1 INFO-VAX	Wed, 26 Jun 2002	Volume 2002 : Issue 348       Contents: $CRMPSC and memory management ! Re: $CRMPSC and memory management ! Re: $CRMPSC and memory management ! Re: $CRMPSC and memory management ! Re: $CRMPSC and memory management 4 Re: A possible shift in the status of VMS ar HP ????4 Re: A possible shift in the status of VMS ar HP ????4 Re: A possible shift in the status of VMS ar HP ????; Re: A possible shift in the status of VMS ar HP ???? ERRATA ; Re: A possible shift in the status of VMS ar HP ???? ERRATA ; Re: A possible shift in the status of VMS ar HP ???? ERRATA ; Re: A possible shift in the status of VMS ar HP ???? ERRATA ; Re: A possible shift in the status of VMS ar HP ???? ERRATA 8 Re: Amdrew wants numbers, here they are!  Blow out baby!8 Re: Amdrew wants numbers, here they are!  Blow out baby!8 Re: Amdrew wants numbers, here they are!  Blow out baby!9 Re: Call or email hp ... they will answer your questions! 9 Re: Call or email hp ... they will answer your questions! ) Re: Happy Anniversary!	:-) (was: Re: wow)  HSZ50 and RZ1DF-CB Re: HSZ50 and RZ1DF-CB Re: HSZ50 and RZ1DF-CB" Re: Intel booting contest - Update Re: More good news RE: More good news Re: OpenVMS on CNBC  Re: OpenVMS on CNBC  Re: OpenVMS on CNBC ; Process Software does SSH for TCP/IP Services - new release ? Re: Process Software does SSH for TCP/IP Services - new release 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 3 Re: Reuters test - Itanium II blows away Sparky III 
 Re: TCO study  Re: VMS port delayed!  Re: VMS port delayed!  Re: WASD startup problems , Re: Why does SMTP RBL filter appear to fail?  F ----------------------------------------------------------------------  # Date: Tue, 25 Jun 2002 19:19:47 GMT % From: Robert Davis <res057y3@gte.net> & Subject: $CRMPSC and memory management' Message-ID: <3D18C119.D74FC50F@gte.net>   E I'm trying to understand a little more about VMS memory management so @ I can pick the 'best' address to map a shared memory section. MyC problem is caused by the fact that our software requires two global E sections. The first to be mapped, the main section, is larger in size E (and may grow quite large based on user config) and is mapped EXPREG. F The second is used by code that has been ported and 'tacked on' from aE different system. It's smaller and must be mapped at a specific known D address ... which is where my problem begins since the first sectionF can be potentially large enough to overlap the second (and cause nasty	 results).   D I'm trying to determine what is the highest address I can safely mapE the second section, on both VAX and Alpha. From the $CRMPSC reference E I see that it is "generally discouraged" to map sections in P1 space. C My understanding is that P0 space runs from  0 to x3FFFFFFF. But it D also appears that VIRTUALPAGECNT determines exactly how much of thisC is available and that "the division of these pages between P0 space 8 and P1 space is totally arbitrary and process-specific".  @ Our fixed section was originally mapped at 0x04000000, which wasB eventually over-run by a larger main section. I attempted to 'fix'? this by moving it to 0x50000000, whch seemed to work OK ... but E recently I've gotten reports that this doesn't work on VAX ... and it E seems that this may be dangerous as it's in P1 space. On one specific ? VAX, nothing above 0x10000000 would work, but I suspect this is C because VIRTUALPAGECNT was too low. I believe that something around E 0x30000000 is probably a good compromise, and should work on both VAX , and Alpha if VIRTUALPAGECNT is properly set.  - Thanks for any clarifications or suggestions.    Rob    ------------------------------  # Date: Tue, 25 Jun 2002 20:16:36 GMT 0 From: prune@ZAnkh-Morpork.mv.com (Paul Winalski)* Subject: Re: $CRMPSC and memory management9 Message-ID: <3d18cf66.1454506870@proxy.news.easynews.com>   C You definitely DON'T want to play around in P1 space.  That's where ; pieces of the CLI and the various process mode stacks live.   A If you don't know the size of the section that must be at a fixed > address, you're prety much out of luck here.  The best you can: do is to choose a VA high enough to accomodate the largest$ expected size for the other section.  ? Would it be possible to map the sections in the opposite order? @ If you mapped the smaller, fixed-address section first, to a low? address, that leaves the rest of virtual address space free for  the other section.      A On Tue, 25 Jun 2002 19:19:47 GMT, Robert Davis <res057y3@gte.net>  wrote:   > F >I'm trying to understand a little more about VMS memory management soA >I can pick the 'best' address to map a shared memory section. My D >problem is caused by the fact that our software requires two globalF >sections. The first to be mapped, the main section, is larger in sizeF >(and may grow quite large based on user config) and is mapped EXPREG.G >The second is used by code that has been ported and 'tacked on' from a F >different system. It's smaller and must be mapped at a specific knownE >address ... which is where my problem begins since the first section G >can be potentially large enough to overlap the second (and cause nasty 
 >results). > E >I'm trying to determine what is the highest address I can safely map F >the second section, on both VAX and Alpha. From the $CRMPSC referenceF >I see that it is "generally discouraged" to map sections in P1 space.D >My understanding is that P0 space runs from  0 to x3FFFFFFF. But itE >also appears that VIRTUALPAGECNT determines exactly how much of this D >is available and that "the division of these pages between P0 space9 >and P1 space is totally arbitrary and process-specific".  > A >Our fixed section was originally mapped at 0x04000000, which was C >eventually over-run by a larger main section. I attempted to 'fix' @ >this by moving it to 0x50000000, whch seemed to work OK ... butF >recently I've gotten reports that this doesn't work on VAX ... and itF >seems that this may be dangerous as it's in P1 space. On one specific@ >VAX, nothing above 0x10000000 would work, but I suspect this isD >because VIRTUALPAGECNT was too low. I believe that something aroundF >0x30000000 is probably a good compromise, and should work on both VAX- >and Alpha if VIRTUALPAGECNT is properly set.  > . >Thanks for any clarifications or suggestions. >  >Rob  
 ---------- Remove 'Z' to reply by email.    ------------------------------  # Date: Tue, 25 Jun 2002 20:31:46 GMT % From: Robert Davis <res057y3@gte.net> * Subject: Re: $CRMPSC and memory management' Message-ID: <3D18D1F8.2F42998B@gte.net>   D Sorry I wasn't clear ... I do know the size of the second section atF the fixed address and it's relatively small. So once I know what's theF 'safe' top end, I can easily decide the start address. I'm mostly justB a little unclear about the relationship between VIRTUALPAGECNT and= P0/P1, and where that makes P0/P1 begin/end, and exactly what  addresses are available.  < I don't believe reversing the order of mapping is an option.   Thanks,    Rob    Paul Winalski wrote: > E > You definitely DON'T want to play around in P1 space.  That's where = > pieces of the CLI and the various process mode stacks live.  > C > If you don't know the size of the section that must be at a fixed @ > address, you're prety much out of luck here.  The best you can< > do is to choose a VA high enough to accomodate the largest& > expected size for the other section. > A > Would it be possible to map the sections in the opposite order? B > If you mapped the smaller, fixed-address section first, to a lowA > address, that leaves the rest of virtual address space free for  > the other section. > C > On Tue, 25 Jun 2002 19:19:47 GMT, Robert Davis <res057y3@gte.net>  > wrote: >  > > H > >I'm trying to understand a little more about VMS memory management soC > >I can pick the 'best' address to map a shared memory section. My F > >problem is caused by the fact that our software requires two globalH > >sections. The first to be mapped, the main section, is larger in sizeH > >(and may grow quite large based on user config) and is mapped EXPREG.I > >The second is used by code that has been ported and 'tacked on' from a H > >different system. It's smaller and must be mapped at a specific knownG > >address ... which is where my problem begins since the first section I > >can be potentially large enough to overlap the second (and cause nasty  > >results). > > G > >I'm trying to determine what is the highest address I can safely map H > >the second section, on both VAX and Alpha. From the $CRMPSC referenceH > >I see that it is "generally discouraged" to map sections in P1 space.F > >My understanding is that P0 space runs from  0 to x3FFFFFFF. But itG > >also appears that VIRTUALPAGECNT determines exactly how much of this F > >is available and that "the division of these pages between P0 space; > >and P1 space is totally arbitrary and process-specific".  > > C > >Our fixed section was originally mapped at 0x04000000, which was E > >eventually over-run by a larger main section. I attempted to 'fix' B > >this by moving it to 0x50000000, whch seemed to work OK ... butH > >recently I've gotten reports that this doesn't work on VAX ... and itH > >seems that this may be dangerous as it's in P1 space. On one specificB > >VAX, nothing above 0x10000000 would work, but I suspect this isF > >because VIRTUALPAGECNT was too low. I believe that something aroundH > >0x30000000 is probably a good compromise, and should work on both VAX/ > >and Alpha if VIRTUALPAGECNT is properly set.  > > 0 > >Thanks for any clarifications or suggestions. > >  > >Rob >  > ---------- > Remove 'Z' to reply by email.    ------------------------------  # Date: Tue, 25 Jun 2002 21:40:43 GMT  From: danco@pebble.org* Subject: Re: $CRMPSC and memory management- Message-ID: <slrnahhpqm.d7o.danco@pebble.org>   ; In article <3D18D1F8.2F42998B@gte.net>, Robert Davis wrote: F > Sorry I wasn't clear ... I do know the size of the second section atH > the fixed address and it's relatively small. So once I know what's theH > 'safe' top end, I can easily decide the start address. I'm mostly justD > a little unclear about the relationship between VIRTUALPAGECNT and? > P0/P1, and where that makes P0/P1 begin/end, and exactly what  > addresses are available.  A So far as I know, VIRTUALPAGECNT has no bearing on where P0 ends, @ at least not in the normal OpenVMS user mode program.  P0 begins? at 0 and ends at 4FFFFFFF.  You can map to any address range in D P0 that isn't already being used for something else.  VIRTUALPAGECNTB is just the limit on the total number of process virtual pages youD can have.  Some process pages may be mapped and some may not.  A bigA jig-saw puzzle with holes.  The number of pages that are actually A mapped must be less than or equal to VIRTUALPAGECNT.  If you very ? badly fragement P0 virtual address space, you can run out of P0 ? virtual address space before you run out of VIRTUALPAGECNT.  If @ expreg sections are allocated after the highest currently mappedC virtual addresses (don't recall if they are or not), then combining A expreg sections with sections at high fixed addresses could cause  a problem like that.  G Then there is PROCSECTCNT that limits how many sections you can have.   A The default limit of 32 sections is rather low.  If you use a lot > of shareable images that are not well consolidated into as few> image sections as possible, or you have a lot of data sectionsC mapped through $CRMPSC and friends in your program, you can easilly  go over 32.    ------------------------------  # Date: Tue, 25 Jun 2002 23:36:29 GMT ( From: Don Sykes <annonymous@pacbell.net>* Subject: Re: $CRMPSC and memory management+ Message-ID: <3D18FEDF.A4CA2CFD@pacbell.net>    Robert Davis wrote:  > F > Sorry I wasn't clear ... I do know the size of the second section atH > the fixed address and it's relatively small. So once I know what's theH > 'safe' top end, I can easily decide the start address. I'm mostly justD > a little unclear about the relationship between VIRTUALPAGECNT and? > P0/P1, and where that makes P0/P1 begin/end, and exactly what  > addresses are available. > > > I don't believe reversing the order of mapping is an option. > 	 > Thanks,  >  > Rob  >  > Paul Winalski wrote: > > G > > You definitely DON'T want to play around in P1 space.  That's where ? > > pieces of the CLI and the various process mode stacks live.  > > E > > If you don't know the size of the section that must be at a fixed B > > address, you're prety much out of luck here.  The best you can> > > do is to choose a VA high enough to accomodate the largest( > > expected size for the other section. > > C > > Would it be possible to map the sections in the opposite order? D > > If you mapped the smaller, fixed-address section first, to a lowC > > address, that leaves the rest of virtual address space free for  > > the other section. > > E > > On Tue, 25 Jun 2002 19:19:47 GMT, Robert Davis <res057y3@gte.net> 
 > > wrote: > >  > > > J > > >I'm trying to understand a little more about VMS memory management soE > > >I can pick the 'best' address to map a shared memory section. My H > > >problem is caused by the fact that our software requires two globalJ > > >sections. The first to be mapped, the main section, is larger in sizeJ > > >(and may grow quite large based on user config) and is mapped EXPREG.K > > >The second is used by code that has been ported and 'tacked on' from a J > > >different system. It's smaller and must be mapped at a specific knownI > > >address ... which is where my problem begins since the first section K > > >can be potentially large enough to overlap the second (and cause nasty  > > >results). > > > I > > >I'm trying to determine what is the highest address I can safely map J > > >the second section, on both VAX and Alpha. From the $CRMPSC referenceJ > > >I see that it is "generally discouraged" to map sections in P1 space.H > > >My understanding is that P0 space runs from  0 to x3FFFFFFF. But itI > > >also appears that VIRTUALPAGECNT determines exactly how much of this H > > >is available and that "the division of these pages between P0 space= > > >and P1 space is totally arbitrary and process-specific".  > > > E > > >Our fixed section was originally mapped at 0x04000000, which was G > > >eventually over-run by a larger main section. I attempted to 'fix' D > > >this by moving it to 0x50000000, whch seemed to work OK ... butJ > > >recently I've gotten reports that this doesn't work on VAX ... and itJ > > >seems that this may be dangerous as it's in P1 space. On one specificD > > >VAX, nothing above 0x10000000 would work, but I suspect this isH > > >because VIRTUALPAGECNT was too low. I believe that something aroundJ > > >0x30000000 is probably a good compromise, and should work on both VAX1 > > >and Alpha if VIRTUALPAGECNT is properly set.  > > > 2 > > >Thanks for any clarifications or suggestions. > > >   L It's important to consider that VIRTUALPAGECNT is Obsolete on Alphas w/VMS >J 7.0. On those platforms the value is always forced to %X7FFFFFFF. For your! VAX's, set it high using AUTOGEN. O And as Paul mentioned, the best way around your problem is mapping the variable / large section AFTER the known smaller section.     --     Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Francisco    ------------------------------  # Date: Tue, 25 Jun 2002 18:47:38 GMT ( From: Don Sykes <annonymous@pacbell.net>= Subject: Re: A possible shift in the status of VMS ar HP ???? + Message-ID: <3D18BB29.E38F059F@pacbell.net>    Dave Gudewicz wrote: > J > Just this week I learned.....  there is a VMS (only) tour in the US comeL > July and August.  Today I heard the news of VMS v7.3-1 on MSNBC and Yahoo.H > I don't ever remember VMS getting this kind of attention.  Something'sI > happening here, what it is ain't exactly clear.  <--- I just rememberedm > words from a song of the 60s.yP Yea. It's from Buffalo Springfield circa 1966. Unfortunately the next line goes:% 	There's a man with a gun over there. ! 	Tell'n me that I got to be ware!o   -- r   Have VMS. Will Travel. Wire Paladin (@alphase.com)H
 San Franciscoo   ------------------------------  # Date: Tue, 25 Jun 2002 19:06:10 GMT 1 From: "Terry C. Shannon" <terryshannon@attbi.com>i= Subject: Re: A possible shift in the status of VMS ar HP ????t? Message-ID: <Ca3S8.176102$6m5.147678@rwcrnsc51.ops.asp.att.net>   5 "Don Sykes" <annonymous@pacbell.net> wrote in messaget% news:3D18BB29.E38F059F@pacbell.net...  > Dave Gudewicz wrote: > >eL > > Just this week I learned.....  there is a VMS (only) tour in the US comeG > > July and August.  Today I heard the news of VMS v7.3-1 on MSNBC anda Yahoo.J > > I don't ever remember VMS getting this kind of attention.  Something'sK > > happening here, what it is ain't exactly clear.  <--- I just remembered5! > > words from a song of the 60s.-L > Yea. It's from Buffalo Springfield circa 1966. Unfortunately the next line goes:E& > There's a man with a gun over there." > Tell'n me that I got to be ware!   But further along we have...   Paranoia strikes deepe Into your life it will creep# It starts when you're always afraid 7 You step out of line, the man come and take you away...   C The announcement of a new VMS release on the anniversary of the IPF J Consolidation announcement would seem to indicate that "The Man" has taken	 VMS away.k   cheers,    terry st3 Man With a Gun (and an IBM S/360) Over There, 70-72t   ------------------------------  % Date: Tue, 25 Jun 2002 21:59:01 +0200w" From: "Hans Vlems" <hvlems@iae.nl>= Subject: Re: A possible shift in the status of VMS ar HP ???? 5 Message-ID: <afahic$cdgcf$1@ID-143435.news.dfncis.de>l  I What about this: Tru64 sites were probably DEC sites until CPQ took over.l	 Some even I had PDP-11's or VAX systems. Now that their favorite o/s was killed it is- assumed that those customers will go HP-UX.J The question is if the "DEC affinity" may be stronger than the o/s one. In other words3I could they be interested in VMS (with proven cluster technology) instead?7L Existing Alpha hardware may even help that decision. In fact that might have been one of  the reasons to kill Alpha.   Hans8 JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message& news:3D17CA36.49A5C3DE@videotron.ca... > David Froble wrote:oL > > The HP Enterprise Technical Symposium 2002 (previously known as CETS) is HP's >eH > > In the above paragraph VMS is mentioned before all but storageworks. >rA > Coinsider that VMS is the only semi-viable product left for the I > group-formerly-known-as-DECUS to handle, since the rest will be done bys HP'sL > Interex user group. Note that what used to be DECUS could also play a hugeK > role in getting Tru64 customers to express their needs and comments abouts HP'sG > conversion plans (from a technical perspective, what utilities/system  services: > need be ported to ease customer's own porting problems).   ------------------------------  # Date: Tue, 25 Jun 2002 19:12:23 GMT.1 From: "Terry C. Shannon" <terryshannon@attbi.com>eD Subject: Re: A possible shift in the status of VMS ar HP ???? ERRATA? Message-ID: <rg3S8.176145$6m5.147186@rwcrnsc51.ops.asp.att.net>   < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message9 news:Ca3S8.176102$6m5.147678@rwcrnsc51.ops.asp.att.net...n >c7 > "Don Sykes" <annonymous@pacbell.net> wrote in message0' > news:3D18BB29.E38F059F@pacbell.net...t > > Dave Gudewicz wrote: > > >dI > > > Just this week I learned.....  there is a VMS (only) tour in the US  comeI > > > July and August.  Today I heard the news of VMS v7.3-1 on MSNBC andc > Yahoo.L > > > I don't ever remember VMS getting this kind of attention.  Something'sB > > > happening here, what it is ain't exactly clear.  <--- I just
 remembered# > > > words from a song of the 60s.nI > > Yea. It's from Buffalo Springfield circa 1966. Unfortunately the next- line > goes:e( > > There's a man with a gun over there.$ > > Tell'n me that I got to be ware! >  > But further along we have... >a > Paranoia strikes deepc > Into your life it will creep% > It starts when you're always afraido9 > You step out of line, the man come and take you away...  > E > The announcement of a new VMS release on the anniversary of the IPF L > Consolidation announcement would seem to indicate that "The Man" has taken > VMS away.t >u	 > cheers,  >c	 > terry sh5 > Man With a Gun (and an IBM S/360) Over There, 70-72  >   . Ummm, meant to say *HAS NOT* taken VMS away...   terry s   Shot in Foot With Own Gun, Again >h   ------------------------------  # Date: Tue, 25 Jun 2002 21:21:49 GMTe# From: "John Smith" <a@nonymous.com>oD Subject: Re: A possible shift in the status of VMS ar HP ???? ERRATAI Message-ID: <N95S8.24626$71t1.21203@news01.bloor.is.net.cable.rogers.com>e  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message9 news:rg3S8.176145$6m5.147186@rwcrnsc51.ops.asp.att.net...d > 	 > terry sa" > Shot in Foot With Own Gun, Again  C At the risk of starting a 'war', perhaps it's time for gun control?    ------------------------------  # Date: Tue, 25 Jun 2002 21:35:48 GMTT1 From: "Terry C. Shannon" <terryshannon@attbi.com>'D Subject: Re: A possible shift in the status of VMS ar HP ???? ERRATA? Message-ID: <Um5S8.177458$6m5.148401@rwcrnsc51.ops.asp.att.net>n  . "John Smith" <a@nonymous.com> wrote in messageC news:N95S8.24626$71t1.21203@news01.bloor.is.net.cable.rogers.com...I >t> > "Terry C. Shannon" <terryshannon@attbi.com> wrote in message; > news:rg3S8.176145$6m5.147186@rwcrnsc51.ops.asp.att.net...C > >r > > terry st$ > > Shot in Foot With Own Gun, Again >sE > At the risk of starting a 'war', perhaps it's time for gun control?o >F  ( Nope. Just commonsense keyboard control.   ------------------------------  # Date: Tue, 25 Jun 2002 23:08:17 GMTi( From: Don Sykes <annonymous@pacbell.net>D Subject: Re: A possible shift in the status of VMS ar HP ???? ERRATA+ Message-ID: <3D18F843.2FF78539@pacbell.net>0   "Terry C. Shannon" wrote:r > > > "Terry C. Shannon" <terryshannon@attbi.com> wrote in message; > news:Ca3S8.176102$6m5.147678@rwcrnsc51.ops.asp.att.net...  > >g9 > > "Don Sykes" <annonymous@pacbell.net> wrote in messagee) > > news:3D18BB29.E38F059F@pacbell.net...  > > > Dave Gudewicz wrote: > > > > K > > > > Just this week I learned.....  there is a VMS (only) tour in the USa > comeK > > > > July and August.  Today I heard the news of VMS v7.3-1 on MSNBC and,
 > > Yahoo.N > > > > I don't ever remember VMS getting this kind of attention.  Something'sD > > > > happening here, what it is ain't exactly clear.  <--- I just > remembered% > > > > words from a song of the 60s.-K > > > Yea. It's from Buffalo Springfield circa 1966. Unfortunately the nextn > line	 > > goes:$* > > > There's a man with a gun over there.& > > > Tell'n me that I got to be ware! > >c  > > But further along we have... > >n > > Paranoia strikes deepr  > > Into your life it will creep' > > It starts when you're always afraid ; > > You step out of line, the man come and take you away...e > >eG > > The announcement of a new VMS release on the anniversary of the IPFaN > > Consolidation announcement would seem to indicate that "The Man" has taken
 > > VMS away.n > >  > > cheers,i > >  > > terry s 7 > > Man With a Gun (and an IBM S/360) Over There, 70-72i > >D > 0 > Ummm, meant to say *HAS NOT* taken VMS away... > 	 > terry s " > Shot in Foot With Own Gun, Again > >i  ( Let's hope your UPDATED version is true.   -- o   Have VMS. Will Travel. Wire Paladin (@alphase.com)h
 San Franciscor   ------------------------------  # Date: Tue, 25 Jun 2002 23:15:09 GMTe1 From: "Terry C. Shannon" <terryshannon@attbi.com> D Subject: Re: A possible shift in the status of VMS ar HP ???? ERRATA. Message-ID: <1Q6S8.151538$nZ3.65801@rwcrnsc53>  5 "Don Sykes" <annonymous@pacbell.net> wrote in messager% news:3D18F843.2FF78539@pacbell.net...e > "Terry C. Shannon" wrote:J > >0   <snip> > > 2 > > Ummm, meant to say *HAS NOT* taken VMS away... > >I > > terry su$ > > Shot in Foot With Own Gun, Again > > >g >n* > Let's hope your UPDATED version is true. >   K Agreed! Time will tell. Perchance there will be tales to tell subsequent tor my next pilgrimage to ZKOland.   cheers,t   terry s,   ------------------------------  % Date: Tue, 25 Jun 2002 15:15:23 -0400 * From: "rob kas" <rob@paychoice.nospam.com>A Subject: Re: Amdrew wants numbers, here they are!  Blow out baby! 3 Message-ID: <3d18c13d$0$1421$8e9e3842@news.atx.net>a       Bobc  5  Do you ever think about what you read or do you justi  regurgitate Web Pages ?  7 You do know that is comparing the lasest GREATEST Alpha I  againest a older SUN chip. And all thing considered not much better theng	 IBM's P4.s  8 By the way I am looking foward to meeting you at the VMS	 Roadshow.t  "                                Rob     >n) > http://www.theinquirer.net/25060207.htmp >. > & > HPQ's ES45 1.26GHz Alpha benchmarked >m > Phew what a scorcha!* > By the INQUIRER, 25/06/2002 09:47:25 BST >s >a$ > A PICTURE SPEAKS A THOUSAND words.   ------------------------------    Date: 25 Jun 2002 15:55:16 -0700( From: bob@instantwhip.com (Bob Ceculski)A Subject: Re: Amdrew wants numbers, here they are!  Blow out baby!0= Message-ID: <d7791aa1.0206251455.5842cd40@posting.google.com>.   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<afa26h$i62$1@new-usenet.uk.sun.com>...k > Bob Ceculski wrote:s > 1 > > Ok Andrew, you wanted numbers, here they are!  > >  > > + > > http://www.theinquirer.net/25060207.htm, > >  > > ) > > HPQ's ES45 1.26GHz Alpha benchmarked o > >  > > Phew what a scorcha! r, > > By the INQUIRER, 25/06/2002 09:47:25 BST > >  > > & > > A PICTURE SPEAKS A THOUSAND words. > >  >  > Really, Bob get a clue.  > 	 > Regardsn > Andrew Harrison   E the clue is Andrew, if you expect someone to believe that a processoraE with triple the floating point performance and double the int doesn't4D mean a squat of beans to any application running under that cpu, letC alone smp configurations, then I expect your next line to be "And ID2 also have some swamp land for sale in Florida" ...   ------------------------------  # Date: Wed, 26 Jun 2002 00:23:59 GMT * From: "Bill Todd" <billtodd@metrocast.net>A Subject: Re: Amdrew wants numbers, here they are!  Blow out baby!tC Message-ID: <zQ7S8.439346$Oa1.31010046@bin8.nnrp.aus1.giganews.com>i  5 "Bob Ceculski" <bob@instantwhip.com> wrote in messagee7 news:d7791aa1.0206251455.5842cd40@posting.google.com...i   ...   G > the clue is Andrew, if you expect someone to believe that a processora? > with triple the floating point performance and double the inta  L Hmmm.  Perhaps you're an idiot in the technical sense of the term as well asF the colloquial one.  You certainly don't appear to have mastered basic arithmetic, anyway.o  I 467 * 2 = 934.  The new Alpha hits only 850 SPECint, so even based on themJ Inquirer's data (without realizing that it just happened to ignore USIII'sH top-of-the-line 1050 MHz processor) the Alpha doesn't double the SPECintI performance.  Use the top-of-the-line USIII SPECint number (which is 610, L not that I expect you to be aware of it, off in your own little world as youH are) and the Alpha edge dwindles farther to a bit under 1.4x:  certainly( decent, but nothing like what you claim.  L Same problem with the SPECfp numbers, of course - at least you're consistentL in your ignorance, though it gets pretty 'old' after far less exposure to itJ than you insist on providing.  And, of course, you've consistently ignoredJ Andrew's point from the beginning that great single-processor SPEC numbers' do not necessarily a great system make.   I Pretty much batting .000 as usual, Bob.  After this long, I'm not holding  out much hope for improvement.   - bill   ------------------------------  # Date: Tue, 25 Jun 2002 18:00:14 GMTu# From: "John Smith" <a@nonymous.com>aB Subject: Re: Call or email hp ... they will answer your questions!E Message-ID: <Oc2S8.9589$mi.4283@news01.bloor.is.net.cable.rogers.com>z  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message  news:3D17DE9B.398A550@fsi.net... > - > d) HP can do nothing about VMS and survive.  >rA > Note, however, that a distinction is made between "succeed" andw > "survive".    ( HP can do nothing about VMS and survive.  L a) Hmmmm, do you mean that, HP will survive if they do nothing about VMS? or  B b) Did you mean that VMS will survive if HP does nothing about it?    C In my view, a) is probably likely, while b) is a certainty - if thea conditions mentioned are met.h   ------------------------------  # Date: Tue, 25 Jun 2002 21:29:03 GMT # From: "John Smith" <a@nonymous.com> B Subject: Re: Call or email hp ... they will answer your questions!G Message-ID: <zg5S8.11148$mi.10631@news01.bloor.is.net.cable.rogers.com>h  	 ooops....e  the conclusion should read......  G In my view, a) is probably likely, while b) is NOT a certainty - if the  conditions mentioned are met.u      . "John Smith" <a@nonymous.com> wrote in message? news:Oc2S8.9589$mi.4283@news01.bloor.is.net.cable.rogers.com...f > > > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message" > news:3D17DE9B.398A550@fsi.net... > > / > > d) HP can do nothing about VMS and survive.d > >nC > > Note, however, that a distinction is made between "succeed" andb > > "survive". >  >h* > HP can do nothing about VMS and survive. >dK > a) Hmmmm, do you mean that, HP will survive if they do nothing about VMS?t or >pD > b) Did you mean that VMS will survive if HP does nothing about it? >b >dE > In my view, a) is probably likely, while b) is a certainty - if the  > conditions mentioned are met.o >f >  >  >    ------------------------------  # Date: Tue, 25 Jun 2002 18:53:11 GMT 1 From: "Terry C. Shannon" <terryshannon@attbi.com>c2 Subject: Re: Happy Anniversary!	:-) (was: Re: wow). Message-ID: <r_2S8.149637$nZ3.65303@rwcrnsc53>  D "Bradford J. Hamilton" <sy18889@rabmbit.famrp.cosm> wrote in message news:CvrZsPerHBcs@rabbit... H > Please notice also the date of the announcement - has it really been a wholes= > year since the incident which some refer to as "Alphacide"?c  * Correctomundo. Exactly one year ago today.   ------------------------------  # Date: Tue, 25 Jun 2002 18:56:16 GMTe From: dittman@dittman.nett Subject: HSZ50 and RZ1DF-CBh8 Message-ID: <k13S8.24101$cE5.17352@nwrddc02.gnilink.net>  < I've got the latest (last?) firmware in my HSZ50 controllers; (V57Z), but the controllers can't use the RZ1DF-CB drives Ih> have installed.  Is there any way to get these drives working?   -- d Eric Dittman dittman@dittman.nett= Check out the DEC Enthusiasts Club at http://www.dittman.net/e   ------------------------------  # Date: Wed, 26 Jun 2002 00:50:35 GMT ! From: rob.buxton@wcc.spam.govt.nz  Subject: Re: HSZ50 and RZ1DF-CB % Message-ID: <3d190d1f.182930930@news>t  5 What type of BA35n Cabinets are the Disks mounted in?o  , I know on our HSD30 can see 9 & 18 GB disks.@ These conenct through to a BA356 (I think) that has an 8-Bit I/O Module.p  F So, If you're using BA350s then you will not be able to see the disks.E Not sure if you'll need the 8-Bit or 16-Bit I/O module for the shelf.T   HTH    Rob.< On Tue, 25 Jun 2002 18:56:16 GMT, dittman@dittman.net wrote:  = >I've got the latest (last?) firmware in my HSZ50 controllersA< >(V57Z), but the controllers can't use the RZ1DF-CB drives I? >have installed.  Is there any way to get these drives working?e >e >-- 
 >Eric Dittmana >dittman@dittman.net> >Check out the DEC Enthusiasts Club at http://www.dittman.net/   ------------------------------  # Date: Wed, 26 Jun 2002 03:15:02 GMTa From: dittman@dittman.netj Subject: Re: HSZ50 and RZ1DF-CBs8 Message-ID: <WkaS8.26545$cE5.23772@nwrddc02.gnilink.net>  " rob.buxton@wcc.spam.govt.nz wrote:  7 : What type of BA35n Cabinets are the Disks mounted in?t  . : I know on our HSD30 can see 9 & 18 GB disks.B : These conenct through to a BA356 (I think) that has an 8-Bit I/O	 : Module.t  H : So, If you're using BA350s then you will not be able to see the disks.G : Not sure if you'll need the 8-Bit or 16-Bit I/O module for the shelf.   < I'm using a BA356 with the 8-bit I/O module.  The controller( can see the RZ29-VW drives in the shelf. -- a Eric Dittman dittman@dittman.net = Check out the DEC Enthusiasts Club at http://www.dittman.net/-   ------------------------------  % Date: Tue, 25 Jun 2002 18:33:15 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)+ Subject: Re: Intel booting contest - UpdateeL Message-ID: <rdeininger-2506021833150001@11cust145.tnt2.nashua.nh.da.uu.net>  < In article <d7791aa1.0206250504.f6fb371@posting.google.com>,) bob@instantwhip.com (Bob Ceculski) wrote:a  > >"Sue Skonetski" <susan.skonetski@compaq.com> wrote in message. news:<o%KR8.30$Zm5.879663@news.cpqcorp.net>...B >> Just letting you know that we have received 727 guesses so far. >>  @ >> http://www.compaq.com/hps/ipf-enterprise/openvms_contest.html >> 0 >> sue > = >Since the HAL code problem has delayed it, does that mean we 
 >can reguess?y  E This question is based on a dangerous assumption -- that the inquirerSH actually came close to the truth with their recent story on the IPF port of VMS.r   ------------------------------  # Date: Tue, 25 Jun 2002 21:03:02 GMT,* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: More good news4B Message-ID: <aU4S8.125269$_j6.6769239@bin3.nnrp.aus1.giganews.com>  = "Sue Skonetski" <susan.skonetski@compaq.com> wrote in message6, news:2a0S8.21$6P5.549981@news.cpqcorp.net...B > Hard to apologize for sending so much good news about VMS today.  F Would have been nicer if the string 'VMS' had appeared in the article.   - bill   ------------------------------  % Date: Tue, 25 Jun 2002 19:21:46 -0400e' From: "Main, Kerry" <Kerry.Main@hp.com>  Subject: RE: More good newsTT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4026607C2@kaoexc01.americas.cpqcorp.net>   Bill,    Re: mention of OpenVMS ..=20   Well, indirectly speaking :-)a  
 Extract from: G <http://www.bobsguide.com/cgi-bin/guide/newsitem.pl?ID=3Dbg&seq=3D1427>eH "Deutsche B=F6rse has been offering participants liquid trading in the =- seven Dutch Euro-STOXX since November 2001.":   ( Pointer to Deutsche B=F6rse testimonial:; http://www.openvms.compaq.com/openvms/brochures/borchesite/    Regards5  
 Kerry Main Senior Consultantd Hewlett-Packard Canada! Consulting & Integration Servicesh Voice: 613-592-4660y Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----2 From: Bill Todd [mailto:billtodd@metrocast.net]=20 Sent: June 25, 2002 5:03 PMc To: Info-VAX@Mvb.Saic.Com  Subject: Re: More good newsw      ? "Sue Skonetski" <susan.skonetski@compaq.com> wrote in message = , news:2a0S8.21$6P5.549981@news.cpqcorp.net...B > Hard to apologize for sending so much good news about VMS today.  F Would have been nicer if the string 'VMS' had appeared in the article.   - bill   ------------------------------  # Date: Tue, 25 Jun 2002 21:06:21 GMTC* From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: OpenVMS on CNBCC Message-ID: <hX4S8.437851$Oa1.30871830@bin8.nnrp.aus1.giganews.com>   = "Sue Skonetski" <susan.skonetski@compaq.com> wrote in messagee, news:Tr_R8.14$HL5.399507@news.cpqcorp.net...J > http://famulus.msnbc.com/famuluscom/businesswire06-25-060938.asp?sym=HPQ  K It's beginning to appear possible that, while your new company is still runtF by scumbags, they may be measurably less incompetent scumbags than theL previous lot.  Whether this will benefit VMS and its customers remains to beL seen:  on the one hand, one might suspect that *any* change would have to beL for the better; on the other, one has only to look a year back from today to! see that this is not always true.e   - bill   ------------------------------  % Date: Tue, 25 Jun 2002 19:51:02 -0400b, From: "Frank Sapienza" <sapienza@noesys.com> Subject: Re: OpenVMS on CNBC, Message-ID: <afavl902bgv@enews4.newsguy.com>  5 "Bill Todd" <billtodd@metrocast.net> wrote in messagey= news:hX4S8.437851$Oa1.30871830@bin8.nnrp.aus1.giganews.com...yI > It's beginning to appear possible that, while your new company is stillo runcH > by scumbags, they may be measurably less incompetent scumbags than the > previous lot.@   ROTFL!  @ Bill, you're so cute when you say nice things about people.  :-)   ------------------------------  # Date: Wed, 26 Jun 2002 01:02:14 GMT * From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: OpenVMS on CNBCB Message-ID: <qo8S8.402418$%y.29461674@bin4.nnrp.aus1.giganews.com>  7 "Frank Sapienza" <sapienza@noesys.com> wrote in messagem& news:afavl902bgv@enews4.newsguy.com... >a7 > "Bill Todd" <billtodd@metrocast.net> wrote in message-? > news:hX4S8.437851$Oa1.30871830@bin8.nnrp.aus1.giganews.com...MK > > It's beginning to appear possible that, while your new company is stillh > run J > > by scumbags, they may be measurably less incompetent scumbags than the > > previous lot.h >  > ROTFL! >hB > Bill, you're so cute when you say nice things about people.  :-)  + I always try to give credit where it's due.o   - bill   ------------------------------   Date: 25 Jun 02 22:29:04 +0200) From: p_sture@elias.decus.ch (Paul Sture).D Subject: Process Software does SSH for TCP/IP Services - new release) Message-ID: <vkYFz5vHpNOi@elias.decus.ch>@  > Noticed this today. No idea yet if there's a Hobbyist version.  ' http://www.process.com/tcpip/sshds.htmlo   SSH FOR OPENVMSs  L Complete SSH Solution for Alpha and VAX Systems using Compaq TCP/IP Services   See url for further details.   __
 Paul Sture Switzerlandn   ------------------------------   Date: 26 Jun 2002 02:38:05 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>H Subject: Re: Process Software does SSH for TCP/IP Services - new release+ Message-ID: <afb9ed02js@enews4.newsguy.com>-  * Paul Sture <p_sture@elias.decus.ch> wrote:@ > Noticed this today. No idea yet if there's a Hobbyist version.  / I don't see any indication, but I sure hope so!)   		Zane   ------------------------------  # Date: Tue, 25 Jun 2002 18:07:33 GMT@5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>@< Subject: Re: Reuters test - Itanium II blows away Sparky III2 Message-ID: <Fj2S8.33$dS5.681020@news.cpqcorp.net>  6 Andrew Harrison SUNUK Consultancy wrote in message ... >E >t& >Wow what else does it do, make toast. >-  I Let's lead the subject farther away.  Let's ignore performance estimates,eK and what makes a system perform better or worse.  The problem you have here.K is that a real customer, real applications, real systems are not making the I Sparc story sound pretty comparing it against the much maligned IA64.  NodG matter what way we cut it, it appears on the surface that a system thatiL costs more (the Sun) and has more CPU's, doesn't meet the performance of the HP IA64 system.-  I You can ignore it as "not having enough information" - but even a cynicalnK view of the report still result in the fact that Sun should start to worry.8   ------------------------------    Date: 25 Jun 2002 13:31:46 -0600+ From: young_r@encompasserve.org (Rob Young)i< Subject: Re: Reuters test - Itanium II blows away Sparky III3 Message-ID: <KSyEiNqNmBip@eisner.encompasserve.org>   j In article <Fj2S8.33$dS5.681020@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes: > 8 > Andrew Harrison SUNUK Consultancy wrote in message ... >> >>' >>Wow what else does it do, make toast.f >> > K > Let's lead the subject farther away.  Let's ignore performance estimates, M > and what makes a system perform better or worse.  The problem you have here M > is that a real customer, real applications, real systems are not making the K > Sparc story sound pretty comparing it against the much maligned IA64.  No I > matter what way we cut it, it appears on the surface that a system thatRN > costs more (the Sun) and has more CPU's, doesn't meet the performance of the > HP IA64 system.r > K > You can ignore it as "not having enough information" - but even a cynicaluM > view of the report still result in the fact that Sun should start to worry.t >   A 	Worry and change their minds.  The price performance curves that D 	Sun business managers are looking at are most likely sounding alarm 	bells.  Prognostication alert:6  + 		Sun becomes an IA64 partner someday soon.1  5 	That are go down like a rock clinging to UltraSparc.t  > 	I suspect we know more before October.  Zander and crew exit 1 	gracefully and change of direction is announced.m  4 	Triumphant UltraSparc soundings prior to Ultracide?   	Deja Vu all over again?    4 http://news.com.com/2100-1001-938995.html?tag=fd_top  # Sun to push UltraSparc V past 3GHz e   By Stephen Shankland t Staff Writer, CNET News.comm June 24, 2002, 11:15 PM PT  E 	"Sun has more than 1,400 designers working on its Sparc processors."3  B 	They can't maintain that investment when their server margins are? 	severely impacted in a down economy.  IA64 - if nothing else - A 	will ensure much cheaper high-end servers and slimmer margins to  	go with it.  E "IBM's p630, to be introduced Tuesday, brings its Power4 processor tot@ comparatively inexpensive systems costing a well under $50,000."  F 	(don't pin me down on that last comment, we will know more later and * 	look back... "much cheaper" is relative).   				Robi   ------------------------------  # Date: Tue, 25 Jun 2002 20:50:20 GMTt* From: "Bill Todd" <billtodd@metrocast.net>< Subject: Re: Reuters test - Itanium II blows away Sparky IIIA Message-ID: <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>e  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:KSyEiNqNmBip@eisner.encompasserve.org...oF > In article <Fj2S8.33$dS5.681020@news.cpqcorp.net>, "Fred Kleinsorge"% <kleinsorge@star.zko.dec.com> writes:G   ...   E > > You can ignore it as "not having enough information" - but even ae cynical H > > view of the report still result in the fact that Sun should start to worry.  K I'd say that might depend on just how cynical one's view was.  For example,3L considering that this 'report' comes from a lot of the same upper managementK that swore that Alpha couldn't maintain a significant (even just 2:1, whichaI is now looking like a major under-estimate) performance edge over Itanic, K suspecting that their performance assertions in this area might be off by at( factor of 3 or more is hardly difficult.   > >  >vB > Worry and change their minds.  The price performance curves thatE > Sun business managers are looking at are most likely sounding alarmo  > bells.  Prognostication alert: >e+ > Sun becomes an IA64 partner someday soon.s  @ That prognostication seems to assume that Sun's management is as0 galactically stupid as Compaq's was.  We'll see.   ...a   > IA64 - if nothing else -B > will ensure much cheaper high-end servers and slimmer margins to
 > go with it.h  F Really?  So far, Sun's servers appear to have the price advantage over- comparable Itanics.  As do IBM's (see below).o   >sG > "IBM's p630, to be introduced Tuesday, brings its Power4 processor topB > comparatively inexpensive systems costing a well under $50,000." >dF > (don't pin me down on that last comment, we will know more later and+ > look back... "much cheaper" is relative).   I Saw a figure of about $30K stated yesterday for a 4-processor 630.  Looks-K like Intel will have to do some radical rethinking about its server pricingsL (since neither Itanic2 nor Xeon seems very competitive with that at present)I if it wants to fulfill your expectations for taking over that part of the > world any time soon.  And that's *before* Hammer comes down...   - bill   ------------------------------  # Date: Tue, 25 Jun 2002 20:53:24 GMTn1 From: "Terry C. Shannon" <terryshannon@attbi.com>B< Subject: Re: Reuters test - Itanium II blows away Sparky III> Message-ID: <8L4S8.148822$R61.53966@rwcrnsc52.ops.asp.att.net>  # "Andrew Harrison SUNUK Consultancy"s> <andrew_nospam.harrison_remove_this@sun#.com> wrote in message* news:afa4h7$iu2$1@new-usenet.uk.sun.com... <snip> >a; > So using the HP platform chipset improves the performancet< > of Itanium II from slightly less than that of an Ultra III; > Intels own 70% performance hike estimate for Itanium madea' > in April to 4 x that of an Ultra III.o >e' > Wow what else does it do, make toast.. >   E That's entirely possible... what's the estimated power consumption ofa Itanium II?s   ------------------------------    Date: 25 Jun 2002 17:43:21 -0600+ From: young_r@encompasserve.org (Rob Young)a< Subject: Re: Reuters test - Itanium II blows away Sparky III3 Message-ID: <ezPgoEFBApo7@eisner.encompasserve.org>d  n In article <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes: >    >  >> > >>C >> Worry and change their minds.  The price performance curves that*F >> Sun business managers are looking at are most likely sounding alarm! >> bells.  Prognostication alert:a >>, >> Sun becomes an IA64 partner someday soon. > B > That prognostication seems to assume that Sun's management is as2 > galactically stupid as Compaq's was.  We'll see. >  > ...1 >  >> IA64 - if nothing else - C >> will ensure much cheaper high-end servers and slimmer margins tor >> go with it. > H > Really?  So far, Sun's servers appear to have the price advantage over/ > comparable Itanics.  As do IBM's (see below).r >   @ 	Maybe on parity or more... but probably not for long.  Besides,7 	they won't keep up with Itanium so what good are they?o   >>H >> "IBM's p630, to be introduced Tuesday, brings its Power4 processor toC >> comparatively inexpensive systems costing a well under $50,000."s >>G >> (don't pin me down on that last comment, we will know more later and2, >> look back... "much cheaper" is relative). > D > Saw a figure of about $30K stated yesterday for a 4-processor 630.  = 	That wasn't a good figure you saw.  A less than loaded 2-wayC 	sells for $28K   ] http://commerce.www.ibm.com/content/home/shop_ShopIBM/en_US/eServer/pSeries/entry/6306C4.htmle    ' Model 		Processor	count	L3	disk		memoryn    7028-6C4C       POWER4:		-, 		1.0 GHz  	2-way  	32MB  	109.2GB  	4GB      2  8 $28,043.00 IBM Web price*a  n  u    7028-6C4D  	POWER4:		:+ 		1.0 GHz  	4-way  	64MB  	220.2GB		8GB    ?     A $50,591.00 IBM Web price*F  5  p6 	Intel is competing with HP/Dell.  IA64 boxes will be A 	plentiful and relatively "cheap."  That $41K 4 processor Bandera8C 	box by Intel.... I'll bet Dell beats that price by a few thousand.s     LooksuM > like Intel will have to do some radical rethinking about its server pricing N > (since neither Itanic2 nor Xeon seems very competitive with that at present)K > if it wants to fulfill your expectations for taking over that part of thee@ > world any time soon.  And that's *before* Hammer comes down... >   @ 	Well actually, I think across the board 4-way IA64 boxes shouldB 	be at or near 5-10K cheaper apples-to-apples (same memory sizing,B 	same CPU count) and much to Andrew's protestations actually very A 	competitive performance-wise.  10K versus IBM, less for Sun (ande2 	yes it will probably take a Dell to ensure that).  > 	This paradigm "scoot" involves more of a shift to services asC 	margins shrink and shrink on the last bastion of margin hardware, i- 	"high-end" servers (64-bit multi-cpu boxes).a   				Rob    ------------------------------  # Date: Tue, 25 Jun 2002 23:19:34 GMTh1 From: "Terry C. Shannon" <terryshannon@attbi.com>a< Subject: Re: Reuters test - Itanium II blows away Sparky III? Message-ID: <aU6S8.178155$6m5.149166@rwcrnsc51.ops.asp.att.net>s  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:ezPgoEFBApo7@eisner.encompasserve.org...oI > In article <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>, "Billu& Todd" <billtodd@metrocast.net> writes: > >s >  > >u > >> > > >>E > >> Worry and change their minds.  The price performance curves that H > >> Sun business managers are looking at are most likely sounding alarm# > >> bells.  Prognostication alert:a > >>. > >> Sun becomes an IA64 partner someday soon. > >VD > > That prognostication seems to assume that Sun's management is as4 > > galactically stupid as Compaq's was.  We'll see. > >e  H Believe it or not, SUNW in the mid-90's or thereabouts actually ponderedL becoming an Alpha partner (a deal involving plenty of quid, and a pro quo inB the form of DEC saying "no" to Digital UNIX and adopting Solaris).  F I don't claim to know all that much about SUNW's processor 'n platformK strategy beyond what's been publicly articulated, but IMHO having IA64 as as( "fallback" might not be such a bad idea.   ------------------------------  # Date: Tue, 25 Jun 2002 23:19:10 GMT * From: "Bill Todd" <billtodd@metrocast.net>< Subject: Re: Reuters test - Itanium II blows away Sparky IIIB Message-ID: <OT6S8.388409$Gs.29522596@bin5.nnrp.aus1.giganews.com>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:ezPgoEFBApo7@eisner.encompasserve.org...aI > In article <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>, "Bill%& Todd" <billtodd@metrocast.net> writes:   ...    > >> IA64 - if nothing else -eE > >> will ensure much cheaper high-end servers and slimmer margins tor > >> go with it. > >lJ > > Really?  So far, Sun's servers appear to have the price advantage over1 > > comparable Itanics.  As do IBM's (see below).  > >g >u7 > Maybe on parity or more... but probably not for long.o  L Right.  I'll pay considerably more attention when you have something solider: in terms of future pricing information than your own SWAG.  
   Besides,8 > they won't keep up with Itanium so what good are they?  I Pardon me while I guffaw.  POWER4 won't 'keep up with' Itanic?  Give me aeI break:  it has enough advantages in both the memory sub-system and the MP9) interconnect to leave it far in the dust.   K You might have more of a prayer with USIII:  it also has its on-chip memorysI controller going for it, but probably less of an MP advantage over Itanic-: than POWER4 does.  But we'll see real numbers soon enough.   >K > >>J > >> "IBM's p630, to be introduced Tuesday, brings its Power4 processor toE > >> comparatively inexpensive systems costing a well under $50,000."  > >>I > >> (don't pin me down on that last comment, we will know more later and . > >> look back... "much cheaper" is relative). > >?F > > Saw a figure of about $30K stated yesterday for a 4-processor 630. >i> > That wasn't a good figure you saw.  A less than loaded 2-way > sells for $28K >s > L http://commerce.www.ibm.com/content/home/shop_ShopIBM/en_US/eServer/pSeries/ entry/6306C4.html   I So it seems.  Of course, a fairly comparable (one less disk) V480 Sun boxi sells for about $23K.3   ...6  6 > Intel is competing with HP/Dell.  IA64 boxes will be# > plentiful and relatively "cheap."T  I Why do you think they'll be any cheaper than existing Xeon servers, whichoD both Sun's and IBM's RISC pricing seems pretty competitive with (see previous Register reference:K  http://www.theregus.com/content/53/25276.html )?  Given the relative sales K of Xeons vs. Itanics, one could easily expect the latter to be considerablyi *more* expensive.V  J I guess you're just full of faith.  Others might want to wait and see whatL the reality is.  The only thing that's certain so far is that current ItanicH and (reported) Itanic2 pricing does not appear to support your claims at all.   - bill   ------------------------------  # Date: Wed, 26 Jun 2002 00:55:53 GMTe* From: "Bill Todd" <billtodd@metrocast.net>< Subject: Re: Reuters test - Itanium II blows away Sparky IIIC Message-ID: <ti8S8.439602$Oa1.31033733@bin8.nnrp.aus1.giganews.com>l  < "Terry C. Shannon" <terryshannon@attbi.com> wrote in message9 news:aU6S8.178155$6m5.149166@rwcrnsc51.ops.asp.att.net...e   ...i  H > I don't claim to know all that much about SUNW's processor 'n platformK > strategy beyond what's been publicly articulated, but IMHO having IA64 aso ar* > "fallback" might not be such a bad idea.  H Given that IA64 still has definite possibilities of falling *over*, if II were Sun I'd keep my options open for a while longer.  Porting Solaris to<J Itanic would likely be fairly easy for them at any point, especially givenL that they supposedly started such a port once already:  it's clear that IA64J isn't going anywhere unexpected (let alone thrilling) for about the next 3A years unless it goes away completely, whereas Hammer has definiteuK possibilities and could be well worth waiting for (and who knows?  USIV and K USV might actually come in on schedule - and if they do, performance shouldbK remain competitive with contemporary Itanics, as USIII likely will be until0 then).   - bill   ------------------------------    Date: 25 Jun 2002 20:14:48 -0600+ From: young_r@encompasserve.org (Rob Young)r< Subject: Re: Reuters test - Itanium II blows away Sparky III3 Message-ID: <reWdbW2POhai@eisner.encompasserve.org>o  o In article <OT6S8.388409$Gs.29522596@bin5.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:. > : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:ezPgoEFBApo7@eisner.encompasserve.org...uJ >> In article <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>, "Bill( > Todd" <billtodd@metrocast.net> writes: >  > ...e >  >> >> IA64 - if nothing else -F >> >> will ensure much cheaper high-end servers and slimmer margins to >> >> go with it.- >> >K >> > Really?  So far, Sun's servers appear to have the price advantage over 2 >> > comparable Itanics.  As do IBM's (see below). >> > >>8 >> Maybe on parity or more... but probably not for long. > N > Right.  I'll pay considerably more attention when you have something solider< > in terms of future pricing information than your own SWAG. >  >   Besides,9 >> they won't keep up with Itanium so what good are they?e > K > Pardon me while I guffaw.  POWER4 won't 'keep up with' Itanic?  Give me aaK > break:  it has enough advantages in both the memory sub-system and the MPo+ > interconnect to leave it far in the dust.r  ? 	That "they" referred to Sun and creative trimming ruined that.g   > M > You might have more of a prayer with USIII:  it also has its on-chip memory K > controller going for it, but probably less of an MP advantage over Itanich< > than POWER4 does.  But we'll see real numbers soon enough. >  >> >> >>2K >> >> "IBM's p630, to be introduced Tuesday, brings its Power4 processor tocF >> >> comparatively inexpensive systems costing a well under $50,000." >> >>SJ >> >> (don't pin me down on that last comment, we will know more later and/ >> >> look back... "much cheaper" is relative).  >> >G >> > Saw a figure of about $30K stated yesterday for a 4-processor 630.t >>? >> That wasn't a good figure you saw.  A less than loaded 2-waye >> sells for $28Ks >> >>N > http://commerce.www.ibm.com/content/home/shop_ShopIBM/en_US/eServer/pSeries/ > entry/6306C4.htmlg >  > So it seems.    < 	Not "seems."  It "is."  You can order then at those prices.  = > Of course, a fairly comparable (one less disk) V480 Sun boxP > sells for about $23K.f   	Not 4-way.    >  > ...m > 7 >> Intel is competing with HP/Dell.  IA64 boxes will beb$ >> plentiful and relatively "cheap." > K > Why do you think they'll be any cheaper than existing Xeon servers, which-F > both Sun's and IBM's RISC pricing seems pretty competitive with (see > previous Register reference:M >  http://www.theregus.com/content/53/25276.html )?  Given the relative salesiM > of Xeons vs. Itanics, one could easily expect the latter to be considerably" > *more* expensive.i > % > I guess you're just full of faith. r  = 	No.  Faith is beyond reason.  I'm reasoning above, sometimes B 	incorrectly.  Back to Xeon... we've kicked that around... without@ 	looking (no time, so feel free) I believe the Dell versus AlphaA 	numbers kicked around a while back had Dell at $27K for a nicely > 	loaded 4 processor box.  I believe IA64 will get down to that9 	range in Madison timeframe (shrinky dinks and all that).i  '  Others might want to wait and see whatwN > the reality is.  The only thing that's certain so far is that current ItanicJ > and (reported) Itanic2 pricing does not appear to support your claims at > all. >   < 	Sure.  But this is Usenet and it is a lot of fun to dig out< 	Zinc Whiskers and what not and have a good laugh.  Meaning A 	prognostications - correct and incorrect - , goofs and blunders,d> 	spot-on successes are what makes these forums very enjoyable.  < 	Regarding not supporting?  Why did you trim the $50K Power4= 	4 processor box?  It clearly shows it is $9K higher than the < 	purported $41K for Intel's Bandera.  I'll bet Dell comes inB 	cheaper than that.  Manufacturing effeciencies and all that gains? 	you.   So maybe you are referring to Sun?  Okay.  But they areS9 	way behind the Itanium price performance curve and theiri 	business managers know it.1   				Rob9   ------------------------------  # Date: Wed, 26 Jun 2002 03:50:05 GMTS* From: "Bill Todd" <billtodd@metrocast.net>< Subject: Re: Reuters test - Itanium II blows away Sparky IIIC Message-ID: <NRaS8.441315$Oa1.31163791@bin8.nnrp.aus1.giganews.com>e  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:reWdbW2POhai@eisner.encompasserve.org...lJ > In article <OT6S8.388409$Gs.29522596@bin5.nnrp.aus1.giganews.com>, "Bill& Todd" <billtodd@metrocast.net> writes: > >i< > > "Rob Young" <young_r@encompasserve.org> wrote in message1 > > news:ezPgoEFBApo7@eisner.encompasserve.org...dL > >> In article <gI4S8.73470$Ny6.4133894@bin2.nnrp.aus1.giganews.com>, "Bill* > > Todd" <billtodd@metrocast.net> writes: > >- > > ...- > >-  > >> >> IA64 - if nothing else -H > >> >> will ensure much cheaper high-end servers and slimmer margins to > >> >> go with it.s > >> >H > >> > Really?  So far, Sun's servers appear to have the price advantage over4 > >> > comparable Itanics.  As do IBM's (see below). > >> > > >>: > >> Maybe on parity or more... but probably not for long. > >6H > > Right.  I'll pay considerably more attention when you have something solider > > > in terms of future pricing information than your own SWAG. > >b > >   Besides,; > >> they won't keep up with Itanium so what good are they?n > >lK > > Pardon me while I guffaw.  POWER4 won't 'keep up with' Itanic?  Give mew asJ > > break:  it has enough advantages in both the memory sub-system and the MP- > > interconnect to leave it far in the dust.m >t@ > That "they" referred to Sun and creative trimming ruined that.  J Bullshit, Rob:  I trimmed absolutely nothing above between my reference toL both Sun and IBM and your response to it.  Try reading what we both actuallyK said instead of just shooting blindly from the hip:  your aim from there ism nothing to brag about.   >e > >nH > > You might have more of a prayer with USIII:  it also has its on-chip memoryF > > controller going for it, but probably less of an MP advantage over Itanic> > > than POWER4 does.  But we'll see real numbers soon enough. > >m > >> > >> >>lJ > >> >> "IBM's p630, to be introduced Tuesday, brings its Power4 processor toH > >> >> comparatively inexpensive systems costing a well under $50,000." > >> >>aL > >> >> (don't pin me down on that last comment, we will know more later and1 > >> >> look back... "much cheaper" is relative).s > >> >I > >> > Saw a figure of about $30K stated yesterday for a 4-processor 630.  > >>A > >> That wasn't a good figure you saw.  A less than loaded 2-wayi > >> sells for $28Kd > >> > >> > >"L http://commerce.www.ibm.com/content/home/shop_ShopIBM/en_US/eServer/pSeries/ > > entry/6306C4.htmlo > >  > > So it seems. >t= > Not "seems."  It "is."  You can order then at those prices.n  I What I meant (but did not state clearly) by 'seems' is that this made theaL number I saw for the 4-processor box appear erroneously low.  I said 'seems'L because  a) street prices aren't always quite as high as list prices and  b)G because prices can vary in non-negligible ways with other configurationCL changes (e.g., memory and storage, though since the 4-P box listed by IBM atJ $50K had only 8 GB of memory and three 73.4 GB disks it would be difficult1 to account for a $20K difference by those alone).a   >_? > > Of course, a fairly comparable (one less disk) V480 Sun boxh > > sells for about $23K.o >. > Not 4-way.  K The IBM box you described and I responded to is not a 4-way either:  I saidr$ comparable, and that's what I meant.  - You really need to pay better attention, Rob.f   >a > >m > > ...n > >l9 > >> Intel is competing with HP/Dell.  IA64 boxes will be & > >> plentiful and relatively "cheap." > >bG > > Why do you think they'll be any cheaper than existing Xeon servers,- which-H > > both Sun's and IBM's RISC pricing seems pretty competitive with (see  > > previous Register reference:I > >  http://www.theregus.com/content/53/25276.html )?  Given the relativeo salesrB > > of Xeons vs. Itanics, one could easily expect the latter to be considerably > > *more* expensive.l > >@& > > I guess you're just full of faith. >  > No.  Faith is beyond reason.  A That's my point:  all too often *you* appear to be beyond reason.o      I'm reasoning above, sometimesC > incorrectly.  Back to Xeon... we've kicked that around... withoutrA > looking (no time, so feel free) I believe the Dell versus Alpha B > numbers kicked around a while back had Dell at $27K for a nicely > loaded 4 processor box.I  I I'll give you that one, after a quick look at Dell's 6650 Xeon prices:  IiI think the Register article seriously overestimated them.  However, Dell'sa' Itanic1 prices are a great deal higher.p  &   I believe IA64 will get down to that: > range in Madison timeframe (shrinky dinks and all that).  H USIV (which includes at least a shrink to 130 nm - I'm not familiar withJ what else it might contain) is due at about the same time as Madison, and,F while POWER5 isn't due until 2005 IIRC, IBM reportedly plans to shrinkG POWER4 to a 130 nm process next year (and likely again after that).  So2L process shrinks just maintain the status quo among these three, since Itanic0 isn't getting any other real boosts before 2005.   ...t  = > Regarding not supporting?  Why did you trim the $50K Power48 > 4 processor box?  C Because after I accepted that the figure I had seen might have been61 erroneous it was irrelevant to what I had to say.r  ,   It clearly shows it is $9K higher than the% > purported $41K for Intel's Bandera.8  I BFD.  While both boxes had 8 GB of memory, we know nothing more about the	I Itanic box's configuration.  E.g., the HP press release about the ReuterspG application performance described a 4-P Itanic box with two 18 GB disksrK rather than the three 73 GB disks the 4-P IBM box has:  that's about $2K ortK more right there (and note that Dell's Itanic1 prices include just *one* 18eJ GB hard drive and it appears only the motherboard controller for it).  AndJ if the Itanic price didn't include 64-bit Win2KAS, that's another $3295 (IE found that price on a Compaq web page, but I've heard $5K elsewhere - E possibly the Compaq price was for fewer processors.  IBM's price doescK include AIX, and Sun's includes Solaris; I suppose you could purchase HP-UXKD for the Intel box if you cared to, but I doubt that's cheap either).     I'll bet Dell comes in > cheaper than that.  H Not if its Itanic1 prices are any guide (and since last I knew Intel wasL charging as much for Itanic2 as for Itanic1, it's not obvious why Dell would lower them).  /   Manufacturing effeciencies and all that gainst@ > you.   So maybe you are referring to Sun?  Okay.  But they are: > way behind the Itanium price performance curve and their > business managers know it.  I More bullshit, Rob.  The only pricing and public third-party test resultsuB currently available show USIII ahead of Itanic in both price *and*L performance, and while Itanic2 will likely close the gap it's far from clearG that it will do anything significantly more than that.  We already know H Intel's Itanic2 boxes won't be ahead of Sun's on price per se at the 4-PI data point, given that the $47K 4-P Sun box had 8 GB more memory than the>E $41K 4-P Itanic2 (even Dell's memory prices for Itanic1 are well over>J $1000/GB, and I suspect given historical precedent that, say, HP's will be considerably higher).a  = Come back when you have something more than hot air to offer.V   - bill   ------------------------------  % Date: Tue, 25 Jun 2002 21:33:24 +0200l) From: labadie <labadie.gerard@wanadoo.fr>e Subject: Re: TCO study* Message-ID: <3D18C584.53C1612D@wanadoo.fr>  % Interesting study, I had not noticed.a   Thanks Kerry   Grard   "Main, Kerry" wrote:  	 > Gerard,a >g > Re: TCO Study .. Check out:iG > http://www.openvms.compaq.com/openvms/whitepapers/enterprise_tco.htmlr > 	 > Regardst >e > Kerry Main > Senior Consultantc > Hewlett-Packard Canada# > Consulting & Integration Servicesh > Voice: 613-592-4660- > Fax   : 613-591-4477 > Email: Kerry.Main@hp.com >1 > -----Original Message-----+ > From: labadie [mailto:labadie_g@decus.fr]s > Sent: June 25, 2002 4:47 AMy > To: Info-VAX@Mvb.Saic.Com0 > Subject: TCO study >u > Some thoughts about TCO. > > I have been recently to Decus Lyon, and talked to a Vms System Engineer. He works now on Vms and a Unix (I will not say which one). He said that, in his opinion, the TCO of his Unix machines was very, very high compared to Vms. He said that he was called too often during the night or the week-end, and that he could not understand why they were working with that Operating System, not reliable at all, a nightmare as he said. >r2 > I guess a honest study about TCO does not exist.n > If  Ibm (or HP, or ...) has one, it will only reflect the relative power of the Mvs, Aix, OS 400 inside Ibm. >a >  I do not talk about  a study by D H Brown, paid by Microsoft, to show that Linux was more expensive than NT, or something like that.  >d4 > I guess a honest study about TCO is an oxymoron... >e	 > RegardsI >i > Grard   ------------------------------  # Date: Tue, 25 Jun 2002 20:10:34 GMTt0 From: prune@ZAnkh-Morpork.mv.com (Paul Winalski) Subject: Re: VMS port delayed!9 Message-ID: <3d18cea2.1454311168@proxy.news.easynews.com>r  3 On Tue, 25 Jun 2002 15:28:50 GMT, "Fred Kleinsorge"r$ <kleinsorge@star.zko.dec.com> wrote:  K >I'm not quite sure who they get their information from, but as the projectTJ >leader for the initial bootstrap of VMS on Itanium, I've not been able toJ >figure out what exactly this "HAL" thing/problem is.  I would *hope* that >*I* would know about it.r >a/ >Perhaps they are referring to our use of ACPI?a  ; Yes, I think they're referring to implementing an interfaceo3 to ACPI.  In NT this part of the code is called the ! Hardware Abstraction Layer (HAL).   5 And as you say, VMS Engineering has been aware of thet7 need for that code from the get-go.  It's by no means a4 'snagette'.i  
 ---------- Remove 'Z' to reply by email.    ------------------------------  % Date: Tue, 25 Jun 2002 21:13:34 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>i Subject: Re: VMS port delayed!, Message-ID: <3D191529.C027CFFA@videotron.ca>   Fred Kleinsorge wrote:K > figure out what exactly this "HAL" thing/problem is.  I would *hope* thatv > *I* would know about it.  N Ahhh, but this is where all the conspiracy theories come in. You see,  you areF purpusefully being misled and kept from the true facts in a covert (noK relation to John Covert) effort to ensure that the VMS engineers are poorlyoL coordinated and that the resulting port of VMS on that Intel vapourware will7 achieve the same quality levels as Microsoft's Windows.o  H Before long, when you call other engineers, the call will be routed to aL windows-based AI module that will emulate their voices so your conversationsJ won't ever really happen, leading to even more  dysfunction within the VMS engineering group.  I Luckily, Microsoft hasn't yet invented holographic persons. When they do,rI expect some of the VMS engineers to be adbucted by aliens and replaced by1N their holographic equivalents who will be controlled by Microsoft weenies fromJ Redmond. You know what will happen to VMS's reliability, security etc whenJ even just one of the VMS engineers will be replaced by a Microsoft Weenie.   ------------------------------   Date: 25 Jun 2002 21:03:25 GMT* From: psy@psy.bronderslev.dk (Bent Wagner)" Subject: Re: WASD startup problems3 Message-ID: <slrnahhqnh.7g6.psy@psy.bronderslev.dk>y  0 Mark Daniel <Mark.Daniel@wasd.vsm.com.au> wrote:    J >As you report in a later item you have fixed the problem by starting the G >queue-manager.  From this I guess you are using a pre-VMS-6.2 version  F >(which uses the batch system to detach the process under a different I >account, post-6.2 that process is created directly by the server image).o  8 You are right, i should have said that i use VAX-VMS 6.2  I >If the VMS version guess is correct you also need the WASD_HTTPD_GMT to  3 >tell the startup procedure what the GMT offset is J   Yes, i have that one too.   9 >There is also the info-WASD mailing list available from tJ >http://wasd.vsm.com.au/other/info-WASD.html if you are planning on using  >WASD to any extent.   I will look into this.   Thank you Mark.a     Bent Wagner,   ------------------------------  + Date: Tue, 25 Jun 2002 18:21:00 +0000 (UTC)a* From: bleau@umtof.umd.edu (Lawrence Bleau)5 Subject: Re: Why does SMTP RBL filter appear to fail?80 Message-ID: <afacac$qi2$1@grapevine.wam.umd.edu>  H In article <3D186960.4DACCE92@fdf.com>, afdfad <mfsggfs@fdf.com> writes: >Lawrence Bleau wrote:G >> An earlier poster corrected my misimpression about the usefulness of Q >> Good-Clients and Bad-Clients.  Perhaps you labor under the same misimpression.vP >> I was informed that, if mail relaying is *enabled*, then it is not permmittedF >> for systems on the Bad-Clients unless unless they also appar on theO >> Good-Clients list.  If mail relaying is *disabled*, then these two lists areu. >> not even used, as no mail relaying is done. >> eR >> In short, Bad-Clients is not a list of systems to exclude; for that, one should7 >> use the Reject-Mail-From and Accept-Mail-From lists.e > F >In order for one to use POP mail, relaying must be ENABLED.  However,F >with my configuration, I am rejecting all mail from the ones I listed! >where my smtp.config looks like:g  G Hi, I'm the original poster.  With all due respect to folks who want tooK discuss email relays and POP usage, that is *not* what my problem is.  RBLsMJ are for spam sites, not open relay sites per se.  You can have open relaysE that aren't spammers, and spammers that aren't open relays.  They are,G different.  The Good-Clients and Bad-Clients lists are red herrings wrt  RBLs.   J I am not using POP mail, and relaying is disabled.  Although I'll admit asJ a separate issue this is worth following through on; perhaps if you change( the subject line and start a new thread.  E My post was about why the RBL list (realtime blockhole list) does noteH appear to work.  I can repost the details if you can't find the original post.c   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edut   ------------------------------   End of INFO-VAX 2002.348 ************************