From msaqer at iastate.edu Sat Jul 8 18:22:12 2006 From: msaqer at iastate.edu (Mohamad Al-Saqer) Date: Sat, 8 Jul 2006 13:22:12 -0500 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB Message-ID: <200607081322.12468.msaqer@iastate.edu> Is it possible to find an AMD 64 socket 939 motherboard which supports 4 memory sticks of 4GB to run at max speed 400MHz? From some available MB manuals the speed is either downgraded to 333MHz when 4 mem sticks are used or there is no mention of the issue at all. >From my search on google, some blame it on limitation of the motherboard chipset. Others blame it on processors prior to the E revision. Nothing on the net I came across would be a clear answer. I would suspect that tyan K8E (and K8E-SLI) doesn't have this limitation, does it? Because I saw at amd website some bechmarks on systems built on this motherboard with 4GB installed. However no say on wether these system were running at 400MHz FSB or less. I appreciate help and any clarification. From b.j.smith at ieee.org Sat Jul 8 18:43:11 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Sat, 08 Jul 2006 14:43:11 -0400 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: <200607081322.12468.msaqer@iastate.edu> References: <200607081322.12468.msaqer@iastate.edu> Message-ID: <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> On Sat, 2006-07-08 at 13:22 -0500, Mohamad Al-Saqer wrote: > Is it possible to find an AMD 64 socket 939 motherboard which supports 4 > memory sticks of 4GB to run at max speed 400MHz? From some available MB > manuals the speed is either downgraded to 333MHz when 4 mem sticks are used > or there is no mention of the issue at all. JEDEC specifications require that only one (1) DDR 200MHz (DDR400/PC-3200) 64-bit DIMM be used per 64-bit channel (2x32-bit banks), or two (2) for registered per 64-bit channel (4x32-bit banks). That means a maximum of (2) DDR DIMMs for unregistered Socket-939. That means a maximum of (4) DDR DIMMs for registered Socket-940. > From my search on google, some blame it on limitation of the motherboard > chipset. Others blame it on processors prior to the E revision. Nothing on > the net I came across would be a clear answer. Hit many discussions on JEDEC DDR specifications. Here's the rule ... DDR 100MHz (DDR200, PC1600) 3 DIMMs/channel, 6 DIMMs for S939, 12 DIMMs for S940 DDR 133/166MHz (DDR266/333, PC2100/2700) 2 DIMMs/channel, 4 DIMMs for S939, 8 DIMMs for S940 DDR 200MHz (DDR400, PC3200) 1 DIMM/channel, 2 DIMMs for S939, 4 DIMMs for S940 > I would suspect that tyan K8E (and K8E-SLI) doesn't have this limitation, does > it? Because I saw at amd website some bechmarks on systems built on this > motherboard with 4GB installed. However no say on wether these system were > running at 400MHz FSB or less. > I appreciate help and any clarification. As above. BTW, your greater concern than signaling is timing. Slower timed DDR400 DIMMs often cause more of an impact that faster timed DDR333 DIMMs. I personally use at least 2.5-3-3-6 for DDR400. If you're using 3-3-3-8 or worse for DDR400, you might as well run at DDR333. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com ---------------------------------------------------------- The existence of Linux has far more to do with the breakup of AT&T's monopoly than anything Microsoft has ever done. From loony at loonybin.org Sat Jul 8 18:53:56 2006 From: loony at loonybin.org (Peter Arremann) Date: Sat, 8 Jul 2006 14:53:56 -0400 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200607081453.56406.loony@loonybin.org> On Saturday 08 July 2006 14:43, Bryan J. Smith wrote: Thanks for the details - I had a hard time finding that... :-) > As above. > > BTW, your greater concern than signaling is timing. Slower timed DDR400 > DIMMs often cause more of an impact that faster timed DDR333 DIMMs. > > I personally use at least 2.5-3-3-6 for DDR400. > > If you're using 3-3-3-8 or worse for DDR400, you might as well run at > DDR333. Have you guys actually measured the impact of the memory speed? I'm running a few servers - dual 280s with 4GB ram each - that run queries against a smallisch in memory DB... this should be about as memory intensive as any application can get (unless you're running UMA graphics). Some systems have 266Mhz memory - others 400. It looks like even in this setup, all I can see is less than 3% difference... Peter. From hansel at hansel.mnstate.edu Sat Jul 8 19:06:18 2006 From: hansel at hansel.mnstate.edu (hansel at hansel.mnstate.edu) Date: Sat, 8 Jul 2006 14:06:18 -0500 (CDT) Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB In-Reply-To: <200607081322.12468.msaqer@iastate.edu> References: <200607081322.12468.msaqer@iastate.edu> Message-ID: On Sat, 8 Jul 2006, Mohamad Al-Saqer wrote: > Is it possible to find an AMD 64 socket 939 motherboard which supports 4 > memory sticks of 4GB to run at max speed 400MHz? From some available MB > manuals the speed is either downgraded to 333MHz when 4 mem sticks are used > or there is no mention of the issue at all. For what it is worth: The current Linux Journal is their "Ultimate Linux Box" issue, featuring AMD 939 dual core. They address the memory speed issue explicitly and claim a bios setting gets 400MHz with no glitches on the ABIT AN8 32x 939 board. The same article said that 3GB did not reset the bus to 333MHz. They also said that the real issue was populating all memory banks. Mark Hansel From b.j.smith at ieee.org Sat Jul 8 20:13:59 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Sat, 08 Jul 2006 16:13:59 -0400 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- timing is everything ... In-Reply-To: <200607081453.56406.loony@loonybin.org> References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> <200607081453.56406.loony@loonybin.org> Message-ID: <1152389639.2868.106.camel@bert64.oviedo.smithconcepts.com> On Sat, 2006-07-08 at 14:53 -0400, Peter Arremann wrote: > Have you guys actually measured the impact of the memory speed? I'm running a > few servers - dual 280s with 4GB ram each - that run queries against a > smallisch in memory DB... this should be about as memory intensive as any > application can get (unless you're running UMA graphics). Some systems have > 266Mhz memory - others 400. It looks like even in this setup, all I can see > is less than 3% difference... Depends on how much you write to memory (essentially no latency) and how much you read from memory, it's burst length (size) and how random they are (the greatest latency hit). Remember, even though the synchronous timing is 2.5ns (400MHz effective) for writes or burst reads (after the initial latency), read latency is typically 20-60ns (only 16-50MHz!). So the more you are reading smaller chunks randomly, the more synchronous timing doesn't mean squat. Although this is a mega-oversimplification (and _not_ actual), at DDR 200MHz (400MHz effective aka DDR400/PC3200), the last timing essentiall mean ... 4 = 20ns (50MHz) 5 = 25ns (40MHz) 6 = 30ns (33MHz) 8 = 40ns (25MHz) 10 = 50ns (20MHz) 12 = 60ns (16MHz!) That's what you're waiting on for the first few bytes! Yikes! At some point, if you're reading a lot of random areas of memory, and your L1+L2 (possibly +L3) cache hit rate is closer to 94% than 97% or so, you're going to be tying up the memory bus with a lot of wait. The _true_ dual-channel interleaved nature of S939/940 helps, but it's still significant. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com ---------------------------------------------------------- The existence of Linux has far more to do with the breakup of AT&T's monopoly than anything Microsoft has ever done. From msaqer at iastate.edu Sat Jul 8 20:33:23 2006 From: msaqer at iastate.edu (Mohamad Al-Saqer) Date: Sat, 8 Jul 2006 15:33:23 -0500 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> Message-ID: <200607081533.23291.msaqer@iastate.edu> On Saturday 08 July 2006 1:43, Bryan J. Smith wrote: > That means a maximum of (2) DDR DIMMs for unregistered Socket-939. > That means a maximum of (4) DDR DIMMs for registered Socket-940. most motherboards support registered memory. Does this mean if ECC non-buffered memory sticks is installed on 939 MB we get around these limitations? From b.j.smith at ieee.org Sun Jul 9 01:05:59 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Sat, 08 Jul 2006 21:05:59 -0400 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: <200607081533.23291.msaqer@iastate.edu> References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> <200607081533.23291.msaqer@iastate.edu> Message-ID: <1152407159.2886.9.camel@bert64.oviedo.smithconcepts.com> On Sat, 2006-07-08 at 15:33 -0500, Mohamad Al-Saqer wrote: > most motherboards support registered memory. Remember, the memory controller and channels are on the processor in AMD Athlon 64 / Opteron. AMD _only_ supports _unregistered_ on Socket-939. And AMD _only_ supports _registered_ on Socket-940. Now there are some claims that some Socket-940 processors will, unofficially, work with unregistered DIMMs. But I haven't seen it personally. Going the other way, I don't think registered works with any Socket-939 processor at all. I certainly have not heard anyone doing that. > Does this mean if ECC non-buffered memory sticks Umm, that's _unregistered_. Did you mean registered? [ BTW, ECC does _not_ matter when it comes to signaling ] > is installed on 939 MB we get around these limitations? You can only use unregistered on Socket-939. At DDR400, that's 1 DIMM per channel, so maximum 2 DIMMs. At DDR333, that's 2 DIMMs per channel, so maximum 4 DIMMs. Expect any unstable system if you do otherwise. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com ---------------------------------------------------------- The existence of Linux has far more to do with the breakup of AT&T's monopoly than anything Microsoft has ever done. From hahn at physics.mcmaster.ca Sun Jul 9 05:40:07 2006 From: hahn at physics.mcmaster.ca (hahn at physics.mcmaster.ca) Date: Sun, 9 Jul 2006 01:40:07 -0400 (EDT) Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: <1152407159.2886.9.camel@bert64.oviedo.smithconcepts.com> References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> <200607081533.23291.msaqer@iastate.edu> <1152407159.2886.9.camel@bert64.oviedo.smithconcepts.com> Message-ID: > At DDR400, that's 1 DIMM per channel, so maximum 2 DIMMs. > At DDR333, that's 2 DIMMs per channel, so maximum 4 DIMMs. > > Expect any unstable system if you do otherwise. I believe this ignores another relevant factor: how many chip-selects the dimm uses. my understanding is that if a dimm is "double-sided", it means that it uses 2 CS's. table 46 on p 182 of AMD's bios guide (doc 26094) appears to allow up to 6 ranks for ddr400/2T operation. afaikt, a double-rank dimm is AKA double-sided. unfortunatley AMD also adds some weaselish text about how MB vendors should validate, etc... From b.j.smith at ieee.org Sun Jul 9 05:54:52 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Sun, 09 Jul 2006 01:54:52 -0400 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB -- JEDEC specifications ... In-Reply-To: References: <200607081322.12468.msaqer@iastate.edu> <1152384192.2868.78.camel@bert64.oviedo.smithconcepts.com> <200607081533.23291.msaqer@iastate.edu> <1152407159.2886.9.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1152424492.2886.81.camel@bert64.oviedo.smithconcepts.com> On Sun, 2006-07-09 at 01:40 -0400, hahn at physics.mcmaster.ca wrote: > I believe this ignores another relevant factor: > how many chip-selects the dimm uses. I was purposely not trying to get into memory controller design. Not only have I ripped out the AMD and Intel engineering specification sheets on various chipsets to verify the _exact_ IC technology, width and number/banking before -- but I've designed memory controllers myself. But yes, if you have a total of 128-bit IC width in a 64-bit DIMM, that's also an issue. But that's just the tip-of-the-iceberg. > my understanding is that if a dimm is "double-sided", it means that > it uses 2 CS's. There are a lot of "double sided" or "double stacked" or "low density" or "double rank" or "low rank" or countless other, non-standard, non-uniform descriptions out there. I _refuse_ to use them, as they are not standardized and often confusing. E.g., does "low density" mean less total width/fewer chips? Or does it mean the lower density in the IC, meaning more total width/more chips? The *ONLY* way to *ACCURATELY* describe this is to sit down and break out each DIMM into the *RAW* number of 32-bit "banks" -- and that means you take the bit-width of the *INDIVIDUAL* ICs on every DIMM and multiply by number (except for parity/ECC chips). So, if you want to get really details, we *CAN* do that! ;-> Furthermore, some chipsets just do _not_ support certain IC technologies, bit-widths and other sizing. The chronic issue was the PC100 256MB DIMM for i440BX versus i810/815 -- they used completely _difference_ DIMMs. The i440BX only supported a 4-bit with in 32-chip (36-chip for ECC) configuration -- clearly a "double sided" registered DIMM. The i810/815 used a 8-bit or 16-bit in a 8 or 4 chip configuration, respectively. And even that is a mega-oversimplification! ;-> > table 46 on p 182 of AMD's bios guide (doc 26094) appears to allow up > to 6 ranks for ddr400/2T operation. afaikt, a double-rank dimm is AKA > double-sided. unfortunatley AMD also adds some weaselish text about > how MB vendors should validate, etc... I only listed the JEDEC specifications and recommendations. If you want to go outside of those, which some vendors may do, by all means, do! Also remember that AMD _might_ be talking about "banks" and not merely "DIMMs." "Banks" are 32-bit (like old SIMMs). "DIMMs" are 64-bit. BTW, I've seen the term "rank" used 3 different ways. Everyone seems to want to simply things with a term or two. Impossible. Especially when it comes to DDR333 v. DDR400 performance. Even reporting 3 or 4 timings is woefully incomplete. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com ---------------------------------------------------------- The existence of Linux has far more to do with the breakup of AT&T's monopoly than anything Microsoft has ever done. From andy at boersma.ca Mon Jul 10 16:47:53 2006 From: andy at boersma.ca (andy at boersma.ca) Date: Mon, 10 Jul 2006 16:47:53 +0000 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB Message-ID: <20060710164753.jbqfas1cgcosw88g@webmail.internic.ca> I'm using a Biostar T6100-939 overclocking no problems with 4 gb PC3200 Andy B From msaqer at iastate.edu Mon Jul 10 19:27:58 2006 From: msaqer at iastate.edu (Mohamad Al-Saqer) Date: Mon, 10 Jul 2006 14:27:58 -0500 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB In-Reply-To: <20060710164753.jbqfas1cgcosw88g@webmail.internic.ca> References: <20060710164753.jbqfas1cgcosw88g@webmail.internic.ca> Message-ID: <200607101427.58892.msaqer@iastate.edu> On Monday 10 July 2006 11:47, andy at boersma.ca wrote: > I'm using a Biostar T6100-939 overclocking no problems with 4 gb PC3200 > > Andy B Can you be more specific? What do you mean overclocking? is it that the RAM running at 400MHz? Maybe you need to benchmark for stability like running memtest86, for instance, for long enough (several hours) to check for errors. According to the replies in this thread, it looks like it is a limitation of the CPU not the motherboard. Searching the net, I came across few posts saying that CPUs revision E or later should work fine with 4 sticks at 400MHz. So what is your CPU model number and specifications? From andy at boersma.ca Tue Jul 11 16:49:44 2006 From: andy at boersma.ca (andy at boersma.ca) Date: Tue, 11 Jul 2006 16:49:44 +0000 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB Message-ID: <20060711164944.5qtwekzeqswsk4o8@webmail.internic.ca> I think the CPU is Rev E. I am using Kingston memory 4 1Gb sticks. The Biostar T6100 allows overclocking of CPU and Memory separately. Voltages can be increased to CPU and Memory seperately. Plus it has a nice auto overclock utility. In windows :-(. ut it gave me a good guide on where to start safely. Have not had it crash with Fedora 5 64 or Ubuntu 6 64bit. Andy B. From msaqer at iastate.edu Tue Jul 11 17:48:58 2006 From: msaqer at iastate.edu (Mohamad Al-Saqer) Date: Tue, 11 Jul 2006 12:48:58 -0500 Subject: 4gb (4 memory sticks) at 400Mhz on socket 939 MB In-Reply-To: <20060711164944.5qtwekzeqswsk4o8@webmail.internic.ca> References: <20060711164944.5qtwekzeqswsk4o8@webmail.internic.ca> Message-ID: <200607111248.58436.msaqer@iastate.edu> On Tuesday 11 July 2006 11:49, andy at boersma.ca wrote: > I am using Kingston memory 4 1Gb sticks. Running at 400MHz? > Have not had it crash with Fedora 5 64 ?or Ubuntu 6 64bit. For how long at high load? good if you can try to do memtest86 for several hours with the 4 sticks at 400Mhz (maybe overnight) and see at the end if memtest would report errors. > Andy B. Mohamad From wam at HiWAAY.net Tue Jul 11 19:27:25 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue, 11 Jul 2006 14:27:25 -0500 Subject: AMD64 chipset Linux support .... Message-ID: <44B3FB9D.9000108@HiWAAY.net> .... I am looking at the Asus A8V Mbd (Via K8T800 chipset, ~$62.00 on pricewatch) for a hypothetical Opteron box, lightweight LAN server. How is Linux support for this chipset ? I have an older (S370 PIII) box, this one as it happens, w/ a Via 694 chipset & love it. Can I look forward to the same great performance & reliability under Linux (FC5 most likely) w/ the K8T800 ? TIA .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton From b.j.smith at ieee.org Wed Jul 12 04:24:36 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 00:24:36 -0400 Subject: AMD64 chipset Linux support .... In-Reply-To: <44B3FB9D.9000108@HiWAAY.net> References: <44B3FB9D.9000108@HiWAAY.net> Message-ID: <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-07-11 at 14:27 -0500, William A. Mahaffey III wrote: > .... I am looking at the Asus A8V Mbd (Via K8T800 chipset, > ~$62.00 on pricewatch) for a hypothetical Opteron box, lightweight > LAN server. You don't want to buy a desktop chipset for a server. People do this all-the-time, even for Intel (e.g., i845/i865), and pay the price in I/O limitations and stability. Paying $200 for an entry-level, but true _server_ mainboard instead of $100 for a desktop mainboard is well worth the added $100. The entry-level ServerWorks HT1000 is a nice solution with a single PCI-X channel for Socket-939 for under $200. Adding a HT2000 typically brings up the cost over $300 to what the nVidia nForce Pro with 40 PCIe channels plus the AMD8131/8132 dual-PCI-X typically costs. Intel's entry-level 72xx series (based on a license of the ServerWorks IV/Grand Champion) starts with dual PCIe x8 I/O in the 7230 for under $200 (sometimes under $150) and you can add an ESB6x00 series for PCI-X for under $300. Again, _never_ use a desktop chipset for server duties. And especially _not_ a "consumer" rated/tested (e.g., i865 instead of i875). > How is Linux support for this chipset ? ViA typically has ATA controller issues (long story). Off-chipset is recommended in many cases. > I have an older (S370 PIII) box, this one as it happens, w/ > a Via 694 chipset & love it. The 694 had many performance and stability issues. Especially early revisions. Latter 694X versions fared better, but were still _not_ a _server_ chipset. E.g., the 694 still had only 1/4-1/8 the I/O of a ServerSet IIILE which started at $250 when they were new (and under $200 later on). I replaced many i440BX/GX and 693A/694X dual-processor systems with the IIILE and file server performance was instantly 3x (literally, +200%). > Can I look forward to the same great performance & reliability under > Linux (FC5 most likely) w/ the K8T800 ? TIA .... I guess if you're idea of performance and reliability is low, yes. ;-> -- Bryan P.S. William -- you're not going to believe this, but I wrote this and was just about to send and just realized you posted it. I honestly wrote it _before_ I noticed it was you. P.P.S. Lay off the desktop chipsets -- seriously. I'll loan you an Intel 72xx (or even a 75xx) series -- you'll see what I mean! ;-> -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From wam at HiWAAY.net Wed Jul 12 04:40:23 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue, 11 Jul 2006 23:40:23 -0500 Subject: AMD64 chipset Linux support .... In-Reply-To: <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> Message-ID: <44B47D37.7020001@HiWAAY.net> Bryan J. Smith wrote: >On Tue, 2006-07-11 at 14:27 -0500, William A. Mahaffey III wrote: > > >>.... I am looking at the Asus A8V Mbd (Via K8T800 chipset, >>~$62.00 on pricewatch) for a hypothetical Opteron box, lightweight >>LAN server. >> >> > >You don't want to buy a desktop chipset for a server. People do this >all-the-time, even for Intel (e.g., i845/i865), and pay the price in I/O >limitations and stability. > >Paying $200 for an entry-level, but true _server_ mainboard instead of >$100 for a desktop mainboard is well worth the added $100. > >The entry-level ServerWorks HT1000 is a nice solution with a single >PCI-X channel for Socket-939 for under $200. Adding a HT2000 typically >brings up the cost over $300 to what the nVidia nForce Pro with 40 PCIe >channels plus the AMD8131/8132 dual-PCI-X typically costs. > >Intel's entry-level 72xx series (based on a license of the ServerWorks >IV/Grand Champion) starts with dual PCIe x8 I/O in the 7230 for under >$200 (sometimes under $150) and you can add an ESB6x00 series for PCI-X >for under $300. > >Again, _never_ use a desktop chipset for server duties. And especially >_not_ a "consumer" rated/tested (e.g., i865 instead of i875). > > > >>How is Linux support for this chipset ? >> >> > >ViA typically has ATA controller issues (long story). >Off-chipset is recommended in many cases. > > > >>I have an older (S370 PIII) box, this one as it happens, w/ >>a Via 694 chipset & love it. >> >> > >The 694 had many performance and stability issues. Especially early >revisions. Latter 694X versions fared better, but were still _not_ a >_server_ chipset. > >E.g., the 694 still had only 1/4-1/8 the I/O of a ServerSet IIILE which >started at $250 when they were new (and under $200 later on). I >replaced many i440BX/GX and 693A/694X dual-processor systems with the >IIILE and file server performance was instantly 3x (literally, +200%). > > > >>Can I look forward to the same great performance & reliability under >>Linux (FC5 most likely) w/ the K8T800 ? TIA .... >> >> > >I guess if you're idea of performance and reliability is low, yes. ;-> > >-- Bryan > >P.S. William -- you're not going to believe this, but I wrote this and >was just about to send and just realized you posted it. I honestly >wrote it _before_ I noticed it was you. > >P.P.S. Lay off the desktop chipsets -- seriously. I'll loan you an >Intel 72xx (or even a 75xx) series -- you'll see what I mean! ;-> > > > Oooohhhhhhh, you're just too kind :-). When I say lightweight, I *mean* lightweight. NO public access, NO apache, NO DNS, NO several dozen users at any 1 time, just CPU, RAM, & code to run ad nauseum. It *will* live at runlevel 3, but the server traits end there, for the most part. I was mostly asking about reasonably complete/stable Linux support for that chipset, decent %-age of full speed for RAM-intensive calculations, etc. e.g. good OS/chipset interaction for my somewhat truncated requirements. Someone else (another list) indicated all was well, care to dispute that ? -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: From b.j.smith at ieee.org Wed Jul 12 04:43:27 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 00:43:27 -0400 Subject: AMD64 chipset Linux support .... In-Reply-To: <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> On Wed, 2006-07-12 at 00:24 -0400, Bryan J. Smith wrote: > Paying $200 for an entry-level, but true _server_ mainboard instead of > $100 for a desktop mainboard is well worth the added $100. > ... > P.S. William -- you're not going to believe this, but I wrote this and > was just about to send and just realized you posted it. I honestly > wrote it _before_ I noticed it was you. > P.P.S. Lay off the desktop chipsets -- seriously. I'll loan you an > Intel 72xx (or even a 75xx) series -- you'll see what I mean! ;-> FYI, NewEgg _does_ have a server mainboard category: http://www.newegg.com/ProductSort/SubCategory.asp?SubCategory=302 An Intel E7210 with two (2) PCI-X 66MHz slots starts at $180: http://www.newegg.com/Product/Product.asp?Item=N82E16813128261 A ServerWorks HT1000 with one (1) PCI-X 133MHz slot starts at $235: http://www.newegg.com/Product/Product.asp?Item=N82E16813182078 There are also various options with PCIe x4 and x8 slots (with or without PCI-X) for storage (in case your storage is a PCIe x4 or x8 card) for both single Socket-478/LGA-775 as well as Socket-939. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From wam at HiWAAY.net Wed Jul 12 04:54:21 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue, 11 Jul 2006 23:54:21 -0500 Subject: AMD64 chipset Linux support .... In-Reply-To: <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> Message-ID: <44B4807D.6070604@HiWAAY.net> Bryan J. Smith wrote: >On Wed, 2006-07-12 at 00:24 -0400, Bryan J. Smith wrote: > > >>Paying $200 for an entry-level, but true _server_ mainboard instead of >>$100 for a desktop mainboard is well worth the added $100. >> ... >>P.S. William -- you're not going to believe this, but I wrote this and >>was just about to send and just realized you posted it. I honestly >>wrote it _before_ I noticed it was you. >>P.P.S. Lay off the desktop chipsets -- seriously. I'll loan you an >>Intel 72xx (or even a 75xx) series -- you'll see what I mean! ;-> >> >> > >FYI, NewEgg _does_ have a server mainboard category: > http://www.newegg.com/ProductSort/SubCategory.asp?SubCategory=302 > >An Intel E7210 with two (2) PCI-X 66MHz slots starts at $180: > http://www.newegg.com/Product/Product.asp?Item=N82E16813128261 > >A ServerWorks HT1000 with one (1) PCI-X 133MHz slot starts at $235: > http://www.newegg.com/Product/Product.asp?Item=N82E16813182078 > >There are also various options with PCIe x4 and x8 slots (with or >without PCI-X) for storage (in case your storage is a PCIe x4 or x8 >card) for both single Socket-478/LGA-775 as well as Socket-939. > > > > 1st link got me the generic server board page, nothing specific, my misfire ? I *am* looking at Opteron, & found the K8T800 board on a price basis alone. Tyan makes a S939 board, NewEgg got 'em for ~$195.00 last I looked. I *don't* need the I/O just now, just good RAM/CPU/reliability .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: From wam at HiWAAY.net Wed Jul 12 04:57:29 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Tue, 11 Jul 2006 23:57:29 -0500 Subject: AMD64 chipset Linux support .... In-Reply-To: <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> Message-ID: <44B48139.6000708@HiWAAY.net> Bryan J. Smith wrote: >On Wed, 2006-07-12 at 00:24 -0400, Bryan J. Smith wrote: > > >>Paying $200 for an entry-level, but true _server_ mainboard instead of >>$100 for a desktop mainboard is well worth the added $100. >> ... >>P.S. William -- you're not going to believe this, but I wrote this and >>was just about to send and just realized you posted it. I honestly >>wrote it _before_ I noticed it was you. >>P.P.S. Lay off the desktop chipsets -- seriously. I'll loan you an >>Intel 72xx (or even a 75xx) series -- you'll see what I mean! ;-> >> >> > >FYI, NewEgg _does_ have a server mainboard category: > http://www.newegg.com/ProductSort/SubCategory.asp?SubCategory=302 > >An Intel E7210 with two (2) PCI-X 66MHz slots starts at $180: > http://www.newegg.com/Product/Product.asp?Item=N82E16813128261 > >A ServerWorks HT1000 with one (1) PCI-X 133MHz slot starts at $235: > http://www.newegg.com/Product/Product.asp?Item=N82E16813182078 > >There are also various options with PCIe x4 and x8 slots (with or >without PCI-X) for storage (in case your storage is a PCIe x4 or x8 >card) for both single Socket-478/LGA-775 as well as Socket-939. > > > > Just tried that last link, sweet. NewEgg had the Foxconn S940 board for $195.00 before they (Foxconn) tanked that board, too bad I waited .... -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: From b.j.smith at ieee.org Wed Jul 12 05:00:03 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 01:00:03 -0400 Subject: AMD64 chipset Linux support .... In-Reply-To: <44B47D37.7020001@HiWAAY.net> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <44B47D37.7020001@HiWAAY.net> Message-ID: <1152680403.2880.42.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-07-11 at 23:40 -0500, William A. Mahaffey III wrote: > Oooohhhhhhh, you're just too kind :-). Seriously William. We had several threads on your issues with the i865 (as well as i845) month after month after month. That wasn't even a "professional/corporate desktop" chipsets, that's the i875. I'm just trying to help you friend. > When I say lightweight, I *mean* lightweight. NO public access, What does "public" have _anything_ to do with it? In fact, LAN I/O is typically far more intense. > NO apache, NO DNS, Apache and DNS typically have more to do with _computational_ performance. If you're throwing files around on a LAN, that's I/O! We've had this discussion before, when you've had compatibility and performance issues. Moving from an Intel i845/i865 (again, not even a i875 that is tested for professional, desktop use -- and totally ignoring the E72xx/75xx) to a ViA KT series is a _lateral_ move. Again, referring back to your experience with the 694 -- back in the late '90s, I used to replace i440BX/GX and 693/694 chipsets with ServerSet IIIs -- often for ~$250. A measly $100 more in the mainboard makes _all_ the difference when you're spending $1K! > NO several dozen users at any 1 time, just CPU, RAM, & code to run ad > nauseum. What is its function? Again, why save $100 on a system that costs $500-1,000 anyway, when another $100 will give you server performance and quality? > It *will* live at runlevel 3, When does "graphics" have _anything_ to do with "storage/communcation" I/O? William, I think this is where you keep missing the point, and why you keep running into issue after issue on your systems. I can't believe you've now taken all those issues and discussions we had months ago with your issues and just thrown them out-the-window. Spend $175+ on an E72xx series mainboard for P4, or $235 for a HT1000 for Opteron. > but the server traits end there, for the most part. I was mostly > asking about reasonably complete/stable Linux support for that > chipset, decent %-age of full speed for RAM-intensive calculations, Can you elaborate on the application? > etc. e.g. good OS/chipset interaction for my somewhat truncated > requirements. Intel 72xx series hsa pretty much been the "Tier-1 PC OEM Gold Standard" for P4 (as well as Xeon) for workstation/entry-level server for 3+ years (and that's not looking at the new 5000 series which just came out a few months ago). On AMD, it's been either the aging AMD8000 series, or the 18 month proven nForce Professional with the AMD8131/8132 added for PCI-X. On the low-end (single Socket-939), the ServerWorks HT1000 does the job for kernel 2.6 just fine. > Someone else (another list) Yes, LUNA. > indicated all was well, care to dispute that ? Yes, I do! And I responded to him as well! It's clear he's used to desktop uses as well. Desktop I/O is _different_ than server I/O, even just for a few clients. William, you've had a _lot_ of issues that I've helped you with. I didn't mind, especially since you said didn't know about the E72xx/75xx series prior. And we discussed this at length. But now you're going right back into the rut. And you're going to run into the same issues. I honestly thought you really appreciated it when we went into the i8xx/9xx v. E72xx/75xx series differences months back. I guess not! @-ppp This isn't a "my way dammit" discussion. This is me saying, "Hey, remember all those issues you had? I thought you weren't going to dot hem again?" Spend an extra $100. It's damn worth it. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From b.j.smith at ieee.org Wed Jul 12 05:06:38 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 01:06:38 -0400 Subject: AMD64 chipset Linux support .... In-Reply-To: <44B4807D.6070604@HiWAAY.net> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <1152679407.2880.24.camel@bert64.oviedo.smithconcepts.com> <44B4807D.6070604@HiWAAY.net> Message-ID: <1152680798.2880.47.camel@bert64.oviedo.smithconcepts.com> On Tue, 2006-07-11 at 23:54 -0500, William A. Mahaffey III wrote: > 1st link got me the generic server board page, nothing specific, my > misfire ? It's a _different_ page than the normal mainboard page. > I *am* looking at Opteron, Just because it's an Opteron does _not_ mean any mainboard that takes it is server-quality! We've had that discussion before too! You'd be better off going with an older P4 with a cheap E7210 mainboard (although they are becoming more scarce as the E7221/7230 are more popular now). > & found the K8T800 board on a price basis alone. Yes! I know! > Tyan makes a S939 board, Tyan makes server, workstation _and_ desktop mainboards too. Socket-939 is a desktop and workstation/entry-level server platform. > NewEgg got 'em for ~$195.00 last I looked. What chipset is on that board? > I *don't* need the I/O just now, just good RAM/CPU/reliability .... And what's your application? Again, an extra $100 _before_ you put the $1K+ system makes all-the-difference. It's not just server I/O. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From wam at HiWAAY.net Wed Jul 12 05:23:53 2006 From: wam at HiWAAY.net (William A. Mahaffey III) Date: Wed, 12 Jul 2006 00:23:53 -0500 Subject: AMD64 chipset Linux support .... In-Reply-To: <1152680403.2880.42.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <44B47D37.7020001@HiWAAY.net> <1152680403.2880.42.camel@bert64.oviedo.smithconcepts.com> Message-ID: <44B48769.9050403@HiWAAY.net> Bryan J. Smith wrote: >On Tue, 2006-07-11 at 23:40 -0500, William A. Mahaffey III wrote: > > >>Oooohhhhhhh, you're just too kind :-). >> >> > >Seriously William. We had several threads on your issues with the i865 >(as well as i845) month after month after month. That wasn't even a >"professional/corporate desktop" chipsets, that's the i875. > >I'm just trying to help you friend. > > > >>When I say lightweight, I *mean* lightweight. NO public access, >> >> > >What does "public" have _anything_ to do with it? >In fact, LAN I/O is typically far more intense. > > > >>NO apache, NO DNS, >> >> > >Apache and DNS typically have more to do with _computational_ >performance. If you're throwing files around on a LAN, that's I/O! > >We've had this discussion before, when you've had compatibility and >performance issues. Moving from an Intel i845/i865 (again, not even a >i875 that is tested for professional, desktop use -- and totally >ignoring the E72xx/75xx) to a ViA KT series is a _lateral_ move. > >Again, referring back to your experience with the 694 -- back in the >late '90s, I used to replace i440BX/GX and 693/694 chipsets with >ServerSet IIIs -- often for ~$250. A measly $100 more in the mainboard >makes _all_ the difference when you're spending $1K! > > > >>NO several dozen users at any 1 time, just CPU, RAM, & code to run ad >>nauseum. >> >> > >What is its function? > > See below. >Again, why save $100 on a system that costs $500-1,000 anyway, when >another $100 will give you server performance and quality? > > Probably more like $700.00 total, apparently. $100.00+ does matter, ~15% .... > > >>It *will* live at runlevel 3, >> >> > >When does "graphics" have _anything_ to do with "storage/communcation" >I/O? William, I think this is where you keep missing the point, and why >you keep running into issue after issue on your systems. > >I can't believe you've now taken all those issues and discussions we had >months ago with your issues and just thrown them out-the-window. Spend >$175+ on an E72xx series mainboard for P4, or $235 for a HT1000 for >Opteron. > > > >>but the server traits end there, for the most part. I was mostly >>asking about reasonably complete/stable Linux support for that >>chipset, decent %-age of full speed for RAM-intensive calculations, >> >> > >Can you elaborate on the application? > > Sure. Finite Element grid/mesh generation & subsequent analysis, often coupled w/ CFD grid generation/analysis. Fluid-structure interaction problems (deforming SRM propellant grains under stress from the flow resulting from their combustion). Reasonably well resolved 2-D (i.e., not 3-D, thus I *can* afford pretty good 2-D resolution, borderline large-eddy type resolution on the fluid side, similar resolution on the solid side since I spend most of my CPU time on the fluid side & don't want/need to answer needling questions about skimping on resolution *anywhere* in the analysis). Turbulent, sub-sonic/transonic, complex internal geometries on the fluid side, largish deformations on the solid side. Definitely CPU/RAM intensive, not quite so I/O intensive. Local drives large enough to catch the results, then processed by another box on the LAN (where the Tecplot license lives), but still by me, i.e. only one user at any 1 time. These boxen are like droids, w/ only 1 master, not dozens like most public servers. I/O is less important, CPU/RAM/stability are paramount. > > >>etc. e.g. good OS/chipset interaction for my somewhat truncated >>requirements. >> >> -- William A. Mahaffey III ---------------------------------------------------------------------- "The M1 Garand is without doubt the finest implement of war ever devised by man." -- Gen. George S. Patton -------------- next part -------------- An HTML attachment was scrubbed... URL: From b.j.smith at ieee.org Wed Jul 12 05:58:04 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 01:58:04 -0400 Subject: AMD64 chipset Linux support .... In-Reply-To: <44B48769.9050403@HiWAAY.net> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <44B47D37.7020001@HiWAAY.net> <1152680403.2880.42.camel@bert64.oviedo.smithconcepts.com> <44B48769.9050403@HiWAAY.net> Message-ID: <1152683884.2880.67.camel@bert64.oviedo.smithconcepts.com> On Wed, 2006-07-12 at 00:23 -0500, William A. Mahaffey III wrote: > Probably more like $700.00 total, apparently. $100.00+ does matter, > ~15% .... Consider your time swapping out a board/chipset that was specifically designated as "does not test to 24x7 tolerances" (e.g., i865 v. i875) in that cost estimate. > Sure. Finite Element grid/mesh generation & subsequent analysis, often > coupled w/ CFD grid generation/analysis. > Fluid-structure interaction problems (deforming SRM propellant grains > under stress from the flow resulting from their combustion). > Reasonably well resolved 2-D (i.e., not 3-D, thus I *can* afford > pretty good 2-D resolution, borderline large-eddy type resolution on > the fluid side, similar resolution on the solid side since I spend > most of my CPU time on the fluid side & don't want/need to answer > needling questions about skimping on resolution *anywhere* in the > analysis). Turbulent, sub-sonic/transonic, complex internal geometries > on the fluid side, largish deformations on the solid side. Definitely > CPU/RAM intensive, not quite so I/O intensive. Local drives large > enough to catch the results, then processed by another box on the LAN > (where the Tecplot license lives), but still by me, i.e. only one user > at any 1 time. > These boxen are like droids, w/ only 1 master, not dozens like most > public servers. I/O is less important, But what's your data rate from the distributed clients to the master(s)? At what point will the combination of possible local storage and network communication saturate each other? That all affects how linearly the application scales. If you're pushing back just 10MBps to the master per client with at least 10 clients, you've already saturated a 32-bit PCI bus for just network -- not including the mass inefficiency of Ethernet or the added throughout to put to disk (if required). That's why even 8 years ago**, we used 64-bit PCI NICs for GbE, typically with all other I/O (storage, etc...) on a separate PCI bus when we were looking at anything close to a dozen nodes**. > CPU/RAM/stability are paramount. They why do you buy i865 solutions when they _fail_ Intel's tolerances for i875? We've been through this before William. I really don't think you understood anything I said before, and I really wish the best for you. Your time easily pays for the premium. Saving $100/node isn't worth your time -- before we even look at the performance/bottleneck aspects. -- Bryan **NOTE: I (among others on this list) rolled out such grid computing for CFD and other, distributed applications with Linux in the late '90s. Even back then, a single 32-bit PCI bus didn't cut it. The options are a little better now with PCIe x1 channels being more standard -- but the data rate has always increased as well. Some are using Infiniband right on the HTX (HyperTransport eXtension), but that's overkill for you. All I'm advocating is that you don't bottleneck yourself to the point where you are sacrificing any scalability beyond 4-5 nodes so you can buy a few more. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From b.j.smith at ieee.org Wed Jul 12 06:14:27 2006 From: b.j.smith at ieee.org (Bryan J. Smith) Date: Wed, 12 Jul 2006 02:14:27 -0400 Subject: AMD64 chipset Linux support .... Elementary costs in CFD and other, distributed apps ... In-Reply-To: <1152683884.2880.67.camel@bert64.oviedo.smithconcepts.com> References: <44B3FB9D.9000108@HiWAAY.net> <1152678276.2880.15.camel@bert64.oviedo.smithconcepts.com> <44B47D37.7020001@HiWAAY.net> <1152680403.2880.42.camel@bert64.oviedo.smithconcepts.com> <44B48769.9050403@HiWAAY.net> <1152683884.2880.67.camel@bert64.oviedo.smithconcepts.com> Message-ID: <1152684867.2880.77.camel@bert64.oviedo.smithconcepts.com> On Wed, 2006-07-12 at 01:58 -0400, Bryan J. Smith wrote: > Consider your time swapping out a board/chipset that was specifically > designated as "does not test to 24x7 tolerances" (e.g., i865 v. i875) in > that cost estimate. Or the cost of your engineer's time and/or deadlines for a failed run. > **NOTE: I (among others on this list) rolled out such grid computing > for CFD and other, distributed applications with Linux in the late '90s. > Even back then, a single 32-bit PCI bus didn't cut it. The options are > a little better now with PCIe x1 channels being more standard -- but the > data rate has always increased as well. If cost is really paramount, then get the cheapest _refurb_ mainboard and Celeron or Sempron you can find, run diskless (if possible -- it does make management easier) keep the cost under $200/unit. The only issue is if and when the I/O becomes a bottleneck for the DTR -- especially when you have masters in the distributed system (which is why you should _never_ skimp there). You also need to ensure your applications -- especially if engineering runs take days at a time -- can handle failover and restart an interation/process if a node fails. That right there is typically the biggest issue and cost (far, far more than hardware and IT time) -- code that is non-interruptable/restartable when runs take days. -- Bryan J. Smith Professional, technical annoyance mailto:b.j.smith at ieee.org http://thebs413.blogspot.com --------------------------------------------------------- The world is in need of solutions. Unfortunately, people seem to be more interested in blindly aligning themselves with one of only two viewponts -- an "us v. them" debate that has nothing to do with finding an actual solution. From jdreese at bucknell.edu Thu Jul 13 14:29:08 2006 From: jdreese at bucknell.edu (Jeremy Dreese) Date: Thu, 13 Jul 2006 10:29:08 -0400 Subject: 32-bit Compiling and Linking Message-ID: <44B658B4.4090103@bucknell.edu> We have user home directories that are shared across 32-bit and 64-bit systems. We have setup a wrapper script around gcc/g++ so that our 64-bit systems compile in 32-bit mode by default; basically, the "-m32" option is just inserted into a user's gcc/g++ command. This works fine if the user is not linking against a library (e.g. "gcc -o hello hello.c"). However, if they need to link against a library, they get an error like this: [user at host dir]$ gcc -o sslnumbers -lssl sslnumbers.c /usr/bin/ld: skipping incompatible /usr/lib64/libssl.so when searching for -lssl/usr/bin/ld: skipping incompatible /usr/lib64/libssl.a when searching for -lssl /usr/bin/ld: cannot find -lssl collect2: ld returned 1 exit status Obviously, it cannot find the 32-bit version of the "ssl" library. This occurs because their in fact is no "/usr/lib/libssl.so", although "/usr/lib/libssl3.so" *does* exist. Further, as far as I know there is no package that includes the necessary "/usr/lib/libssl.so" link. Is there a way to have the linker find the necessary 32-bit library (other than manually creating a link)? Thanks, Jeremy Dreese