Createdt? !%  NCL Script SYS$SPECIFIC:[SYSMGR]NET$SEARCHPATH_STARTUP.NCL;- !%  26-AUG-1999 16:00:05.15- !%@ !%  Edited by MVB 26-AUG-1999 to add support for SAIC.COM domain defaults !%/ SET NODE 0 SESSION CONTROL NAMING SEARCH PATH -o7         ([DIRECTORY SERVICE = LOCAL, TEMPLATE = "*"], -f<         [DIRECTORY SERVICE = LOCAL, TEMPLATE = "local:*"], -=         [DIRECTORY SERVICE = LOCAL, TEMPLATE = "LOCAL:.*"], -c5         [DIRECTORY SERVICE = DOMAIN, TEMPLATE = "*"])t- SET NODE 0 SESSION CONTROL BACK SEARCH PATH - 6         ([DIRECTORY SERVICE = LOCAL, TEMPLATE = ""], -4         [DIRECTORY SERVICE = DOMAIN, TEMPLATE = ""]); SET NODE 0 SESSION CONTROL NAMING CACHE TIMEOUT 30-00:00:00yD SET NODE 0 SESSION CONTROL NAMING CACHE CHECKPOINT INTERVAL 08:00:00  C Hopefully, this will provide enough of a guideline to show you whate yours needs to look like.1  G You will probably want to add proxies for your IP connections.  If yoursH DNS system is properly set up to provide PTR records you can add proxies1 using fully-qualified domain names.  For example:s  ? $ mcr authorize add/proxy remote-host.mydomain.com::remote-userp local-user/default  D If you don't have reverse pointers in your DNS, you must use the ip$ form:o  D $ mcr authorize add/proxy ip$x.x.x.x::remote-user local-user/default  A The easiest way to find out which one you must use is as follows:    $ set host ip$192.168.0.1n  * <sign in to any valid interactive account>   $ show terminalS  < and use the contents of the Remote Port Info: as your guide.    ! Note that using a syntax such as:V   $phone ip$192.168.0.1::user  or $phone host.mydomain.com::user  / will force DECnet over ip but a syntax such as:y   $phone nodeo  E will use a transport based on whichever directory service the node isa first found in.o   I hope this helps,  
 Mark Berrymanl   ------------------------------  # Date: Thu, 28 Jun 2001 23:19:08 GMTs From: <fgherthj@yahoo.com>I Subject: Email Marketing Tools - Bullet-Proof Hosting! Sending! Software!d- Message-ID: <MXO_6.29452$rO.12304@news.rt.ru>e  - EXPLODE YOUR BUSINESS WITH INTERNET MARKETING'5 Direct Email Works!  You're Reading This Email Right?hE Just Imagine That We Sent Out Another 9,999,999 Emails Like This One!m  o WHY IT WORKSH If you have a product or service that people want at a desirable price, 7 then Direct Email works because it is cost effective.   A You can deliver your message to 1 Million people for only $449.  vB That's right it's less than a classified ad that would only reach  a fraction of the audience. B http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/   WHAT WE OFFER YOUiA We offer a suite of services to make your Direct Email Marketing 5? efforts truly "effortless" for you .  It's a complete turnkey s:> olution for you...but you can obviously choose to participate < in any one aspect of it if you like.  This services include:  uA -Direct Email Sending Service - We can send out up to 10 Million rO emails a week for you alone for only $449/million and we supply email addressespB http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/  A -Email Creation Service - We can create an effective email piece   for your campaigneB http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/  D -Bullet-Proof Direct Hosting - We can host your site so it does not C get shut down for Direct marketing for as little as $399 per month nB http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/  F -Remove List Management - We will manage your remove requests for FREEB http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/  g EASY AS 1...2...3... It's easy to get started!   ? Simply place your order today and start mailing the same day.   > Hosting is usually set up the same day or the very next day.  ; Creative work can usually be turned around same-day also.  tH So you can actually be mailing your message to 1 Million people today!  A Simply call (800) 322-0251 or click here to get more information:bB http://Direct-depot.com-cgi.bin.url.js@209.84.189.222/bulletproof/   ------------------------------  # Date: Thu, 28 Jun 2001 23:38:58 GMTh4 From: "Terry C. Shannon" <terryshannon@mediaone.net>M Subject: Re: Email Marketing Tools - Bullet-Proof Hosting! Sending! Software!o; Message-ID: <meP_6.1895$UT1.956553@typhoon.ne.mediaone.net>u   abuse@rt.rut    % <fgherthj@yahoo.com> wrote in messaget' news:MXO_6.29452$rO.12304@news.rt.ru...-/ > EXPLODE YOUR BUSINESS WITH INTERNET MARKETING 7 > Direct Email Works!  You're Reading This Email Right?jG > Just Imagine That We Sent Out Another 9,999,999 Emails Like This One!6   ------------------------------   Date: 28 Jun 2001 21:46:53 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)) Subject: Re: Full port of VMS to Itanium.p, Message-ID: <9hg8kd$jpk@gap.cco.caltech.edu>  R In article <9hg0ab$opv$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes:M >Boards of Directors have ways of fixing management/marketing deficiencies in < >companies, rather than simply accepting their incompetence.  K That assumes a certain level of board competence.  I think we're seeing nowiD why Mr. Rosen took his marbles and walked away from Compaq.  Can youI imagine the frustration of being in the minority "dump the turkeys" groupm on a BOD at Compaq or Digital?    $ >  If BoDs don't, stockholders can.   E It is very, very rare for stockholders to replace a board in order tooK replace managment.  Except in those instances where a takeover is involved.o  ' > If no one else does, the market will.e  H Bingo.  Well, it does weed out the weaker players. The sad truth is that@ poorly run companies die far more often than they receive better management.      David Mathog mathog@caltech.edu? Manager, sequence analysis facility, biology division, Caltech sJ **************************************************************************J *                       RIP VMS & ALPHA                                  *J **************************************************************************   ------------------------------  % Date: Thu, 28 Jun 2001 18:33:54 -0400r' From: "Bill Todd" <billtodd@foo.mv.com>n) Subject: Re: Full port of VMS to Itanium.6( Message-ID: <9hgb60$4dm$1@pyrite.mv.net>  J "Brian Schenkenberger, VAXman-" <system@SendSpamHere.ORG> wrote in message* news:009FE371.E75A7DEC@SendSpamHere.ORG...   ...s  9 > What were you smoking JF?  Architecture???  From intel?a >oL > Can you say 8086? ...80186? ...80286? ...80386? ...80486^H^H^H^H^HPentium?  2 I think you inadvertently omitted ...80586' above.   >o8 > Intel has never been concerned with any compatibility!  J I'm no expert, but I'd have to see explicit examples (e.g., non-privilegedD instructions in some generation that aren't supported in some futureH generation) before I believed this.  I've got XT DOS software that still+ runs under Pentiums and Win9x, for example.l     In fact, it behoovedJ > them to be incompatible to boost their profits.  Out with the old and in withJ > the new every generation.  How many landfills are full of PC-AT and XTs?  K They're full mostly because  a) those systems are now 15 or more years old,lK components have died, and replacing them would be *really* cost-ineffectivepC and  b) most of today's software requires rather more in the way ofPL processor and memory resources than were available on systems made back thenE (not to mention using instructions that weren't available back then).i  A New instructions really do constitute features, not objectionablerI incompatibilities:  it's *upward* compatibility that's usually important,1H and AFAICT Intel has always done an excellent job of maintaining it.  IfH *downward* compatibility was important to any application, it could haveH achieved it simply by not using the newer instructions (and watching itsL memory requirements) - but that's not an Intel responsibility but a software one.     CanV: > I run Weendoze on an early 1980's vintage Compaq PeeCee?  H You could if Microsoft had wanted to package Windows that way:  the factL that it didn't is not Intel's fault (nor Microsoft's, either:  it would haveA been *really dumb* and really ignored what its customers wanted).   B You mention emulation of new instructions below.  Do you think anyK reasonable person would want to run something the size of today's operatingaC systems on 16-bit hardware emulating 32-bit instructions (and stillrI requiring that the code be structured to use 64KB overlays), even if such-H emulation were available?  (Don't forget that there's a good chance thatL they can run their old OS and applications on current hardware, by the way.)  F And new instruction additions aren't all that hard to avoid for lesserL architectural jumps:  if you want your application to run on everything backL through 386s, compiling it with a compiler of suitable vintage should ensureK that (assuming your current compiler doesn't have switches controlling suchgI choices:  I've never checked, but their existence wouldn't surprise me atbK all, and it certainly would seem to be an appropriate approach), as long as I the OS offers adequate upward-compatibility (which has nothing to do witho Intel).s  G Win95 will run on 386s, by the way.  I think Win98 as well.  Don't knowyJ about Win98SE or later, but it wouldn't surprise me:  the reason NT didn'tJ (though it did early-on) was IIRC performance, including the use of one or) two 486 instructions it wanted available.    >lA > Will IA64 have an "escape mechanism" whereby older chips in thed architectureC > can run software built for the newer? (ie. instruction emulation)h  K I don't know, but it would seem desirable:  unlike the PC space, higher-end D system hardware can outlast enough software (and follow-on hardware)I generations to make this desirable.  But, as noted above, any application L concerned with such downward compatibility could achieve it on its own if it
 wanted to.     Will thereE > be an "architecture manual" for IA64 the likes of that which we are  familiarG > with in the VAX and Alpha realm?  I simply don't see intel making diei changeseL > to add in necessary support for VMS as their true focus is toward Redmond.  I They might not have a problem merging in a couple of hooks in a couple oftC years, if they really were necessary:  it's the suggestion that anyeA significant Alpha *architectural* features will appear that seems  ridiculous.a  H Damn - didn't mean to get so diverted from my Truth Squad efforts, but IK guess this response falls into that category even if it's pretty tangential  to the porting flap.   - bill   >  > --4 > VAXman- OpenVMS APE certification number: AAA-0001 VAXman(at)TMESIS(dot)COM >lK > city, n., 1. a place where trees are cut down and streets are named after  them.    ------------------------------    Date: 28 Jun 2001 16:51:41 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)-) Subject: Re: Full port of VMS to Itanium.O, Message-ID: <zxBLxECW9Hy8@malvm5.mala.bc.ca>  ) In article <9hg43n$rr4$1@pyrite.mv.net>, n-     "Bill Todd" <billtodd@foo.mv.com> writes:   0 >>       The fate of Alpha was sealed years ago. > N > This ridiculous assertion just refuses to disappear.  But so do I:  it's too+ > much fun being part of the 'truth squad'.u > G     Have it your way, the fact remains Compaq has abandoned Alpha, much-; as many predicted they would given it's failure to become alD commodity item. You may feel Compaq should have continued developingB Alphas strictly for the Tru64 and VMS market ( the only places its; use was assured ), but they apparently don't feel that way.e  K     My assertion is that if Windows on Alpha had been a success we wouldn'ta today be arguing about this.  " >  If Microsoft had actually givenH >> a d**n about producing an OS not tied to a single processor ( as they	 > claimedi! >> when they first announced NT )  > E > As time went on, the value of other processors to MS was seen to be M > marginal, and MS not-entirely-unreasonably decided that the owners of thosew. > processors should pay their support costs.    G     Why was the ability of Windows to run on other processors ever seen $ as a useful goal then? What changed?  A     How much does Intel pay each year to support Windows on theiraC processors? Maybe you can argue they shouldn't have to since x86 is E the de-facto Windows standard, but what about 64 bit Windows? If thattD needs to be developed from the ground up anyway why should Intel get$ any different treatment than Compaq?  ! >Several owners decided that thisnF > was not worthwhile for them, either (including, eventually, Compaq). >   G     It was MS's lacklustre commitment to Alpha that killed NT on Alpha.cK You couldn't even get a full set of "MS Office" software for it - let alonekM all the MS backend utilities like Exchange server, SQL server, etc. How could F one expect any of the 3rd party vendors to support Alpha when even MS $ obviously had little interest in it.   > K > Compaq always had a lot more to gain having Windows on Alpha than MS did,a$ > and should have acted accordingly. >   D     Perhaps they saw their was little point in doing so. Without anyE indication from MS that they saw Windows on Alpha as a good thing whycC would Compaq expect a significant number of customers to buy it. IfnB nobody's going to buy it anyway what's the point of supporting it?   >>L >>       Without a volume market for Alpha there's no way Compaq ( or anyoneM >> else ) could afford to put the dollars into research necessary to maintaint# >> technical superiority for Alpha.  > B > Sort of like there's no way IBM can justify continuing the PowerK > architecture, or SUN justify SPARC (frankly, even *I* would be willing towN > question the latter, but SUN manages to make it work quite well, thank you)? > I      That's a good question! I don't know much about Power. I thought SuntK was having problems getting the new SPARCs out, but I admit that's hearsay.uI Perhaps the corporate culture at Sun has a lot to do with it also - ScottiG McNealy strikes me of much less of a beancounter than the folks running G Compaq and Sun has bet their entire business on SPARC whereas Alpha hastH been the poor stepsister ever since Compaq bought Digital. Don't forget,I Compaq is first and foremost a hawker of Intel boxes running MS software,uC when times are troubled companies fall back on what they know best.r  I      Please note I'm not saying Alpha couldn't have succeeded, just that iC the events of this week are not surprising given the way Compaq hasn treated Alpha.    F > couple of orders of magnitude more revenue - and offers at least theL > possibility of generating another order of magnitude more if the resultingN > leadership (coupled with something more than minimal marketing by its owner)J > causes it to take over *major* market share rather than just remaining a > popular boutique product.. >   H      OK, lets agree that Alpha is better than IPF and that VMS is betterI than Windows. What is your marketing plan to convince the masses of that?   I    At some point the old "if you can't beat em, join em" adage comes into  play.f  ) >  Without that lead Alphas current glory K >> in tht HPTC market would be short lived ( number crunchers have no brand=B >> loyalty - they'll always buy the fastest box they can afford ). > L > By all reasonable accounts, that would have been Alpha for the foreseeable	 > future.e  D     The benchmark I saw showed that the current Itanium is faster onG floating point than the fastest shipping Alpha. Granted this lead wouldxB change with 1GHz Alphas, but we're also constantly told what a dogF Merced is and how much better McKinley will be. This may or may not beF true, but the masses are going to believe it and buy Merceds ( or holdJ off buying anything ) on the promise that in a year or two they'll be ableI to upgrade to McKinleys that will blow away anything the competition has.d  F >  And while that's probably not all that an important market in terms+ > of revenue, it does help with mind-share.s >   H     It may not be that important, but it's one of the few markets CompaqM seems to have a chance of winning these days ( other than us VMS loyalists ).A  5 >>      Given that Intel isn't likely to adopt Alpha,E >  > No argument there. >  >  and that Compaq couldn'tG >> afford to keep up,P > 
 > Balderdash.1 > H      Your opinion. Compaq felt differently, in the end we have to accept6 that and carry on with what we have left to work with.  5 > it seems that moving VMS to IA64 is the best way toM >> insure it has a future. > M > Not only is one of your premises above flawed, but neither continuing AlphaCM > development, nor porting VMS to IA64, nor even *both* would be *sufficient* M > to 'insure' VMS's future:  Compaq would still have to show some interest inVK > the product, which it never has nor shows any sign of doing in the future M > (hey - you even say something like that below!  another post that can close0 > with agreement!).N >   I    Let me add to the agreement by saying I don't think there was anythinggE inherently wrong with Alpha, but that I felt it was doomed because ofaE management decisions made over the years - starting at Digital in theiL Palmer era and carrying on into Compaq. You're quite right that things couldH have turned out differently, but eventually the cumulative effect of tooF many wrong decisions makes it impossible to turn things around. CompaqE apparently felt they reached that point last week, I'm afraid I can'tr disagree with them.x   ------------------------------  # Date: Fri, 29 Jun 2001 00:25:08 GMTn4 From: "Terry C. Shannon" <terryshannon@mediaone.net>) Subject: Re: Full port of VMS to Itanium. ; Message-ID: <EVP_6.1951$UT1.992680@typhoon.ne.mediaone.net>e  > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message& news:zxBLxECW9Hy8@malvm5.mala.bc.ca...* > In article <9hg43n$rr4$1@pyrite.mv.net>,/ >     "Bill Todd" <billtodd@foo.mv.com> writes:e >d2 > >>       The fate of Alpha was sealed years ago. > >eL > > This ridiculous assertion just refuses to disappear.  But so do I:  it's too-- > > much fun being part of the 'truth squad'.  > >2I >     Have it your way, the fact remains Compaq has abandoned Alpha, muchB= > as many predicted they would given it's failure to become a4F > commodity item. You may feel Compaq should have continued developingD > Alphas strictly for the Tru64 and VMS market ( the only places its= > use was assured ), but they apparently don't feel that way.o >HD >     My assertion is that if Windows on Alpha had been a success we wouldn't > today be arguing about this.  C A good guy to talk to about this would be Bill Strecker, the Palmero? underling who negotiated the "organ donor" deal with Microsoft.i   ------------------------------  % Date: Thu, 28 Jun 2001 22:28:03 -0400e' From: "Bill Todd" <billtodd@foo.mv.com>l) Subject: Re: Full port of VMS to Itanium.s( Message-ID: <9hgpj5$3lk$1@pyrite.mv.net>  > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message& news:zxBLxECW9Hy8@malvm5.mala.bc.ca...* > In article <9hg43n$rr4$1@pyrite.mv.net>,/ >     "Bill Todd" <billtodd@foo.mv.com> writes:l >:2 > >>       The fate of Alpha was sealed years ago. > >tL > > This ridiculous assertion just refuses to disappear.  But so do I:  it's tooi- > > much fun being part of the 'truth squad'.r > >2I >     Have it your way, the fact remains Compaq has abandoned Alpha, muchf= > as many predicted they would given it's failure to become ao > commodity item.9  L Rather, given Compaq's failure to make any attempt to make it anything closeH to a commodity item, even in the limited higher-than-desktop 'commodity' sense.  J Note that IA64 isn't a commodity item either:  in fact it is not yet *any*K kind of an item (save for a units supplied to people so that they can startDH developing things on it) - and shows little sign of becoming much of anyD kind of an item until McKinley ships, since Itanic/Merced is so, er, uninspiring.  5  You may feel Compaq should have continued developingeD > Alphas strictly for the Tru64 and VMS market ( the only places its= > use was assured ), but they apparently don't feel that way.g  ( One might infer that from their actions.  F Of course, it's kind of difficult to make comparative statements aboutF IA64's assured markets since it isn't used *anywhere* at the moment inH measurable quantities (nor does it seem very likely to be until McKinleyJ appears next year) and the only OSs ready or near-ready to support it haveJ other platforms they can run on (unless you call 32-bit Windows and 64-bitE Windows two separate OSs - and even 64-bit Windows likely has another C platform it could run on if Microsoft were inclined to support it).S  K Fortunately, we don't have to be able to evaluate IA64's potential in orderlL to be able to project whether Alpha *already* commands sufficient markets toI make further development viable.  Alpha owns the VMS market (which itselfrH has been large enough to easily fund on-going Alpha development over theK years, despite active attempts by DEC to kill VMS and at an at-best neutraleG attitude toward VMS - read:  still no marketing - by Compaq), the Tru64 L market (which though modest - perhaps due to historical qualms about awkwardI Ultrix -> OSF and VAX -> MIPS -> Alpha transitions in the past - has beenmK growing at a respectable rate, at least until Monday), and the lion's shareoI of the 64-bit Linux market (whose potential is difficult to evaluate, buts non-negligible).  H So it seems clear that with even modest encouragement from Compaq (read:L visible development and marketing) VMS alone (with the potential for severalL billion USD annually in *profit* if any real marketing were performed) couldK sustain Alpha development (Terry's estimate of $250 million annually soundse7 credible) if Compaq had any interest in Alpha *OR VMS*.e  C The lack of such even modest support for VMS over the past 3 years,eI conversely, indicates that Compaq has absolutely no long-term interest insE VMS, and that the current 'committed' porting effort exists solely to*J protect short-term VMS revenue against the drastic reduction that dropping  Alpha would otherwise guarantee.   > D >     My assertion is that if Windows on Alpha had been a success we wouldn't > today be arguing about this. >-$ > >  If Microsoft had actually givenJ > >> a d**n about producing an OS not tied to a single processor ( as they > > claimed # > >> when they first announced NT )$ > >iG > > As time went on, the value of other processors to MS was seen to berI > > marginal, and MS not-entirely-unreasonably decided that the owners oft thoseo. > > processors should pay their support costs. > I >     Why was the ability of Windows to run on other processors ever seen & > as a useful goal then? What changed?  I Pre-NT-release market guesses turned into post-NT-release market reality.  It didn't happen overnight.    >eC >     How much does Intel pay each year to support Windows on theirtE > processors? Maybe you can argue they shouldn't have to since x86 is G > the de-facto Windows standard, but what about 64 bit Windows? If that,F > needs to be developed from the ground up anyway why should Intel get& > any different treatment than Compaq?  L Perhaps because Intel is a far more credible long-term supplier than Compaq,I given Compaq's obvious attitude (certainly since Pfeiffer's departure, ifyH not before) that Alpha is far from a bet-the-company product for Compaq?  I Microsoft has learned over time that it doesn't really need more than onewJ platform for Windows, as long as that platform's acceptance is rock-solid.H It was willing, probably even happy, to work with Compaq to create AlphaI support, but if Compaq wasn't interested, why should Microsoft have been,rE given that Intel can pretty much guarantee at least *adequate* marketaH acceptance of IA64 (and one platform is after all easier to support than two)?d   >r# > >Several owners decided that this H > > was not worthwhile for them, either (including, eventually, Compaq). > >o >(I >     It was MS's lacklustre commitment to Alpha that killed NT on Alpha.lG > You couldn't even get a full set of "MS Office" software for it - leti aloneoI > all the MS backend utilities like Exchange server, SQL server, etc. Howa could G > one expect any of the 3rd party vendors to support Alpha when even MSo& > obviously had little interest in it.  J You just don't get it.  Microsoft didn't need Alpha, Alpha needed Windows.H If Compaq hadn't blown it about 2 years ago, Alpha would have had 64-bitK Windows plus the important associated Microsoft products for something like I a year already (Microsoft was in no screaming rush to get them out before H IA64 arrived, but 64-bit Windows *was* out in the field being tested twoH years ago and stretching out that field test until today would have been! embarrassing even for Microsoft).   H Has the absence of such support on Alpha hurt Microsoft in the meantime?K Hard to see how - it gave them a lot of breathing room to get a more stable B 64-bit product ready, in fact, and the slower-than-expected marketL acceptance of Win2K suggests that enterprise sales of 64-bit Windows systemsG would have been limited.  But has the absence of 64-bit Windows supportaL (combined with the blow Compaq administered to 32-bit Windows on Alpha) hurtK the market's perception of Alpha's future viability over that period?  Duh.o  H If Compaq was interested in Alpha, it was Compaq's job to make damn wellK sure that it gave Microsoft the support it needed to get Windows - both 32-NK and 64-bit - there.  Instead, it fired a double-barreled blast into Alpha's H feet two years ago (the second chamber was Microsoft's, but Compaq couldK have averted that and didn't) - which in retrospect raises doubts about theoI level of its commitment to Alpha even then, despite the last 22 months ofu$ continued reassurances in that area.   >p > >hH > > Compaq always had a lot more to gain having Windows on Alpha than MS did,& > > and should have acted accordingly. > >s >t: >     Perhaps they saw their was little point in doing so.  = Certainly not if they were already planning to scuttle Alpha.l    Without anyG > indication from MS that they saw Windows on Alpha as a good thing why B > would Compaq expect a significant number of customers to buy it.  K That hadn't been much of a problem up to that point, even lacking all thosewJ applications:  there was a clear market, though the larger opportunity wasJ the one of having exclusive use of 64-bit Windows for a significant periodL of time (since Alpha running 32-bit Windows didn't really have that distinct& an advantage over IA32-based Windows).  K Nonetheless, Microsoft's level of support for Windows up to the time of theoJ acquisition was enough for the customers - about 15% of total Alpha sales,J according to Terry (though more applications would have likely resulted inI more sales).  If customers had a confidence problem, it was with Compaq's I commitment, not Microsoft's:  8 months after the acquisition, NT on Alpha ' sales had dropped to 2% of Alpha sales.@    IfsD > nobody's going to buy it anyway what's the point of supporting it? >  > >>G > >>       Without a volume market for Alpha there's no way Compaq ( or. anyoneF > >> else ) could afford to put the dollars into research necessary to maintain% > >> technical superiority for Alpha.l > >sD > > Sort of like there's no way IBM can justify continuing the PowerJ > > architecture, or SUN justify SPARC (frankly, even *I* would be willing toJ > > question the latter, but SUN manages to make it work quite well, thank you)?o > >mK >      That's a good question! I don't know much about Power. I thought SunlD > was having problems getting the new SPARCs out, but I admit that's hearsay.K > Perhaps the corporate culture at Sun has a lot to do with it also - Scott I > McNealy strikes me of much less of a beancounter than the folks runningiI > Compaq and Sun has bet their entire business on SPARC whereas Alpha hasoJ > been the poor stepsister ever since Compaq bought Digital. Don't forget,K > Compaq is first and foremost a hawker of Intel boxes running MS software,eE > when times are troubled companies fall back on what they know best.  >eJ >      Please note I'm not saying Alpha couldn't have succeeded, just thatE > the events of this week are not surprising given the way Compaq hasu > treated Alpha.  K Well, blow me down and call me scuttlebutt!  Why didn't you put it that way  in the first place?    >r > H > > couple of orders of magnitude more revenue - and offers at least theD > > possibility of generating another order of magnitude more if the	 resultingaI > > leadership (coupled with something more than minimal marketing by itsn owner)L > > causes it to take over *major* market share rather than just remaining a > > popular boutique product.. > >u >eJ >      OK, lets agree that Alpha is better than IPF and that VMS is betterK > than Windows. What is your marketing plan to convince the masses of that?s  J Raising the level of marketing activity - even otherwise prosaic marketingI activity - above zero doesn't take all that much planning and would be an  excellent starting point.y  I That would at least let the people who already appreciate VMS (and likelyhF Tru64 to some extent) see *something* in the way of Compaq commitment:J given how many people have stuck with VMS *despite* Compaq's utter lack ofG demonstrated commitment to it, there must be a similar number out theremK who'd still be willing to hitch their own futures to VMS if Compaq's futurer seemed more tied to it.W  D Note, incidentally, that the current situation decreases rather thanL increases VMS's importance to Compaq's future (and I can hardly believe thisA is an accident):  instead of having a group of three products allaK significantly inter-dependent (Alpha, VMS, and Tru64), Compaq will now have(G only two completely independent products (VMS and Tru64), making it farnJ easier to drop one if it seems convenient without impacting other markets.  K Start increasing VMS's market (which should be really easy, given how smallhL it is) and you start getting at least a bit of attention.  Do enough to makeH it appear that Compaq *really* is committed to VMS, and you could likelyK (before last Monday) at least triple VMS revenue (there really are a lot oftL people out there who haven't forgotten it, even though many of them may well not know it's still available).e  J That's as far as I'll go:  I'm not a marketeer, nor do I have any interestI in becoming one - and tripling a $4 billion/year market is enough for one H post, anyway.  But I'm reasonably sure a good marketeer could do a great
 deal more.   >sK >    At some point the old "if you can't beat em, join em" adage comes intop > play.e  K But usually only after you've first made at least some attempt to beat 'em.t   >n+ > >  Without that lead Alphas current glorygG > >> in tht HPTC market would be short lived ( number crunchers have no  brand D > >> loyalty - they'll always buy the fastest box they can afford ). > > B > > By all reasonable accounts, that would have been Alpha for the foreseeable  > > future.V >iF >     The benchmark I saw showed that the current Itanium is faster on1 > floating point than the fastest shipping Alpha.n  H Was that the one they later changed?  My impression was that the currentG stats showed some flavor of 833 MHz Alpha marginally ahead of Itanic onnL floating-point (and *way* ahead on integer), but I don't keep careful track.  I As for the future, go read the architectural prognostications and make upt" your own mind:  that's what I did.    Granted this lead wouldD > change with 1GHz Alphas, but we're also constantly told what a dogH > Merced is and how much better McKinley will be. This may or may not beH > true, but the masses are going to believe it and buy Merceds ( or holdL > off buying anything ) on the promise that in a year or two they'll be ableK > to upgrade to McKinleys that will blow away anything the competition has.   J I wouldn't have expected Alpha and VMS to take the world by storm over theL next year:  tripling VMS's revenue would have been a nice start, and made itF clear that IA64 did not have the larger field (only a relatively small< portion of which would put Compaq on Easy Street) to itself.   >pH > >  And while that's probably not all that an important market in terms- > > of revenue, it does help with mind-share.l > >a >tJ >     It may not be that important, but it's one of the few markets CompaqB > seems to have a chance of winning these days ( other than us VMS loyalists ).  G Winning would likely be easier if actual attempts to compete were made.t   >k7 > >>      Given that Intel isn't likely to adopt Alpha,w > >M > > No argument there. > >  > >  and that Compaq couldn'tr > >> afford to keep up,n > >r > > Balderdash.C > >eJ >      Your opinion. Compaq felt differently, in the end we have to accept8 > that and carry on with what we have left to work with.  L We certainly have to accept it, but we equally certainly don't have to agree with it, or defend it.   >y7 > > it seems that moving VMS to IA64 is the best way to  > >> insure it has a future. > > I > > Not only is one of your premises above flawed, but neither continuingn AlphaeB > > development, nor porting VMS to IA64, nor even *both* would be *sufficient*L > > to 'insure' VMS's future:  Compaq would still have to show some interest inF > > the product, which it never has nor shows any sign of doing in the futureI > > (hey - you even say something like that below!  another post that cani closes > > with agreement!).u > >o >qK >    Let me add to the agreement by saying I don't think there was anythingaG > inherently wrong with Alpha, but that I felt it was doomed because of0G > management decisions made over the years - starting at Digital in theuH > Palmer era and carrying on into Compaq. You're quite right that things couldoJ > have turned out differently, but eventually the cumulative effect of tooH > many wrong decisions makes it impossible to turn things around. CompaqG > apparently felt they reached that point last week, I'm afraid I can'td > disagree with them.f  K I clearly can, and do - aside from any suspicions that they might have beentI planning on this for far longer.  This industry is in many ways amazinglytL tolerant of incompetence and false steps, as long as the light is eventuallyB seen and something is done *right* that helps correct past wrongs.  K By contrast, when one blunders toward oblivion, step after step, for years,iL and then says, "We just have to take these actions, given the circumstances"F (said circumstances being wholly of one's own creation and yet *still*F leaving room for recovery if only one bites the bullet and admits past9 screw-ups), there's little room for anything but disgust.u   - bill   ------------------------------  % Date: Fri, 29 Jun 2001 12:38:57 +1000"- From: "Paul Nankervis" <paulnank@au1.ibm.com>-) Subject: Re: Full port of VMS to Itanium. / Message-ID: <9hgqe7$7g3a$1@poknews.pok.ibm.com>   : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3B3AB68A.3222E975@videotron.ca... > Paul Nankervis wrote:.L > > Give it a go! Think **positive** and PROMOTE  VMS!  After all the peopleK > > in this group influence far more people than Compaq marketing ever willl (!). >>L > You see this an an investment in VMS. Technically yes. But politically no.G > Compaq is spending a bit fo dollars in order to allow the sale of thebL > architecture that will give it a whole wad of money. The whole impetus forJ > this move is not to push VMS but rather to get rid of Alpha. The Port of VMSa1 > is a necessary evil to allow the sale of Alpha.i > L > If Compaq had though that VMS on IA64 would have been great, how come theyL > didn't announce the port a long time ago, as it had announced for True64 ?  J I kind of agree with you on this. I spent some time posting on the subject
 of portingE VMS to other platforms years ago, and typically got back rather shortg sighted-F responses about how Alpha was the premier performance platform, and it	 thereforesJ wasn't worth porting VMS to any other platform. What a turn around!!! I DO= think that Digital/Compaq have been hopelessly short sighted!n  L I really wish that I had kept the note from Digital which said "that Digital was a D hardware company who only provide operating systems in order to sell
 hardware".K As a customer I had real problems with this as my ONLY interest was in VMS.iL I did not care whether it was on VAX or Alpha or whatever - provided that it? had all the features of VMS and a reasonable price/performance!o  J > Porting VMS isn't enough. It needs to be marketed especially in a period ofK > uncertainty. If VMS continues to be neglected by Compaq from the point of  viewL > of Marketing, then the port to IA64 will be meaningless and without effectD > because nobody will notice and Compaq won't start to push VMS as a solution
 > against NT.r  G Well marketing isn't enough! You can try and sell something, but if thek customerJ doesn't want it then it will not sell. What does matter is how people like you andrK me talk about VMS. If your boss comes in and asks what platform you thoughtaL was suitable for a new project, and you give back some dialogue about how itJ should have been VMS but it is not marketed properly - then he is going to thinksF about using Windows or Unix (and wonder what drugs you are on!) If you respond K that VMS has all the reliability features and tools that are needed then hea	 has to ati least consider it!  L Compaq have a responsibility to market ALL of their products if they want toK make money (which I'm sure that they do!). But ultimately it is going to beg the J customer who decides what gets installed. Personally I think Compaq shouldK focus on putting a few more decent tools onto VMS - it would be really neatwI to be able to use VMS to provide ISP services for example. Some jobs suchiK as this default to Windows/Linux when the reliability of VMS would be nice!*  L It would be interesting to compare the marketing approaches of the differentL computer companies. I currently live in a small country city where we do notG see computer advertisements in papers or on TV for *ANY* vendor. I note I that whenever I travel I see lots of stuff for IBM and Microsoft. In thiso contextwE VMS and Compaq don't seem any better or worse off than for most otherd computer products!  J As for my view on porting to Itanium it looks like this is an injection of new H money for VMS which at the very least is going to give customers a widerK choice of platforms to run on.   I only hope that they consider the portingoA issues carefully because with luck VMS will outlive Itanium too!!l   Paul Nankervis   ------------------------------  % Date: Thu, 28 Jun 2001 22:23:01 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>l) Subject: Re: Full port of VMS to Itanium.v' Message-ID: <3B3BF495.F1A74F39@fsi.net>h   "Terry C. Shannon" wrote:e > @ > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message( > news:zxBLxECW9Hy8@malvm5.mala.bc.ca..., > > In article <9hg43n$rr4$1@pyrite.mv.net>,1 > >     "Bill Todd" <billtodd@foo.mv.com> writes:t > > 4 > > >>       The fate of Alpha was sealed years ago. > > >bN > > > This ridiculous assertion just refuses to disappear.  But so do I:  it's > toou/ > > > much fun being part of the 'truth squad'.a > > >eK > >     Have it your way, the fact remains Compaq has abandoned Alpha, muchi? > > as many predicted they would given it's failure to become arH > > commodity item. You may feel Compaq should have continued developingF > > Alphas strictly for the Tru64 and VMS market ( the only places its? > > use was assured ), but they apparently don't feel that way.c > >cF > >     My assertion is that if Windows on Alpha had been a success we
 > wouldn't  > > today be arguing about this. > E > A good guy to talk to about this would be Bill Strecker, the PalmertA > underling who negotiated the "organ donor" deal with Microsoft.n   Yeah - "brain donor" as well!r   -- p David J. Dachtera  dba DJE Systemsn http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/s  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.a   ------------------------------  % Date: Thu, 28 Jun 2001 23:43:14 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>n) Subject: Re: Full port of VMS to Itanium.m, Message-ID: <3B3BF949.A576C3AA@videotron.ca>   Bill Todd wrote:L > You just don't get it.  Microsoft didn't need Alpha, Alpha needed Windows.J > If Compaq hadn't blown it about 2 years ago, Alpha would have had 64-bit > Window     My take of NT on Windows:o  N Initially, Microsoft wanted NT on many platforms, especially the ones that had4 the potential to go high end computing (aka: Alpha).  J I think that Microsoft gave Alpha a good chance because it felt it was its& door to high end enterprise computing.  N Unfortunatly, neither Digital nor Compaq decided to build/market Alpha systemsL as part of their mainstream products and Digital in fact prevened Alpha from! competing against wintel systems.   J It is no surprise then that Microsoft would have pulled the plug on NT andL told Compaq that if it wanted NT on Alpha, Compaq would have to pay for it.   G But had Compaq seen the potential of Alpha and begun to push Alpha as amK "industry standard" machine and sold both low end and high ens servers, and H perhaps high end workstation, I have a feeling that Microsoft would haveF provided much better support for Alpha-NT including more applications.  H It is pretty obvious why MS didn't port Office since neither digital norI compaq wanted to produce low end workstations where Office would be used.    ------------------------------    Date: 28 Jun 2001 17:55:09 -0400/ From: mark@hubcap.clemson.edu (Mark Smotherman)   Subject: Re: IA64 Rocks My World- Message-ID: <9hg93t$cof$1@hubcap.clemson.edu>i  + John Bayko <jbayko@sk.sympatico.ca> writes:sA >    The Intel i960, at one point the world's most popular 32-bit H >embedded CPU, was tossed without regard when Intel switched to ARM. ARMF >may have been a better architecture, but the i960 was not at all bad,D >and had much life left in it - even superscalar versions, which ARM >doesn't have.  L The i960CA was actually the first single-chip superscalar processor in 1990.  H I'm not completely familiar with the ARM family, but a blurb in May said6 they are working on a superscalar core called Cheetah.   -- dH Mark Smotherman, Computer Science Dept., Clemson University, Clemson, SC- http://www.cs.clemson.edu/~mark/homepage.html    ------------------------------    Date: 28 Jun 2001 16:15:17 -0700/ From: Brannon_Batson@yahoo.com (Brannon Batson)-  Subject: Re: IA64 Rocks My World= Message-ID: <4495ef1f.0106281515.557c1454@posting.google.com>r  j bengtl.net@telia.nospam.com (Bengt Larsson) wrote in message news:<3b3b4181.30126558@enews.newsguy.com>...D > In comp.arch, Eric Smith <eric-no-spam-for-me@brouhaha.com> wrote: > S > >Casper.Dik@Holland.Sun.Com (Casper H.S. Dik - Network Security Engineer) writes:uC > >> Engineers shouldn't blame everything on marketing; Digital hast= > >> made serious mistakes when it came to building hardware.  > >h > >Such as?o > " > - Not building large enough SMPs   EV7l  5 > - Not focusing on memory performance (esp. latency)n   EV7B  A > - Aversion to obsoleting their own systems (esp. the 8200/8400)S  ? What do you mean?  Is Digital too bad about supporting existinge customers with legacy hardware?k   Brannon,   ------------------------------  % Date: Fri, 29 Jun 2001 00:51:42 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>e Subject: Inform comedy...., Message-ID: <3B3C0950.4FD154AF@videotron.ca>  5 I think they should rename the magasine "MAD" :-) :-)   K Inform Magasine, Canadian edition (Summer 2001) received today in the mail.   > Alphaserver mentioned in 2 colums in the first compaq advert.   8 Page 8 features article about Compaq-Microsoft alliance.     page 12:I AlphaServer systems power world class research (Autralian partnership for 8 advanced Computing) who ordered 450 AlphaServer systems.    G Page 54 has a 2 page article about AlphaServer TS for telecom industry.tK (only tyru64 is mentioned in the supported OS on that page, but there is ann interesting text:uN 	"...TS40 systems combine Compaq's powerful industry-standard Alpha technology# with (telephony software products)"e  K Interesting use of "industry standard". Perhaps a last attempt by the alphaaK group to convince Compaq that Alpha could be an industry stanard had it not  been squandered :-(r  K Page 56 lists some of the "fastest CPU" benchmarks acheived bty the 833 mhza8 Alpha, as well as design awards for alphaserver systems.  @ Page 57 contains an ad for Oeek&Spy, a third party VMS software.  M Page 59 mentions the $7 million contract with the Qubec govermmeht to supplyhL Alphaserver systems that shoudl regroup a lot of the government's IT needs .J (Trucluster and Tru64, Oracle and will serve up to 66,000 govt employees).  N Then  announcement of a deal with Entrust (the proprietary encryption company,L once used by a canadian bank) that will port its software to Tru64 on Alpha.  L There is also some bragging about how many SAP systems based on Alphaservers) and Tru64 Compaq has sold. (over 10,000).s  K Page 62 has article about Life Sciences revolution and how AlphaServers arem0 deployed in large numbers at Sanger, Celera etc.  K Interesting sentence: "The joint R&D agreeement will use future versions of K AlphaServer supercomputer systems. The Goal is to create a new prototype bye@ 2004". (this is part of a project to have a 100 TeraOPS system).  - Sanger Centre in europe has 590 Alphaservers.f  K "The leader in market revenue for the technical systems and servers market,mM Compaq has won almost every major supercomputing contract since the beginninga
 of 2000."   N The above sentence makes it hard to explain why Compaq would so suddently drop	 Alpha :-)i      D Page 66 has article on how Compaq has 65% of the world's electricity distribution computing. J "Utilities that installed applications on the OpenVMS operating systems 20I years ago have been able to scale through VAX and the AlphaServer productV0 families without re-engineering their software".   2 paragraphs later:oJ "Most Compaq-base utility projects installed over the past 5 years utilize< Compaq AlphaServer servers running the Coimpaq Tru64 Unix...  L Article about hydro Qubec. Only mentions AlphaServers and how they are able& to keep costs down. No mention of VMS.  M There is a 2 page article about VMS being used for an agriculture applicationiN in england. Again, Alphaserver's advantages are touted. Interestingly, again a8 mention of how easy it was to migrate from VAX to Alpha.  L There is also an artocle about how a Winnpieg Health Care organisation choseM Alphaservers and OpenVMS. One of the advantages touted was the ability of theyE wildfires to run different operatin systems so they can run differenta" applications on the same "system".   E*TRADE:  I There is a 2 page article about ETRADE. Most about the PCs featured it istN "public store" in NYC, but a brief mention of all the AlphaServers used to run, its enterprise computing. No mention of VMS.      M I find it quite interesting that the Canadian edition would have shipped thisgN edition at such a wrong time, and I find it most interesting that this editionJ seems to be the first one to feature Alpha so prominently and showcase its successes so much. e  N Obviously, the canadian editor was unaware that Compaq would be killing Alpha.   ------------------------------  % Date: Thu, 28 Jun 2001 23:34:04 -0400o  From: John Santos <JOHN@egh.com>2 Subject: Re: lsedit or whatever xwindow on my pc ?6 Message-ID: <1010628232019.61768A-100000@Ives.egh.com>  ) On Thu, 28 Jun 2001, Chris Sharman wrote:t  K > For my sins (apparently multitudinous) I have a new desktop, owing to thecB > need to read/write Office compatible documents & browse the web. > L > What's the best (cheap/free) way to run an xwindow on my PC, running, say, > lsedit, or perfmeter.l > ; > Or would there be a better place to ask these questions ?a > L > Of course, it may all be immaterial - I was planning to place an order for" > new Alpha/VMS boxes next week :( > 	 > Thanks,e > Chris Sharmanf  A Maybe this will cause a storm of protest, but how about the otherb way around?a   1) Wait 3-4 years. 2) Buy an IA-64 VMS system. @ 3) Buy Softwindows-64 (or what ever they decide to call it.) [1]< 4) But off-the-shelf Win-64 versions of whatever you need at?    CompUSA (Office, Quicken, Netscape, Flight Simulator, Doom).l4 5) Run them in the Softwindows-64 sandbox under VMS.? 6) Run multiple virtual Softwindows so Office can't report back ?    to Redmond that you are running Quicken and thus subject yout5    to massive "MS Money is better" spam attacks.  [2]i  @ [1] Remember Softwindows (or was it called SoftPC?) - We used to= have a demo version of this on our first VMS Alpha.  It was ae= PC (not windows) emulator that you could run the then-currentw? Windows (3.1?) under.  It worked okay, but was slow as molassesf= because it was emulating the x86 instruction set in software.   ? There is now something like this available for Macs.  The womans@ in the office next to mine used to have it on her Mac (G3?), and> I guess was happy with it, at least to the extent she could be  happy with anything Windows-ish.  ; On an IA-64 VMS system, it should be as fast as native moden= because it *would* be native mode (not emulated in software.)   ? [2] Since paranoia rules the day, I thought I had to contributea	 a little.    -- t John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Thu, 28 Jun 2001 17:45:21 -0400a* From: "Joshua Harding" <jharding@sscc.org>- Subject: Re: My take on the Compaq-Intel deal / Message-ID: <tjn92tftgd9ta4@corp.supernews.com>f  I In recent times, I've seen 100x the amount of porting _AWAY_ from OpenVMS K vs. porting _TO_ OpenVMS. So, JF, I accept your analysis as correct. Compaq'H will "unadvertise" VMS until it is long past dead. Eventually there is aK point where the user community cannot sustain their parent company anymore.sG One does not have to look much farther than the Amiga crazies to find a D fanatical fanbase clustered around a very cold and very dead corpse.  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3B3B991A.D807E52D@videotron.ca... > BK Toh wrote:hL > > Before this announcement, many ISVs would not have been keen to port any! > > applications to OVMS nor T64.  > F > Why would they not port ? Because VMS is a small market of remainingK > customers, with Compaq making no credible commitment to expand and markets this> > product and make it part of its maintream product portfolio. >0L > The only somewhat credible saying from Compaq is that it would continue toJ > support VMS for some time. But god only knows what went on behind closed doors K > in terms of strategy to get rid of VMS which doesn't fit in Compaq's core2	 strategy.o >nL > I do not beleive that the recent announcements show Compaq changing any of its K > commitment to VMS nor Compaq changing in any way its marketing commitmentC ofI > VMS. They are just taking money from Intel to port VMS to Intel so thate IntellI > can claim monopoly in the high end chip. I do not beleive that Compaq'st) > handling of VMS will change in any way.a >iJ > So if you look back at how Compaq handled VMS in the last few years, youG > should, in my opinion, see how Compaq will handle VMS in the next fews years:D > VMS will remain a hidden product known only to the installed base. >6J > There is an additional compelling reason for VMS not to be advertised in theoK > next few years: Until it runs on IA64, it is pointless to spend any moneynJ > trying to gain new customers on VMS since they would not spend any money on a > dead platform (Alpha). > K > Compaq is working hard to gain new revenus trhough the services/solutionsoJ > project. But its VMS plans haven't really changed. Just maintain VMS and that > is what they are going to do.t >C >aJ > Now, if Compaq were to come out and drop Tru64 and announce the same day thatL > it has gotten SAP to port to VMS, I might beleive that things would reallyK > change for VMS and that it might becomes a real Compaq product. But untilo thatI > happens, VMS remains a Digital product that Compaq maintains because itn has to3 > maintain it, not because it wants to maintain it.t   ------------------------------    Date: 28 Jun 2001 22:30:10 -0400/ From: jordan@lisa.gemair.com (Jordan Henderson) ? Subject: Origin of iVMS? (was Re: Full port of VMS to Itanium.)m* Message-ID: <9hgp7i$e1l$1@lisa.gemair.com>  8 In article <ckjmjtkv2ato1dlaoigviim8ske4totlss@4ax.com>,' Alan Greig  <a.greig@virgin.net> wrote:aA >On 28 Jun 2001 09:17:03 -0500, Kilgallen@eisner.decus.org.nospamr >(Larry Kilgallen) wrote:n >y^ >>In article <3B3AAC72.2E1EE619@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >>> Keith Brown wrote:P >>>> > Could someone please post some link to an official Compaq statement aboutG >>>> > that iVMS thing and what it will actually do that VMS won't do ?u >>>> > >>>> d= >>>> http://www.compaq.com/newsroom/pr/2001/pr2001062501.htmln >>> ' >>> That page has no mention of "iVMS".w >>= >>Perhaps because "iVMS" is a name made up in this newsgroup.e >l$ >But first used by Hoff I believe...  B Thank goodness for dejagoogle ( http://groups.google.com/ ), it's 6 searches and "Sort by date" to answer these questions.  E The first use of iVMS I can find in comp.os.vms was totally unrelatedo= to the current Itanium discussions, but wasn't that long ago.y  I Christopher Smith, in a discussion about dropping the _OPEN_ in OpenVMS, t on June 7, 2001 said:   4 > Maybe "AjarVMS," since it's not quite "open"... ;) >  > eVMS?s > F > iVMS?  (I do have a VAXStation 2000 at home attached to an LSI ADM-5. > terminal with a sign that says "iVAX" on it) > 
 > Regards, >  > Chrisd > # > Christopher Smith, Perl Developer  > Amdocs - Champaign, IL >  > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");   @ In the current discussion, it appears that the first use was by E John B. Macallister, on June 26th, answering fabio_compaq's concerns n about the OpenVMS job market:u  I > >I believe I will be out of this newsgroup soon. I cant trust my careero
 > >anymore > >in OpenVMS. c > M > If you change platforms make sure you get on the right train. Any VMS trainoL > will get you to the right station. If you wait for an iVMS train you'll beN > able to get off at any station you want. The Windows 2000 train will explodeI > at an intermediate station if too many people get on and off. The Tru64eN > train will have to be built on the journey collecting pieces at each stationN > on the way: you won't be allowed to board unless you (a) have a new piece ofK > train in your luggage or (b) know how to assemble some of the parts lyingaM > all over the floor in each acrriage. The Solaris train will overshoot every J > station as it's just too fast for ordinary people and stations: your SUNG > salesperson will be able to tell you where the nearest SUN-compatiblesM > station is located but you won't be able to obtain that information at yourr5 > station as it does not have a SUN support contract.t >  > John > D > Name: John B. Macallister  E-mail: j.macallister1@physics.ox.ac.ukJ > Post: Nuclear and Astrophysics Laboratory, Keble Road, Oxford OX1 3RH,UKC > Phone: +44-1865-273388 (direct)  273333 (reception)  273418 (Fax)t  D But, Hoff's use was close to the same time, 7 hours later.  I can't E tell if the time listed by google is the time received at google, or aC the time from the header.  Either way, the time is open to dispute.   > I would have posted the links to the google pages, but they're? monstrous (really long and messy).  You can research the use ofME iVMS on comp.os.vms yourself by doing a search for "iVMS comp.os.vms" 7 at http://groups.google.com/ and select 'Sort by date'.A   >--l >AlanM   -Jordan Hendersong jordan@greenapple.come   ------------------------------  % Date: Thu, 28 Jun 2001 22:26:38 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>oC Subject: Re: Origin of iVMS? (was Re: Full port of VMS to Itanium.)c' Message-ID: <3B3BF56E.19265031@fsi.net>o   Jordan Henderson wrote:s > : > In article <ckjmjtkv2ato1dlaoigviim8ske4totlss@4ax.com>,) > Alan Greig  <a.greig@virgin.net> wrote:eC > >On 28 Jun 2001 09:17:03 -0500, Kilgallen@eisner.decus.org.nospamm > >(Larry Kilgallen) wrote:3 > >d` > >>In article <3B3AAC72.2E1EE619@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > >>> Keith Brown wrote:R > >>>> > Could someone please post some link to an official Compaq statement aboutI > >>>> > that iVMS thing and what it will actually do that VMS won't do ?e > >>>> > > >>>>? > >>>> http://www.compaq.com/newsroom/pr/2001/pr2001062501.html  > >>>h) > >>> That page has no mention of "iVMS".s > >>? > >>Perhaps because "iVMS" is a name made up in this newsgroup.n > > & > >But first used by Hoff I believe... > C > Thank goodness for dejagoogle ( http://groups.google.com/ ), it'sl8 > searches and "Sort by date" to answer these questions. > G > The first use of iVMS I can find in comp.os.vms was totally unrelated ? > to the current Itanium discussions, but wasn't that long ago.a > J > Christopher Smith, in a discussion about dropping the _OPEN_ in OpenVMS, > on June 7, 2001 said:  > 6 > > Maybe "AjarVMS," since it's not quite "open"... ;) > > 	 > > eVMS?i > >dH > > iVMS?  (I do have a VAXStation 2000 at home attached to an LSI ADM-50 > > terminal with a sign that says "iVAX" on it)  G IIRC, Chris used that line well before the Deja/Google debacle. He justl+ repeated himself in that post on that date.a   -- h David J. Dachtera  dba DJE Systems- http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/R  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.M   ------------------------------  # Date: Fri, 29 Jun 2001 02:33:14 GMT . From: "Duane Sand" <duane.sand@mindspring.com>? Subject: Re: Question to Charlie Matco - reply to Terry Shannonh@ Message-ID: <KNR_6.138833$%i7.92840739@news1.rdc1.sfba.home.com>   "Terry C. Shannon" wroteK > What would be useful is a detailed description of the VMS porting effort.eF > NSK and Tru64 can pretty much be taken for granted since preliminary Merced5 > ports of both OSes exist. VMS is a different story.h  K A rought plan for porting NSK to Merced some years ago exists.  Little codemH was written, and none survived the Intel lawyers.  The former plan won't workE as is, since we now have several fewer years available to do the job.l  I But we do have a sizeable expert team all set to go, from the interrupted8E port to Alpha,  and can recycle some of that source code, and recycle@ some of its detailed plan.  D Starting from zero and within corporate & team chaos with very tight8 time constraints would be much much harder.   Good luck!    L > Equally useful would be a detailed description of what components of Alpha= > get embedded in IA-64, and when. As near as I can tell, thes Alpha-betization? > doesn't take place until two IA-64 interations past McKinley.v  K My opinion: no such thing as Alpha-betization; certainly nothing that couldnJ help the os ports, beyond perhaps a GEM IL interface into Intel's compiler	 back end.r   ------------------------------  % Date: Thu, 28 Jun 2001 17:40:54 -0400 * From: "Joshua Harding" <jharding@sscc.org>Y Subject: Re: Question to Charlie Matco - reply to Terry Shannon (andsomegeneral comments)h/ Message-ID: <tjn8qhia4ldhc8@corp.supernews.com>   J yeah, I'd love to have a USB joystick on my VAX 4000 VLC for some good old- angband. tell me when the driver is finished.f  6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OFB87C08C2.0D01A1C3-ON03256A79.003DC96F@ep-bc.petrobras.com.br... > Well >hG > I just gave a suggestion and you explained everything in your answer:r > 2 > a) Kids which dont know how to develop a driver;F > b) Kids dont want to learn C and operating systems internals becauseI > they dont like to read "old books" and these informations are not readyt > over the internet; >r > L > PS: I am not a developer, but reading this NG and seeing a lot of requests > fornA > drivers (ex. DVD, CD-RW, USB), I believe more than 70% here are  developers.i > Why not they create ?l >e	 > Regardsy >a > FC >i >o >  >nE > rdeininger@mindspring.com (Robert Deininger) em 27/06/2001 16:31:05a >i@ > Favor responder a rdeininger@mindspring.com (Robert Deininger) >  >t >. >       Info-VAX@Mvb.Saic.ComO >& >, >tF > Assunto: Re: Question to Charlie Matco - reply to Terry Shannon (and  >          somegeneral comments) >l >s > In articleC > <OF8BB15099.76A624F6-ON03256A78.0056184B@ep-bc.petrobras.com.br>, , > fabio_compaq@ep-bc.petrobras.com.br wrote: >t > > Like to have for example:e >o > G > > c) A tool / compiler to develop OpenVMS Device Drivers more easily.n >aK > What do you find so hard about VMS device drivers?  This is an inherentlybL > complex problem.  You have to interact with the innards of the OS, and the > quirks of a hardware widget. >aE > Digital provided the tools (C compiler and extensive driver supportlH > routines and macros) the documentation (VMS manual set and the DigitalI > Press book "Writing OpenVMS Alpha Device Drivers in C") and examples ofeF > working drivers.  They've also provided extensive debugging support. >oH > You have to understand your device.  I don't see any "tool" that could > solve that for you.T >l7 > Then you have to follow the clearly documented rules.v >  > What's missing.s >nK > If there are fewer folks able to write drivers these days, I suspect it's > > because there are fewer good programmers than in olden days.L > Script-kiddies probably can't write decent device drivers on any platform. >tG > If you have ideas to make this complex task easier, I'm interested ine > hearing them.  >  > -- > Robert Deininger > rdeininger@mindspring.comt >c >  >o >  >  >    ------------------------------    Date: 28 Jun 2001 17:57:04 -05004 From: cornelius@encompasserve.org (George Cornelius)2 Subject: Re: Sending command to programs with pipe3 Message-ID: <0GchsB8pC1ND@eisner.encompasserve.org>U  c In article <9hfej4$dl8t4$1@ID-46415.news.dfncis.de>, "James Gessling" <jgessling@yahoo.com> writes:o > Greetings, > E > Ran across a cute trick to eliminate usage of temp file in dcl (onerH > of my pet peeves).  In sda, I wanted to find out a process logical for= > another process on the system.  Here's the snippet of code:m >  > $ say :== write sys$output8 > $ pipe (say "set erase off" ; say "set proc ''p1'" ; -. >         say "clue proc/log" ) | anal/sys | - >         search sys$input 'p2'y  = Glad to see someone else has found that pipe can simplify SDAf operations.c  0 A simpler example, here to display LAN adapters:     $ w:=write sys$outputAO   $ pipe w "sho lan/full" | anal/sys | sear sys$input "device name","line spee"n  ? Or clusterwide (my cluster_do does it the hard way, but a pipedm implementation follows):     $ cluster_do -@      "pipe ''w' ""sho lan/full"" | anal/sys | sear sys$input ...  G where a piped implementation of cluster_do.com would check that only P1 G is present (if more, you have your quotes wrong), and then do somethingc like     $ w:=write sys$outpute9   $ pipe ( w "set env/clus" ; w "do ''p1'" ) | mcr sysmanl   --9 George Cornelius              cornelius@encompasserve.org20                               cornelius@mayo.edu   ------------------------------    Date: 28 Jun 2001 17:58:21 -05004 From: cornelius@encompasserve.org (George Cornelius)2 Subject: Re: Sending command to programs with pipe3 Message-ID: <AZpnGWTl7m6p@eisner.encompasserve.org>h  j In article <fdd7874.0106280853.1ba54755@posting.google.com>, stephane_paquin@hotmail.com (SPaquin) writes:5 > What is pipe ? What VMS version needed ? Freeware ?r  = Implements Unix-style command line pipes.  VMS 7.1 and later.    --9 George Cornelius              cornelius@encompasserve.orgl0                               cornelius@mayo.edu   ------------------------------  # Date: Thu, 28 Jun 2001 23:47:23 GMT . From: "Alphaman" <alphaman64@nixspam-home.com>2 Subject: Re: Sending command to programs with pipe; Message-ID: <fmP_6.15568$P5.5129592@news1.rdc1.tn.home.com>o  . While on that topic, I've tried the following:  K  $ Pipe echo "This is the body text." | mail sys$input aaron /subj="testinga pipe"o  L This works great from the command line, but fails from a .com file (and yes,: I've tried it with Sys$Pipe, too.)  Instead, I have to do:  H  $ Pipe echo "This is the body text." > temp.file ; mail temp.file aaron( /subj="Testing pipe" ; delete temp.file.  K Has anyone got a better way of accomplishing this?  Why does it fail withinf" a .com file but not interactively?   Aaron( --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)t   ------------------------------  % Date: Fri, 29 Jun 2001 05:59:13 +0200u2 From: martin@radiogaga.harz.de (Martin Vorlaender)2 Subject: Re: Sending command to programs with pipe; Message-ID: <3b3bfd11.524144494f47414741@radiogaga.harz.de>s  + James Gessling (jgessling@yahoo.com) wrote:oE > Ran across a cute trick to eliminate usage of temp file in dcl (onerH > of my pet peeves).  In sda, I wanted to find out a process logical for= > another process on the system.  Here's the snippet of code:- >- > $ say :== write sys$output8 > $ pipe (say "set erase off" ; say "set proc ''p1'" ; -. >         say "clue proc/log" ) | anal/sys | - >         search sys$input 'p2'  >n > Used like this:p >e" > $ @showlog jgessling fieldtbls32   I use a symbol  .   $ ps == "pipe show system | search sys$pipe"  F that works like "show system /search" but does only show the processesK searched for. Quite handy when you have several hundered processes running.    cu,m   Martin -- uD                     | Martin Vorlaender    |    VMS & WNT programmer-   Smert' Spamionem  | work: mv@pdv-systeme.deiD                     |       http://www.pdv-systeme.de/users/martinv/4                     | home: martin@radiogaga.harz.de   ------------------------------   Date: 29 Jun 2001 01:32:03 GMT! From: jlahman1@aol.com (Jlahman1)l, Subject: Thanks Compaq for the new business!: Message-ID: <20010628213203.19564.00001430@ng-ff1.aol.com>  O The announcements made by Compaq regarding VMS was the biggest bone-headed move L in the history of computing.  Due to this accouncement, we just got an orderM from a large automobile company to convert their VMS systems to windows 2K.  s  N And, I've been talking to a large steel mill that has been a VMS shop for manyO years.  They are now seriously considering a migration to windows 2K from VMS. n    O This only means business for us since we've already made the migration from VMSm to windows NT/2K.a  + Thanks Compaq for shooting VMS in the head!p   ------------------------------  # Date: Fri, 29 Jun 2001 01:53:34 GMTt4 From: "Terry C. Shannon" <terryshannon@mediaone.net>0 Subject: Re: Thanks Compaq for the new business!< Message-ID: <ycR_6.2019$UT1.1063038@typhoon.ne.mediaone.net>  . "Jlahman1" <jlahman1@aol.com> wrote in message4 news:20010628213203.19564.00001430@ng-ff1.aol.com...L > The announcements made by Compaq regarding VMS was the biggest bone-headed moveH > in the history of computing.  Due to this accouncement, we just got an orderdI > from a large automobile company to convert their VMS systems to windowsw 2K.s  G And what car company might that be? I want to make sure to avoid buyinga their wares.   > K > And, I've been talking to a large steel mill that has been a VMS shop for  manyK > years.  They are now seriously considering a migration to windows 2K fromj VMS. >i  * I guess nuclear power is next, hehehehehe!   ------------------------------  % Date: Thu, 28 Jun 2001 19:34:35 -0700p! From: Tom Linden <tom@kednos.com>a0 Subject: RE: Thanks Compaq for the new business!9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEBMCOAA.tom@kednos.com>   > I wonder what the programming language was, that was used at ?   > -----Original Message-----; > From: Terry C. Shannon [mailto:terryshannon@mediaone.net] ' > Sent: Thursday, June 28, 2001 6:54 PMt > To: Info-VAX@Mvb.Saic.Com-2 > Subject: Re: Thanks Compaq for the new business! >  >  > 0 > "Jlahman1" <jlahman1@aol.com> wrote in message6 > news:20010628213203.19564.00001430@ng-ff1.aol.com...C > > The announcements made by Compaq regarding VMS was the biggest t
 > bone-headedi > moveJ > > in the history of computing.  Due to this accouncement, we just got an > order5K > > from a large automobile company to convert their VMS systems to windowsa > 2K.< > I > And what car company might that be? I want to make sure to avoid buying: > their wares. >  > > A > > And, I've been talking to a large steel mill that has been a a > VMS shop for > many> > > years.  They are now seriously considering a migration to  > windows 2K from> > VMS. > >o > , > I guess nuclear power is next, hehehehehe! >  >  >    ------------------------------  % Date: Thu, 28 Jun 2001 23:51:54 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)0 Subject: Re: Thanks Compaq for the new business!L Message-ID: <rdeininger-2806012351550001@user-2ivebg3.dialup.mindspring.com>  : In article <20010628213203.19564.00001430@ng-ff1.aol.com>," jlahman1@aol.com (Jlahman1) wrote:  @ > The announcements made by Compaq regarding VMS was the biggest bone-headed moveN > in the history of computing.  Due to this accouncement, we just got an orderO > from a large automobile company to convert their VMS systems to windows 2K.    > P > And, I've been talking to a large steel mill that has been a VMS shop for manyF > years.  They are now seriously considering a migration to windows 2K
 from VMS.  >  > H > This only means business for us since we've already made the migration from VMS > to windows NT/2K.! > - > Thanks Compaq for shooting VMS in the head!t  H Or, you made this up as a troll.  Unless AOL is in the porting business,G you didn't post from your company.  You didn't sign your post.  Who are  you and why are you credible?I  D I congratulate your firm on negotiating a large contract in 3 days. C Please post the contract.  I am interested in what level of servicei guarantees you are offering.   -- o Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Thu, 28 Jun 2001 23:30:08 -0400a2 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: The death of VMS has been greatly exaggeratedL Message-ID: <rdeininger-2806012330090001@user-2ivebg3.dialup.mindspring.com>  G In article <9hfton$nhm$3@info.cs.uofs.edu>, bill@cs.scranton.edu wrote:   A > By the way, how many people are running production VMS machinesh* > with 128 MB or less of memory in them??   H We have several, but they have almost 128 MB.  And they are not the main9 workhorses.  Still important, but not the most important.a   -- N Robert Deininger rdeininger@mindspring.comn   ------------------------------    Date: 28 Jun 2001 21:18:31 -0700" From: lyngwyst@aol.com (Jay Braun): Subject: Re: The death of VMS has been greatly exaggerated= Message-ID: <4ce97a1a.0106282018.708cb087@posting.google.com>t  l Vance Haemmerle <vance@toyvax.Tucson.AZ.US> wrote in message news:<3B3A9D5A.77C7D32A@toyvax.Tucson.AZ.US>...$ > Eric <etailor@jpl.nasa.gov> wrote: > > ? > > I used to be a vms kernel hacker. I still have 10 copies ofm@ > > that book that says "Vax, Architecture for the '80s".  I was8 > > one of maybe 20 people who knew about special kernelC > > mode ast's. I sold a few million dollars worth of software thats > > did these and other tricks.M >  >   Hi Eric, > G >   Are you really an employee of JPL?  At New Employee Orientation thelE > Ethics people specifically told us not to post on world-wide boardsg< > from JPL accounts unless it was specifically work related.  B Do you think Eric hacks kernels just for fun?  He does it at work,: Vance.  Right next door to me, in fact.  This post is 100%E work-related.  I would have made my posts on this newsgroup from worko, as well, but my computer was being repaired.  C Jay, aka jay.braun@jpl.nasa.gov (uh-oh, I'm really in trouble now).v   ------------------------------  # Date: Thu, 28 Jun 2001 23:33:00 GMTc2 From: "James Hall" <james.h.hall@blueyonder.co.uk>( Subject: Re: VAX-11/780 boot disk needed9 Message-ID: <M8P_6.17835$4i5.1306396@news1.cableinet.net>>  - I believe its an 8 inch floppy (not 5 1/4)...s  = "Didier Morandi" <Didier.Morandi@Compaq.com> wrote in messaget$ news:3B3B943B.3CAED199@Compaq.com... > "Richard W. Schauer" wrote:> > >  > > Hi-  > >hK > > I have a VAX-11/780 in need of the boot disk for the console processor.p I L > > would like to run the machine, as it's in excellent condition except forK > > this missing disk.  If anyone has one they no longer need or can spare,nK > > please let me know.  Also if this is the sort of thing that still mightd beI > > available from Compaq, let me know where to find it because I haven'ti had  > > any luck >o > You mean the 5 1/4 floppy? > Well... try DECUS.% > Or maybe Portobello Rd (London, UK)h >t > D.   ------------------------------  % Date: Thu, 28 Jun 2001 22:10:56 -04002- From: JF Mezei <jfmezei.spamnot@videotron.ca> E Subject: Re: Wailing and moaning... (was: Question to Charlie Matco.) , Message-ID: <3B3BE3AD.21D0278A@videotron.ca>   Bill Todd wrote:L > anything containing any noticeable Alpha features.  I've never figured outL > whether Kerry is just a cheerfully ignorant (and persistent) parrot of theE > Compaq party line or a cheerfully shrewd (and persistent) corporate  > manipulator, a    K Compaq employees were either taken by surprise and have no hard informationsL but try to paint the compaq move in a positive way, or they are NDA'ed up toN their necks, know excatly what is going on but cannot talk and have to provideU vague information.that might look contradictory because we don't know the full story.r  L If Hoff said that he doesn't yet know what is involved in booting VMS on theN IA64 architecture, it is rather pointless to speculate on what changes to IA64 might be required.  J *IF* someone at Compaq actually does know what changes to IA64 Compaq willN request to Intel, it would be interesting to know at what level that might be.F I would have assumed that it would have been someone at Hoff's level.   M Then again, it is also possible that Hoff knows exactly what is involved, buteH cannot reveal that information at this time (since it would be revealing# futures of IA64, an Intel product).n  I In other words, we may have to wait for Intel to reveal that information.o   ------------------------------  # Date: Thu, 28 Jun 2001 23:51:58 GMTs. From: "Alphaman" <alphaman64@nixspam-home.com>- Subject: Re: Where is Samsung in this story ?o; Message-ID: <yqP_6.15622$P5.5133911@news1.rdc1.tn.home.com>   6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OF247F2680.CCD75CBD-ON03256A79.00446122@ep-bc.petrobras.com.br...3 > I didnt see any comment about Suamsung's positiono- > during this week. Do you have any idea ????A >,	 > RegardsA >e > FC  K Yup, they've committed to building EV6 for as long as Compaq wants them to,tJ out to 2012 at least, if necessary.  IBM has also committed to fab'ing EV7 for at least as long.n  < Basically as long as there are orders, they'll build Alphas.   Aaronl --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)e   ------------------------------  % Date: Thu, 28 Jun 2001 20:38:07 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca>0- Subject: Re: Where is Samsung in this story ? 9 Message-ID: <l2Q_6.6340$lp3.401609@news20.bellglobal.com>o  6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OF247F2680.CCD75CBD-ON03256A79.00446122@ep-bc.petrobras.com.br...3 > I didnt see any comment about Suamsung's position - > during this week. Do you have any idea ????i >;	 > Regards  >t > FC  K Since 100% of the Alpha hardware stays with Compaq, I assume that Samsung'slJ position is no different now than it was when Compaq asked IBM to FAB some copper Alphas.  
 Neil Rieck Kitchener/Waterloo/Cambridge,a Ontario, Canada.! http://www3.sympatico.ca/n.rieck/U   ------------------------------  % Date: Thu, 28 Jun 2001 18:27:35 -0500 * From: cjt & trefoil <cheljuba@prodigy.net>4 Subject: Re: Why Compaq won't succeed on IA64 either+ Message-ID: <3B3BBD67.542133A1@prodigy.net>d   JF Mezei wrote:. >  > "Terry C. Shannon" wrote:.N > > etc). If Compaq doesn't communicate this grand plan rapidly, convincingly,M > > and coherently, it will be labeled as a Risky Scheme and it will fail. So % > > will Compaq Computer Corporation.0 > M > Compaq did very well without Digital. It may have a bad few quarters as therM > VMS/Alpha revenus dry up, and Capellas might be replaced by Winkler, but ineN > the end, Compaq will survive without VMS once it is allowed to return to its? > core competency of building on behalf od Microsoft and Intel.   , You think?  Dell seems to have that job now.   ------------------------------  # Date: Fri, 29 Jun 2001 00:28:59 GMTs. From: "Alphaman" <alphaman64@nixspam-home.com>4 Subject: Re: Why Compaq won't succeed on IA64 either; Message-ID: <fZP_6.15785$P5.5166095@news1.rdc1.tn.home.com>i  5 Arne Vajhj <arne.vajhoej@gtech.com> wrote in message # news:3B3B2242.33AE171C@gtech.com...pG > IA-32 has gained a lot on Alpha regarding performanve in the last 3-4h > years.G > And I can understand that Compaq was worried about IA-64 versus Alphat > in the future.  H You know, I had the distinct opportunity to talk and listen to some veryJ knowledgable folks today.  I don't know that it was so much that the Q wasJ worried about IA64 beating Alpha, but if it even came close, what would beI the point?  The average non-computer grunt would buy 2 IA64's at half the 3 price of one Alpha.  Just like today with the IA32.D  J But even better than that, I think that the more interesting thing is thatA Intel has finally recognized that college grads cannot design newlL architectures.  Itanium almost failed completely because of Intel's juvenileI design team; without McKinley from HP, there wouldn't have been any hope.a  G Intel would not want to have to rely on HP for every generation -- theygG would rather have their own designers in house.  Where were they to geti1 experienced chip designers, tho?  Where indeed...e  J The Q has gotten BIG bucks for their contribution.  You won't see a 21064aL counter in the corner of Itanium-II (the follow on to the current generationJ of EPIC-based systems), but they will be designed by Alpha designers, withD plenty of Q-input for Q-specific needs (e.g., NSK lockstep support).  L It may not be called Alpha, but with a little reading between the lines, youK can sure as hell see an admission of failure by Intel.  I do have to wonder L exactly how much money Q is getting today for this, and how much they'll get2 in "preferred pricing" or royalties down the road.   Aaron  --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)f   ------------------------------  # Date: Fri, 29 Jun 2001 01:28:42 GMT:. From: "Alphaman" <alphaman64@nixspam-home.com>4 Subject: Re: Why Compaq won't succeed on IA64 either; Message-ID: <eRQ_6.15806$P5.5220938@news1.rdc1.tn.home.com>   = Terry C. Shannon <terryshannon@mediaone.net> wrote in message 4 news:sfH_6.744$Bp5.423334@typhoon.ne.mediaone.net...K > As for Compaq, they are limited to whatever they can wring out of the EV7wJ > design, which apparently is EV78 and EV79. So if this migration does notK > work, or the new VMS- (and T64 and NSK) ready silicon is not available inm& > the 2004 timeframe, Compaq is toast.  H That's not true.  EV7, EV78, and EV79 are simply what the Q has publiclyH _promised_ to deliver.  If they want, or find that they need, to deliverK more than that, they can continue with shrinks and speed bumps beyond that.-E Development OS's will be available before Itanium-II -- they have to,AF because that's all that's available for the folks in Nashua to work onJ today.  It may not perform as well as Alpha-optimized Itanium-II will, butC work can start today, and customers will be able to start their ownoL development on iVMS in 2003, again, before Itanium-II and independent of it.  J With Alpha engineers working on Itanium-II (read: fixing what's wrong withI Itanium-I with Alpha technology), we can hope that there won't be as manyhC slips and delays as what we saw with the college kids doing Itanic.i  D The Q has committed to support Alpha until at least 2012.  They haveL promises from Samsung for EV6 and IBM for EV7 until at least that date.  Too< bad they didn't tell customers and the press that on Monday.  K What the buffoons<^J> Q marketing types should have done is had the partner L presentation on Monday and the public press conference on Tuesday.  And they should have announced:  ?  . Compaq has changed the Alpha plan from a 25 year lifespan toaB    a promise for a minimum 20 years (2012), although it may extendB    longer than that, if we deem it necessary.  We will fully staffC    for EV7, EV78, EV79, and any subsequent improvements beyond thatsB    as necessary, and all current EV7 development remains on track.  @  . Compaq is starting work on adding yet another 64 bit platform'    for Tru64, OpenVMS, and NSK -- IA64.e  C  . Compaq is helping Intel design the next generation of IA64 whichtG    will be based on Alpha technology and optimized for Compaq operating A    systems.  Intel will pay Compaq an undisclosed amount for thiss    technology transfer.s  D  . Intel will hire Alpha engineers from Compaq as their design tasksC    are finished at Compaq, based on the above schedule.  Intel willh!    pay Compaq for this expertise.t  E  . Compaq will open up OpenVMS to run on any company's IA64 platform,tD    although we will offer value-added features for Compaq platforms.C    Support on other vendors' platforms will be left to the hardware D    vendor, just as Microsoft does today with their Windows products,<    unless that hardware is specifically qualified by Compaq.  L  . Translation tools for both Alpha and VAX and MIPS applications are in theH    investigation stage; initial research makes us feel confident that weF    will be able to provide such tools for initial OpenVMS and NSK/IA64 support.L    (Translation tools are not critical in the open source UNIX environment.)  ?  . Support for VAX and Alpha OpenVMS applications will continuen=    indefinitely.  Your applications will continue to run in aeA    supported VMS environment on any supported VAX, Alpha, or IA64 C    platform with full support from Compaq.  Today, we can guaranteef@    that support until at least 2012; we will extend that date as+    long as customers need it in the future./  L You know what folks?  That's what they should have said, because that's whatK they meant.  The Intel brainwashed numbies just didn't phrase it like that.rI They have a real problem understanding how to make stuff palatable for useL enterprise consumers.  I think they honestly believe everyone just loves theK Blue Man Group, Bunny suited mimes, and the stoopid 5 note arpeggio.  Get a.I grip, Mr. C -- such just AIN'T the case.  We were NOT amused to hear thatnA you'd sold Alpha to Intel as the opening to your news conference.e  L Somebody at Planet Houston needs to seriously think about hiring someone who1 knows how to speak to enterprise computer people.a   Aaront --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/H "The supersonic boom should hit you in just a few seconds." (Apollo 440)   ------------------------------    Date: 28 Jun 2001 21:57:21 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)u4 Subject: Re: Why Compaq won't succeed on IA64 either3 Message-ID: <NiPpP$lk7m1Q@eisner.encompasserve.org>!  l In article <eRQ_6.15806$P5.5220938@news1.rdc1.tn.home.com>, "Alphaman" <alphaman64@nixspam-home.com> writes:  M > What the buffoons<^J> Q marketing types should have done is had the partnertN > presentation on Monday and the public press conference on Tuesday.  And they > should have announced:  7 	<snip different expression of the facts known to date>   ; Certainly the timing and wording of the Monday announcement2; had to be something agreed with Intel.  It emphasized total < committment to Intel, whereas newsgroup discussion has shown> several areas where Compaq has other options.  Certainly Intel> lawyers are aware of options open to Compaq if things go sour,; but that is something Intel would rather not see publicizedn= because it weakens public faith in IA64.  And Intel is payingb money for this.a   ------------------------------  # Date: Fri, 29 Jun 2001 02:04:08 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>4 Subject: Re: Why Compaq won't succeed on IA64 either< Message-ID: <smR_6.2031$UT1.1070821@typhoon.ne.mediaone.net>  F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:NiPpP$lk7m1Q@eisner.encompasserve.org...sH > In article <eRQ_6.15806$P5.5220938@news1.rdc1.tn.home.com>, "Alphaman"% <alphaman64@nixspam-home.com> writes:1 >0G > > What the buffoons<^J> Q marketing types should have done is had the4 partner@K > > presentation on Monday and the public press conference on Tuesday.  Anda they > > should have announced: >g8 > <snip different expression of the facts known to date> >o= > Certainly the timing and wording of the Monday announcementb= > had to be something agreed with Intel.  It emphasized total > > committment to Intel, whereas newsgroup discussion has shown@ > several areas where Compaq has other options.  Certainly Intel@ > lawyers are aware of options open to Compaq if things go sour,= > but that is something Intel would rather not see publicized ? > because it weakens public faith in IA64.  And Intel is payinge > money for this.e  C Still, Aaron brings up a valid point: if Compaq doesn't effectivelytG communicate this transition scheme, Compaq is toast. And given the richhJ marketing heritage CPQ inherited from DEC Classic and CPQ Classic, one has$ to wonder if they're up to the task.   ------------------------------  % Date: Thu, 28 Jun 2001 22:47:21 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca>r4 Subject: Re: Why Compaq won't succeed on IA64 either, Message-ID: <3B3BEC34.E29F082B@videotron.ca>   Alphaman wrote:RN > You know what folks?  That's what they should have said, because that's whatM > they meant.  The Intel brainwashed numbies just didn't phrase it like that.n  L But there may be good reasons why they didn't make the announcement in a wayE that would not have caused so many customer's hearts to miss beats.  8  L Face the fact: Compaq's actions (or lack thereof) have demonstrated that VMSL is not a key product to its strategy. It may be a cash cow right now but not in the future.  M I view the WAY in which Compaq made the announcement as yet another hint that-M Compaq doesn't intend to make VMS a core Compaq product. A strong hint to thec remaining customer base.  L In my opinion, the only remaining question is whether Compaq will eventuallyN sell VMS or just milk the remaining customers until the revenus don't generate. any profits and they then officially kill VMS.  N The fact that they are porting it to IA64 only means that they aren't ready toN make that decision yet and that VMS generates enough revenus and doesn't steal? sales from Microsoft so it is benign enough to be kept for now.l   ------------------------------   Date: 28 Jun 2001 21:35:02 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog). Subject: Re: Windows Images Running Under iVMS, Message-ID: <9hg7u6$jpk@gap.cco.caltech.edu>  _ In article <tjmrd75ved0746@news.supernews.com>, wspencer@ap.nospam.org (Warren Spencer) writes:fD >Is there any credence to the notion that Windows executables might * >eventually run under OpenVMS on Itanium?   I It might happen after Microsoft takes OpenVMS away from Compaq - but thent it will be called Windows too.  ! > That fact that DEQ has already p0 >ported Win32 to OpenVMS 7.x has me wondering...  7 Huh?  You mean that affinity nonsense?  Fuhgeddaboutit.-   Regards,   David Mathog mathog@caltech.edu? Manager, sequence analysis facility, biology division, Caltech 1J **************************************************************************J *                       RIP VMS & ALPHA                                  *J **************************************************************************   ------------------------------  # Date: Thu, 28 Jun 2001 23:41:27 GMTe. From: "Alphaman" <alphaman64@nixspam-home.com>. Subject: Re: Windows Images Running Under iVMS; Message-ID: <HgP_6.15521$P5.5123479@news1.rdc1.tn.home.com>s  8 Warren Spencer <wspencer@ap.nospam.org> wrote in message) news:tjmrd75ved0746@news.supernews.com...tD > Is there any credence to the notion that Windows executables mightJ > eventually run under OpenVMS on Itanium?  That fact that DEQ has already1 > ported Win32 to OpenVMS 7.x has me wondering.... >  > ws  K Under?  Well, I guess that's a possibility -- you might want to drop a noteaB to Mark Gorham and let him know you've got a requirement for that.  % But let me answer that with one word:d    Galaxy.   Regards, Aaronh --> Aaron Sakovich  http://members.home.net/sakovich/alphaman.html> Make April 15 just another day:        http://www.fairtax.org/K "Oh my God, they killed Alpha!!!  You bastards!" (Anonymous Alpha Engineer)d   ------------------------------    Date: 28 Jun 2001 19:46:22 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)e. Subject: Re: Windows Images Running Under iVMS3 Message-ID: <tChzgHQ1t7XU@eisner.encompasserve.org>e  l In article <HgP_6.15521$P5.5123479@news1.rdc1.tn.home.com>, "Alphaman" <alphaman64@nixspam-home.com> writes:: > Warren Spencer <wspencer@ap.nospam.org> wrote in message+ > news:tjmrd75ved0746@news.supernews.com...fE >> Is there any credence to the notion that Windows executables mighttK >> eventually run under OpenVMS on Itanium?  That fact that DEQ has already 2 >> ported Win32 to OpenVMS 7.x has me wondering... >> >> wss > M > Under?  Well, I guess that's a possibility -- you might want to drop a noteoD > to Mark Gorham and let him know you've got a requirement for that. > ' > But let me answer that with one word:a > 
 >  Galaxy.  * Let me respond to Aaron with another word:    Uniprocessor.   ------------------------------  % Date: Thu, 28 Jun 2001 20:40:22 -0500 ' From: "john oxley" <john@johnoxley.com>w% Subject: www.vms-support.com for saler* Message-ID: <9hg4p1$puk1@sv5.cwpanama.net>    Anyone have an interest in this? john@johnoxley.com   ------------------------------   End of INFO-VAX 2001.357 ************************