INFO-VAX Mon, 12 Feb 2007 Volume 2007 : Issue 86 Contents: Re: DVD writer for DS10L ? Re: Guidelines for converting programs to ODS-5? Re: Guidelines for converting programs to ODS-5? Re: Help wanted with Pathworks 5.0a Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Intel 80 core chip revealed in full detail Re: Mark Daniel and/or Hein van den Huevel or anyone really Re: Migrating C application from VMS to LINUX Re: Migrating C application from VMS to LINUX Re: SPANNING BACKUP TAPES Re: Understanding DSL10L cabinet/hardware Re: Understanding DSL10L cabinet/hardware Re: Understanding DSL10L cabinet/hardware Re: Understanding DSL10L cabinet/hardware Re: Understanding DSL10L cabinet/hardware ---------------------------------------------------------------------- Date: Mon, 12 Feb 2007 07:07:56 -0500 From: JF Mezei Subject: Re: DVD writer for DS10L ? Message-ID: Ok, back to this topic. Island's offering is a DVD-Read and CD read-write. I looked around for IDE Slimline CD/DCD RW. One example: http://www.logicsupply.com/product_info.php/cPath/27/products_id/673 Is there documentation on which drives (or what features are required) that would be supported by VMS 8.3 on Alpha ? I took a peek inside a DS10L, and it appears like the second drive bay can support a slimline device on the top with some screw holes which I would assume/hope are standard, Those slimline drives seem to have a proprietary plug, so I need to buy an adaptor between slimline and normal IDE cable (as well as an adaptor from conventional power to that adapter's "mini" plug for floppies. Is IDE pretty standard in terms of protocols so that all IDE CD/DVD devices appear to be the same to VMS in terms of software/drivers ? om6 posted stuff before stating it appears to be pretty much plug and play. But I think he was refering to an external SCSI drive. ------------------------------ Date: 12 Feb 2007 07:58:14 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Guidelines for converting programs to ODS-5? Message-ID: <5BbeoVmaJGWR@eisner.encompasserve.org> In article , Paul Sture writes: > Hi folks, > > Are there any guidelines out there for ensuring that programs developed > before the days of ODS-5 are ODS-5 aware? Programs developed before ODS-5 should run just fine on ODS-5 without any modification unless you specifically turn on the new per-process case-sensitive option. What you have to look out for are programs that assume they know that VMS file names are done in a specific way. I had a program developed under ODS-2 that wouldn't let me type a lowercase character despite case insensitivity of the OS. I fixed it when ODS-2 added a couple characters that program was prohibiting. The program should have taken the name as given and dealt with bad names only if the open failed. ------------------------------ Date: Mon, 12 Feb 2007 09:58:37 -0600 (CST) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Guidelines for converting programs to ODS-5? Message-ID: <07021209583700_202B6533@antinode.org> From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) > Programs developed before ODS-5 should run just fine on ODS-5 without > any modification unless you specifically turn on the new > per-process case-sensitive option. Not if you give them ODS5 extended file names to work with, obviously, which seems almost inevitable if you use them with an ODS5 file system. Or did you mean to say something which made some sense? ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: 12 Feb 2007 00:50:07 -0800 From: "Wilm Boerhout" Subject: Re: Help wanted with Pathworks 5.0a Message-ID: <1171270207.801035.32910@v33g2000cwv.googlegroups.com> On 11 feb, 10:38, "Antoniov" wrote: > 1) The PCs in not registered in VMS license [See pwrk event on PC] > 2) The user of PC is not registered in VMS database (or password is > changed) [Try access via NET LOGON on VMS box] > 3) PC used NetBEUI and now NetBEUI is disabled on PC [Check for > MyNetwork property on PC] > 4) DNS in not right configurated [type ping vms_servername from PC] > 5) PC now cannot access to uncripted channel [PC controlpanel] Resolved. It turned out to be number 4 in your list! Apparently, Pathworks somehow does a lookup of its server name in DNS. On friday night, this still translated to the old IP address. This morning, with DNS entries properly replicated throughout the customers's worldwide network, access to the Pathworks shares turned out to be possible. Thanks to all who replied for your support! /Wilm ------------------------------ Date: 12 Feb 2007 05:58:57 -0800 From: "n.rieck@sympatico.ca" Subject: Intel 80 core chip revealed in full detail Message-ID: <1171288737.922279.230170@s48g2000cws.googlegroups.com> Intel 80 core chip revealed in full detail http://www.intel.com/research/platform/terascale/teraflops.htm?iid=homepage+80core http://www.theinquirer.net/default.aspx?article=37572 Wether "marketing" or "marketecture", this makes you wonder where they'll be in 36 months. Neil Rieck http://www3.sympatico.ca/n.rieck/ ------------------------------ Date: 12 Feb 2007 14:23:55 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Intel 80 core chip revealed in full detail Message-ID: <53bbjrF1s00vlU1@mid.individual.net> In article , "Richard Brodie" writes: > > wrote in message > >> Wether "marketing" or "marketecture", this makes you wonder where >> they'll be in 36 months. > > ...enough for me not to take some AMD fans too seriously when > they write off Intel for not having a quad core chip out yet. You do realize this is not 80 Pentiums. I would not hold my breath waiting for it. This is not something targeted at the PC market which is where Intel still has to compete with AMD. Intel has always made other chips. Most were for small niche markets and never gained real popularity outside their niche (hmmmm.... sounds like Itanium!) bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Mon, 12 Feb 2007 09:29:50 -0500 From: JF Mezei Subject: Re: Intel 80 core chip revealed in full detail Message-ID: n.rieck@sympatico.ca wrote: > Wether "marketing" or "marketecture", this makes you wonder where > they'll be in 36 months. Note that this 80 core thing is not an 8086 architecture, it is some primitive architecture. And it is unable to actually do anything useful because they have yet to develop a way to hook 80 cores to memory. This is a nice research project. But it does nothing to she light on what is in store for IA64 versus 8086. ------------------------------ Date: Mon, 12 Feb 2007 14:44:10 -0000 From: "Richard Brodie" Subject: Re: Intel 80 core chip revealed in full detail Message-ID: "Bill Gunshannon" wrote in message news:53bbjrF1s00vlU1@mid.individual.net... > You do realize this is not 80 Pentiums. I would not hold my breath > waiting for it. Of course, Bill. It's a research project, not a product. I'm just rather tired of all the "our next product will crush the opposition" claims prevalent in the industry. I look forward to some healthy competition in the next few years. If AMD can pull ahead of Intel a bit, good luck to them. How long they stay that way, I'd not like to put money on. ------------------------------ Date: 12 Feb 2007 15:08:26 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Intel 80 core chip revealed in full detail Message-ID: <53be7aF1sakmmU1@mid.individual.net> In article , "Richard Brodie" writes: > > "Bill Gunshannon" wrote in message > news:53bbjrF1s00vlU1@mid.individual.net... > >> You do realize this is not 80 Pentiums. I would not hold my breath >> waiting for it. > > Of course, Bill. It's a research project, not a product. I'm just > rather tired of all the "our next product will crush the opposition" > claims prevalent in the industry. I look forward to some healthy > competition in the next few years. If AMD can pull ahead of > Intel a bit, good luck to them. How long they stay that way, > I'd not like to put money on. AMD only needs to stay ahead of (or even with) Intel in the commodity market. You don't see them trying to best Intel over the IA64? Intel doing other processors (as they have all along) just requires them to have more resources (and the associated costs) to support them. Look at how they had to scale back Itanium development in order to stay in the game with AMD over the Pentium class processors. This 80 core thing is just another red-herring. It means nothing in the market where Intel and AMD compete. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: 12 Feb 2007 08:33:05 -0800 From: "n.rieck@sympatico.ca" Subject: Re: Intel 80 core chip revealed in full detail Message-ID: <1171297984.814812.294070@a34g2000cwb.googlegroups.com> On Feb 12, 9:29 am, JF Mezei wrote: > n.ri...@sympatico.ca wrote: > > Wether "marketing" or "marketecture", this makes you wonder where > > they'll be in 36 months. > > Note that this 80 core thing is not an 8086 architecture, it is some > primitive architecture. And it is unable to actually do anything useful > because they have yet to develop a way to hook 80 cores to memory. > > This is a nice research project. But it does nothing to she light on what > is in store for IA64 versus 8086. > You are correct. But this announcement is no different than the time IBM's "deep blue" beat Kasparov at chess. You couldn't buy one of those machines either but IBM stock jumped 35% on the announcment. Neil Rieck http://www3.sympatico.ca/n.rieck/ ------------------------------ Date: Mon, 12 Feb 2007 08:31:31 -0800 From: "Tom Linden" Subject: Re: Intel 80 core chip revealed in full detail Message-ID: On Mon, 12 Feb 2007 08:33:05 -0800, n.rieck@sympatico.ca wrote: > On Feb 12, 9:29 am, JF Mezei wrote: >> n.ri...@sympatico.ca wrote: >> > Wether "marketing" or "marketecture", this makes you wonder where >> > they'll be in 36 months. >> >> Note that this 80 core thing is not an 8086 architecture, it is some >> primitive architecture. And it is unable to actually do anything useful >> because they have yet to develop a way to hook 80 cores to memory. >> >> This is a nice research project. But it does nothing to she light on >> what >> is in store for IA64 versus 8086. >> > You are correct. But this announcement is no different than the time > IBM's "deep blue" beat Kasparov at chess. You couldn't buy one of > those machines either but IBM stock jumped 35% on the announcment. Actually, I think they did sell some such systems based on the T2 fabric switch. > > Neil Rieck > http://www3.sympatico.ca/n.rieck/ > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 12 Feb 2007 10:47:43 -0600 From: Dan Foster Subject: Re: Intel 80 core chip revealed in full detail Message-ID: In article <1171297984.814812.294070@a34g2000cwb.googlegroups.com>, n.rieck@sympatico.ca wrote: > On Feb 12, 9:29 am, JF Mezei wrote: >> n.ri...@sympatico.ca wrote: >> > Wether "marketing" or "marketecture", this makes you wonder where >> > they'll be in 36 months. >> >> Note that this 80 core thing is not an 8086 architecture, it is some >> primitive architecture. And it is unable to actually do anything useful >> because they have yet to develop a way to hook 80 cores to memory. >> >> This is a nice research project. But it does nothing to she light on what >> is in store for IA64 versus 8086. >> > You are correct. But this announcement is no different than the time > IBM's "deep blue" beat Kasparov at chess. You couldn't buy one of > those machines either but IBM stock jumped 35% on the announcment. Well... I'd have to beg to differ, simply because my employer did, in fact, own one of these SP/2 systems at the time. :) (Guess who got to administer it?) Deep Blue rematch was in late '96; IBM brought out the SP (and its successor, the SP/2) a few years earlier, I think. We got our SP/2 in early '96. Though, Deep Blue was special; they had experienced developers (with deep MPI / parallel coding experience) working on it as well as 512 chips dedicated to nothing but chess computations. The chess-specific ASICs were pretty much unique and one-offs. Everything else was more-or-less standard and purchasable. The MPI stuff was popular with other SP/2 customers whom tended to use its scalability to do massively parallel stuff -- oil/gas exploration, weather forecasting, nuclear weapon research, etc. SP/2 was a good fit because it had a bounded latency with hard guarantees for inter-node communications regardless of if it was 1 node or 1024, amongst other things. Made certain aspects of parallel programming easier for planning the message passing and coordination and being able to use predictable, bounded O() figures in planning. (IBM capped the validated setup at 1024 nodes, though, because the latency was starting to climb past the hard guaranteed times amongst other reasons. Like, not many customers really wanted more than 1024. Cornell had such a large setup for their CITC center.) I now return you to comp.os.vms pontificating. ;) -Dan ------------------------------ Date: Mon, 12 Feb 2007 22:29:29 +0800 From: "Richard Maher" Subject: Re: Mark Daniel and/or Hein van den Huevel or anyone really Message-ID: Hi Ken, I liked the MAXBUF theory but no joy unfortunately. Cheers Richard Maher wrote in message news:1170859488.729794.87480@p10g2000cwp.googlegroups.com... > On Feb 5, 7:37 pm, "Richard Maher" > wrote: > > Gentlemen, Good-morning, > > > > Can you please help me with a little Webserver/RMS conundrum? > > > > I an using a Mickey Mouse DCL http server (see below) listening on port 80 > > to transfer some html files and a Java Applet Archive file. Once the file > > exceeds somewhere between 7231 and 7244 bytes big, the receiving IE Browser > > on W2K falls over with Unexpected end of file on Socket and a subsequent > > CLassNotFound error :-( > > > > Hein > > ----- > > > > The .JAR file has been $set file/attr=(rfm:udf) and as you can see is a > > straight forward DCL copy to the _BG: device. Anything RMS might do with > > fragmentation? > > > > Mark > > ------ > > > > Does something other that a standard streamed socket read on the client, and > > an RMS copy on the server, have to happen here to satisfy HTTP semantics? Am > > I hitting a socket quota or TCP/IP seg size here? > > > > Everyone: > > ---------- > > > > I really don't want to have to install and maintain a full-blown Webserver > > just to test my applets so if someone knows of a SYSGEN, TCP/IP or Quota I > > should be looking at and adjusting then please let me know! > > > > Cheers Richard Maher > > Here are a couple of suggestions, perhaps not helpful... > > 1) What value do you have for the sysgen parameter MAXBUF? Most > systems I have seen have this set at 8192. This is a dynamic > parameter; should be simple to change and test it. One can not $TYPE > out a file with a record size that exceeded MAXBUF, so perhaps it's > related??? > > 2) You did not say what TCP/IP package you are using; for TCP/IP > services from HP, you can look at $TCPIP SHOW PROTOCOL TCP /PARAMETER > and look at the SEND quota. New sockets that are created should > inherit this value. > > Ken Randell > > You did not say what TCP/IP package you are using, but > ------------------------------ Date: 12 Feb 2007 07:35:55 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Migrating C application from VMS to LINUX Message-ID: In article <53627pF1r0u3cU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: > I was refering to things like RMS calls and QIO calls, which define "VMS" > vs. "C" programs. Most of the non-VMS "C" programs I've dealt with were full of UNIX or Windows calls (not those included in the ANSI library). So I think the other fellow was right about there not being many "C" programs out there by that standard. ------------------------------ Date: 12 Feb 2007 07:38:06 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Migrating C application from VMS to LINUX Message-ID: In article <45CE3C43.6080500@comcast.net>, "Richard B. gilbert" writes: > > I think the difference is that a C program that calls only the standard > C library routines is portable. When a C program directly calls VMS > system services or RMS services or VMS RTL routines, it is no longer > readily portable. The same is true of any C program that makes calls > into any O/S specific service or RTL routine. > > Some O/S dependencies are fairly easy to fix since many O/Ss provide > similar services in similar ways. Others may not be so easy. I agree, but I'd say the same thing of UNIX or Windows calls more strongly than you did. And I'd say the same thing about Fortran, PASCAL, COBOL, ... ------------------------------ Date: 12 Feb 2007 07:41:13 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: SPANNING BACKUP TAPES Message-ID: In article , Paul Sture writes: > > Were failures with Legato recoverable? Sadly that is still under investigation. Several months ago I would have said yes, but that's not what I'm seeing now. May be a Legato problem, or may be a personnel problem. Not my personnel. ------------------------------ Date: Mon, 12 Feb 2007 14:56:01 +0100 From: Michael Unger Subject: Re: Understanding DSL10L cabinet/hardware Message-ID: <53bh9eF1s1pfiU1@mid.individual.net> On 2007-02-12 13:15, "JF Mezei" wrote: > [...] > > I have to conclude that it is a glorified resistance designed to generate heat. Or to provide the "base load" absolutely required by most of the switch-mode power supplies to prevent the output voltage from going above the upper limit ... > Does anyone know if this gizmo has a real purpose in life ? It is attached > to a removable assembly, which, when removed, gives you just a normal power > connector of a length sufficient to reach a drive in the drive bay. > > AKA, plugged into the normal power cable is a resistor. (and a power outlet). Which might just have been the cheapest solution to the problem ... > [...] Michael -- Real names enhance the probability of getting real answers. My e-mail account at DECUS Munich is no longer valid. ------------------------------ Date: 12 Feb 2007 09:34:10 -0800 From: "Rich Jordan" Subject: Re: Understanding DSL10L cabinet/hardware Message-ID: <1171301650.445957.127200@a34g2000cwb.googlegroups.com> On Feb 12, 7:56 am, Michael Unger wrote: > On 2007-02-12 13:15, "JF Mezei" wrote: > > > [...] > > > I have to conclude that it is a glorified resistance designed to generate heat. > > Or to provide the "base load" absolutely required by most of the > switch-mode power supplies to prevent the output voltage from going > above the upper limit ... > > > Does anyone know if this gizmo has a real purpose in life ? It is attached > > to a removable assembly, which, when removed, gives you just a normal power > > connector of a length sufficient to reach a drive in the drive bay. > > > AKA, plugged into the normal power cable is a resistor. (and a power outlet). > > Which might just have been the cheapest solution to the problem ... > > > [...] > > Michael > > -- > Real names enhance the probability of getting real answers. > My e-mail account at DECUS Munich is no longer valid. The SZ12/BA42 cabinets for external disks in the VS/MV3100 timeframe had a large heat-sinked resistor mounted in the chassis for this purpose. Rich ------------------------------ Date: Mon, 12 Feb 2007 18:16:43 GMT From: Tad Winters Subject: Re: Understanding DSL10L cabinet/hardware Message-ID: Michael Unger wrote in news:53bh9eF1s1pfiU1@mid.individual.net: > On 2007-02-12 13:15, "JF Mezei" wrote: > >> I have to conclude that it is a glorified resistance designed to >> generate heat. > > Or to provide the "base load" absolutely required by most of the > switch-mode power supplies to prevent the output voltage from going > above the upper limit ... > [..snip..] What a terrible idea for a system that would likely be packed into a rack with many other system, all generating too much heat. An extra fan would be a much better choice, even if it costs a little more. Tad ------------------------------ Date: Mon, 12 Feb 2007 19:03:36 GMT From: "FredK" Subject: Re: Understanding DSL10L cabinet/hardware Message-ID: "Tad Winters" wrote in message news:Xns98D5688FC5952staffordnospamwinter@130.81.64.196... > Michael Unger wrote in > news:53bh9eF1s1pfiU1@mid.individual.net: > >> On 2007-02-12 13:15, "JF Mezei" wrote: >> >>> I have to conclude that it is a glorified resistance designed to >>> generate heat. >> >> Or to provide the "base load" absolutely required by most of the >> switch-mode power supplies to prevent the output voltage from going >> above the upper limit ... >> > [..snip..] > > What a terrible idea for a system that would likely be packed into a rack > with many other system, all generating too much heat. An extra fan would > be a much better choice, even if it costs a little more. > Unpredictable load, and subject to failure. Short of smashing it with a hammer, I've never seen one of those big-ass resistors fail. ------------------------------ Date: Mon, 12 Feb 2007 18:43:59 -0000 From: "John Wallace" Subject: Re: Understanding DSL10L cabinet/hardware Message-ID: <45d0b56f$0$8710$ed2619ec@ptn-nntp-reader02.plus.net> "Tad Winters" wrote in message news:Xns98D5688FC5952staffordnospamwinter@130.81.64.196... > What a terrible idea for a system that would likely be packed into a rack > with many other system, all generating too much heat. An extra fan would > be a much better choice, even if it costs a little more. > Extra fan doesn't solve the real problem, which is that the switched mode power supplies commonly used to provide high efficiency in electronic equipment don't work right (may not work at all) if the load doesn't draw enough current. Anyone who's ever tried a BA23 chassis with a single low power card in it (eg KXT11 Falcon) may have come across this oddity, though it isn't a DEC-specific problem. Adding a dummy load is certainly a simple and probably relatively reliable approach to making sure enough current is drawn, though it does sound rather like a last-minute fix and as noted doesn't do much for heat load. As an outsider, I might speculate that maybe someone forgot to specify the minimum likely operating load for the PSU, and by the time the omission was noticed (e.g. when small configuration DS10s didn't power up), it was too late to redesign the (bought-in) PSU to handle a realistic minimum load current, so a cheap and not very cheerful "workaround" was identified to ensure that the minimum load current was always in the PSU's "safe zone". But that's pure speculation. ------------------------------ End of INFO-VAX 2007.086 ************************