INFO-VAX Mon, 04 Jun 2007 Volume 2007 : Issue 303 Contents: Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: HP wasting millions of dollars on itanium! Re: Infoserver 150 woes Re: Infoserver 150 woes Re: OpenVMS Support for C-class Blades Question about Mozilla crash (Flash) Re: Whom administers openvmshobbyist.org forums? ---------------------------------------------------------------------- Date: Sun, 03 Jun 2007 17:28:57 -0400 From: JF Mezei Subject: Re: HP wasting millions of dollars on itanium! Message-ID: <63b6e$466332c8$cef8887a$23356@TEKSAVVY.COM> Arne Vajhøj wrote: > The point is that it is not possible to go from whatever nm > (I think someone said 180) to 45 nm and from 1 core to 16 core > is not something you just do. Funny, this is something that is happening in the 8086 marketplace which is now up to 4 64-bit cores for what used to be a toy controller. Where there is a will, there is usually a way (and accompanying budgets). > You start with a white piece of paper and X years later you > have a CPU in production. If you start with an existing core, and want to just produce it with a newer/smaller process, you don't start with a blank sheet of paper. Restarting Alpha need not involve 5 years before a new chip is produced. A speed bump on existing process is likely possible. Then you can work on a process shrink in a year or two. And that would span the time until an revamped EV8 could come out. Also, I would not be surprised to find out that development efforts for EV7 had not been given high priority under Curly (and perhaps even under Palmer who had done all those deals giving Intel access to Alpha technology). Consider that Intel is now in a path where it hopes to get new 8086 cores every 2 years. Would it be correct to state that in terms of design/logic/technology, Alpha is not yet behind, but it is behind in process ? If, today, Alpha were to be given a process shrink to current capabilities, would it be in the race ? Or is it now a requirement to be multi-core to be in the race ? ------------------------------ Date: Sun, 03 Jun 2007 17:38:52 -0400 From: JF Mezei Subject: Re: HP wasting millions of dollars on itanium! Message-ID: <8da1e$4663351b$cef8887a$27242@TEKSAVVY.COM> Arne Vajhøj wrote: > I just doubt that HP or the ISV's are that interested in porting. If VMS were to be ported to the 8086 architecture, it would send a very strong signal that HP is serious about growing VMS. It would also put VMS on a platform whose future is not in question. It would be on an equal footing with Linux, Solaris and Windows as well as many other operating systems. Like it or not, IA64 has been plagued by rumours of its demise even before Merced first booted Windows which is one reason the announcement of porting VMS to IA64 was so poorly received. For all its technical weaknesses, the 8086 has market share and an assured future. The saying used to be: Nobody has ever been fired for choosing IBM The saying is now: Nobody will be fired for porting to the 64 bit 8086 architecture. And right now, HP's refusal to port VMS to a viable platform is also interpreted as HP abandonning VMS and allowing key ISVs like Cerner to drop ou of VMS. HP has no duty to help Intel's financials. HP has a duty to its own shareholders and its own customers. Don,t force customers to a platform they don,t want just because you want to please Intel's CEO. ------------------------------ Date: Sun, 03 Jun 2007 17:02:54 -0500 From: Ron Johnson Subject: Re: HP wasting millions of dollars on itanium! Message-ID: On 06/03/07 16:38, JF Mezei wrote: > Arne Vajhøj wrote: > >> I just doubt that HP or the ISV's are that interested in porting. > > > If VMS were to be ported to the 8086 architecture, it would send a very > strong signal that HP is serious about growing VMS. It would also put > VMS on a platform whose future is not in question. It would be on an > equal footing with Linux, Solaris and Windows as well as many other > operating systems. I *guarantee* you that VMS, Linux, Solaris and WNT will *never* be ported to the 8086. Never. Ever. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Sun, 03 Jun 2007 18:13:14 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: HP wasting millions of dollars on itanium! Message-ID: <46633cf0$0$90276$14726298@news.sunsite.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> I just doubt that HP or the ISV's are that interested in porting. > > If VMS were to be ported to the 8086 architecture, it would send a very > strong signal that HP is serious about growing VMS. It would also put > VMS on a platform whose future is not in question. Yes. But that is not enough reason to choose the platform. > It would be on an > equal footing with Linux, Solaris and Windows as well as many other > operating systems. No. There are still a few issues like apps and people. Arne ------------------------------ Date: Sun, 03 Jun 2007 18:11:15 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: HP wasting millions of dollars on itanium! Message-ID: <46633c7b$0$90276$14726298@news.sunsite.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> The point is that it is not possible to go from whatever nm >> (I think someone said 180) to 45 nm and from 1 core to 16 core >> is not something you just do. > > Funny, this is something that is happening in the 8086 marketplace which > is now up to 4 64-bit cores for what used to be a toy controller. No. Both Intel and AMD had to come up with new designs to support that. > Where there is a will, there is usually a way (and accompanying budgets). True. But it takes time. >> You start with a white piece of paper and X years later you >> have a CPU in production. > > If you start with an existing core, and want to just produce it with a > newer/smaller process, you don't start with a blank sheet of paper. Maybe not blank. But I am not convinced that it is so simple. > Restarting Alpha need not involve 5 years before a new chip is produced. > A speed bump on existing process is likely possible. Then you can work > on a process shrink in a year or two. And that would span the time until > an revamped EV8 could come out. > > Also, I would not be surprised to find out that development efforts > for EV7 had not been given high priority under Curly (and perhaps even > under Palmer who had done all those deals giving Intel access to Alpha > technology). Consider that Intel is now in a path where it hopes to get > new 8086 cores every 2 years. > > Would it be correct to state that in terms of design/logic/technology, > Alpha is not yet behind, but it is behind in process ? > > If, today, Alpha were to be given a process shrink to current > capabilities, would it be in the race ? > > Or is it now a requirement to be multi-core to be in the race ? I think multi core is mandatory for server CPU's today. Arne ------------------------------ Date: Sun, 03 Jun 2007 22:29:18 GMT From: "John Wallace" Subject: Re: HP wasting millions of dollars on itanium! Message-ID: <2hH8i.62552$4a.51758@fe3.news.blueyonder.co.uk> "Bill Todd" wrote in message news:D66dnbYqebNvyf_bnZ2dnUVZ_hisnZ2d@metrocastcablevision.com... > 1. AMD's on-chip memory controller does not IIRC support as much memory > per chip, as much memory bandwidth per chip (though I'd have to check to > see where their recent migration to DDR-2 changed this), or (mirrored or > parity) redundant memory. > Not sure I've read you right here, but someone else has apparently read it the same way I did... When the question of "what's missing from Opteron RAS that Itanium has?" last arose and was as usual largely unanswered in any real detail, I did a bit of research. I've since forgotten the details, but do recollect that Opteron Revision F has some small but worthwhile improvements in terms of main-memory RAS vs its predecessors. A real commercial OS implementors view of some of those improvements can be found on Gavin Maltby's blog entry at http://blogs.sun.com/gavinm/entry/fma_support_for_amd_opteron which, if I read it correctly, describes a variety of recent Opteron-related changes including the concept of online spare memory chips. It's worth a look if you haven't already been there. Given that Opteron main memory already (aiui) had ECC capability for the usual historic reasons, and (according to Maltby) the current hardware already appears able to deliver other useful RAS functionality once suitable OS support exists e.g. Chipkill/Chipspare-like functionality, is there really any significant market outside the Nonstop world willing to pay for fully mirrored memory? Does Itanic offer it, did EV7 offer it? Does it matter :) ------------------------------ Date: Mon, 04 Jun 2007 00:59:50 +0200 From: Michael Kraemer Subject: Re: HP wasting millions of dollars on itanium! Message-ID: JF Mezei schrieb: > > If VMS were to be ported to the 8086 architecture, it would send a very > strong signal that HP is serious about growing VMS. > > And right now, HP's refusal to port VMS to a viable platform is also > interpreted as HP abandonning VMS and allowing key ISVs like Cerner to > drop ou of VMS. > > HP has no duty to help Intel's financials. HP has a duty to its own > shareholders and its own customers. And what if they carry out their duty by *not* burning money in an inefficient effort to convince people to use VMS, be it on Itanic or x86 or whatever ? Given the marginal (if any) incentives this OS has to offer over other platforms these days ? I think this is what HP managers have already expressed publicly: it's not worthwhile. And so they offer lacklustre committment by porting it to an architecture about as dead as alpha, but at least they might keep rich customers like US military and a handful of stock exchanges happy. Good enough for their stock holders, and who cares about smaller customers. I bet that, if HP would decide to withdraw VMS from market tomorrow, just providing minimal support for selected rich customers, they still would make sizeable profits. And probably they wouldn't still stick to their contractual obligations. > Don,t force customers to a platform > they don,t want just because you want to please Intel's CEO. And just why should they please the inmates of c.o.v ? ------------------------------ Date: Mon, 04 Jun 2007 00:24:45 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: Infoserver 150 woes Message-ID: In article , Dan Foster wrote: >In article , Curtis Rempel wrote: >>> >>> Thank you thank you thank you! That did the trick! Interestingly, >>> that flag isn't really 'documented' in any InfoServer literature I >>> could find - mentioned, but not discussed or described. >> >> Have a look in the "InfoServer System Operations Guide", order number >> AA-PJXJC-TE. Mine is October 1994. Pages 5-16 and 5-17 in the >> troubleshooting section talk about the boot flags. > >That particular manual (also October 1994) can be found here: > >http://www1.aclabs.com/MasterIndex/installation_guide/installation_guide_006a2764.txt > >Just bring up this document then do a browser search function for '5-16' >and you'll quickly land on the correct spot. I *highly* recommend Curtis >and other interested people consider saving a copy of this locally as >this stuff is becoming non-easy to find as time goes on. > >They also have a huge number of manuals for various non-current VMS >stuff also in text format, just one step higher in the URL hierarchy. > >That's about 1,400 manuals comprising about 200 MB. Most of them are >from the late '80s and early '90s, though there are some from later. > >-Dan > >P.S. Thank you, AC Labs. I'll pass that along. >P.P.S. Thank you, Digital. What does Digital have to do with it? ------------------------------ Date: Sun, 03 Jun 2007 22:20:03 -0500 From: Dan Foster Subject: Re: Infoserver 150 woes Message-ID: In article , Robert Deininger wrote: > >>P.P.S. Thank you, Digital. > > What does Digital have to do with it? They created the documentation. Digital may not exist as a company, but the product of their labors lives on, and is still useful to this day. -Dan ------------------------------ Date: Mon, 04 Jun 2007 00:30:43 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: OpenVMS Support for C-class Blades Message-ID: In article , "Martin Vorlaender" wrote: >JF Mezei wrote: >> Does anyone know if they have added specific support for cluster >> interconnect between IA64 cards in a c-class box ? Or is it still just >> ethernet through the normal ethernet cards assigned to each slot in the >> C box ? > >That specific question came up in one of the bootcamp's blade sessions. >IIRC, there's no support for an in-box interconnect. The in-box ethernet interconnect is supported. What other in-box interconnect are folks talking about here? ------------------------------ Date: Sun, 03 Jun 2007 23:14:15 -0400 From: JF Mezei Subject: Question about Mozilla crash (Flash) Message-ID: <405c6$466383b5$cef8887a$13809@TEKSAVVY.COM> The http://www.theinquirer.net often causes Mozilla to crash. This also happened with Netscape 4.8 on an old mac. I have the output below. Can anyone confirm that it crashes while trying to process a JPEG image inside a flash module ? (I ask this so I can complain to the inquirer webmaster and provide as many details as possible). > Improper call to JPEG library in state 202 > Unable to read JPEG data > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000000000000 > 0000, PC=0000000008540FD4, PS=0000001B > %TRACE-F-TRACEBACK, symbolic stack dump follows > image module routine line rel PC abs PC > libflash4plugin DISPLAYLIST bbox 11152 0000000000000284 0000000008540FD4 > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=000000010000 > 0000, PC=0000000100000000, PS=0000001B > > Improperly handled condition, image exit forced. > Signal arguments: Number = 0000000000000005 > Name = 000000000000000C > 0000000000010000 > 0000000100000000 > 0000000100000000 > 000000000000001B > > Register dump: > R0 = 0000000000000001 R1 = 0000000000000001 R2 = 000000007BF68590 > R3 = 000000007AE34500 R4 = 000000007AE344F0 R5 = 000000007AE344D8 > R6 = 000000007AE34570 R7 = FFFFFFFF8194DF20 R8 = 000000000E3DBE58 > R9 = 0000000000000001 R10 = 0000000000008000 R11 = FFFFFFFFFFFF8000 > R12 = 000000000852A310 R13 = FFFFFFFF8194E3F0 R14 = 0000000000000000 > R15 = 0000000000000000 R16 = 0000000000000EE0 R17 = FFFFFFFF77773700 > R18 = 002B6D980939E780 R19 = 000000007AE34240 R20 = 00000000000052EA > R21 = 000000007B66A848 R22 = 002B6D980BB7BF68 R23 = 000000007AE34230 > R24 = 000000000850A000 R25 = 0000000000000001 R26 = 0000000100000001 > R27 = 000000007B62D590 R28 = 000000007BF7C660 R29 = 000000007AE34200 > SP = 000000007AE34200 PC = 0000000100000000 PS = 000000000000001B > Mozilla stopped at 3-JUN-2007 23:10:39.84 ------------------------------ Date: Sun, 03 Jun 2007 15:45:10 -0500 From: Dan Foster Subject: Re: Whom administers openvmshobbyist.org forums? Message-ID: In article , Tom Linden wrote: > On Sun, 03 Jun 2007 02:06:28 -0700, Dan Foster wrote: > >> I am in need of contacting a forum administrator for openvmshobbyist.org >> to resolve a minor technical issue, but there is no contact information >> listed anywhere on the forums, nor even a name, other than the >> omnipotent 'Board Administrator'. :-) >> >> If someone knows whom administers that site, please contact me privately >> or post here -- doesn't matter. >> >> Thanks! >> >> -Dan > > > who not whom, it is the subject not the object > Thank you -- consider myself duly chastised and corrected. :-) (Likewise, a thank you also proffered to others whom mentioned it.) -Dan ------------------------------ End of INFO-VAX 2007.303 ************************