Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-ASUS&...
Any suggestions will be appreciated....
Todd
I'd suggest going for an Athlon 64... lol But maybe let's not start up this discussion...
On Thu, 15 Dec 2005, Todd Cary wrote:
Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-ASUS&...
Any suggestions will be appreciated....
Todd
Maciej Żenczykowski wrote:
I'd suggest going for an Athlon 64... lol
Well, the OP probably is stuck with a processor capable of frying eggs...which is not possible with your suggestion. My colleagues take their sweaters/warmers off due to the Dell cum heater box besides their feet in the office.
an Intel chipset motherboard seems to be the safest bet. Say no to VIA. (warning: biased opinion from a Chinese guy who has been burnt too many times by chipsets from said Taiwanese company in both consumer and server boards and therefore has not tried any of the latest chipsets from said company)
OK...now I go to the Intel section, and I am presented with these:
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-INTEL&...
Are there some salient factors I should be looking at?
Todd
Feizhou wrote:
Maciej Żenczykowski wrote:
I'd suggest going for an Athlon 64... lol
Well, the OP probably is stuck with a processor capable of frying eggs...which is not possible with your suggestion. My colleagues take their sweaters/warmers off due to the Dell cum heater box besides their feet in the office.
an Intel chipset motherboard seems to be the safest bet. Say no to VIA. (warning: biased opinion from a Chinese guy who has been burnt too many times by chipsets from said Taiwanese company in both consumer and server boards and therefore has not tried any of the latest chipsets from said company) _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Todd Cary wrote:
OK...now I go to the Intel section, and I am presented with these:
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-INTEL&...
Are there some salient factors I should be looking at?
Depends on what you are going to use the box for. What other stuff are you going to put on this board? Like how much RAM and its type, disk(s) and their type...
Todd Cary wrote:
Are there some salient factors I should be looking at?
Feizhou feizhou@graffiti.net wrote:
Depends on what you are going to use the box for.
Yes, is this just going to be a desktop?
Feizhou feizhou@graffiti.net wrote:
Well, the OP probably is stuck with a processor capable of frying eggs...which is not possible with your suggestion. My colleagues take their sweaters/warmers off due to the
Dell
cum heater box besides their feet in the office.
Socket-478/LGA-775 Netburst architecture (Pentium 4) is the absolute worst in heat generation, even at 90nm. They've brought it down some with the dual-core solutions, but it's still way too high.
Newer Socket-754/939/940 Athlon 3000-3500+ and Opteron "HE" (or 150+) only generate 31-55W heat. Dual-core versions are 70-110W.
The best is the newer, but little known Socket-479 Pentium Pro-III architecture (Pentium M), and uses as little as 21W. The 2.0-2.26GHz versions will typically best all but the highest clock Pentium 4. They even offer it with the PCIe/DDR2 i915 chipset, although Socket-479 is a major mark-up (but far better than it was just a little bit ago).
an Intel chipset motherboard seems to be the safest bet.
Not always. But for the most part, the new ICH7 peripherals on the i9x5 are fairly well supported now.
Ironically enough, Intel does _not_ make good server chipsets, they _never_ have. The only Intel chipsets for servers that are worthy are the E7200/7500 series -- designed by ServerWorks (now owned by Broadcom). The last server chipset Intel designed was the NX450 -- well over 5 years ago.
The ServerWorks ServerSet III series was a godsend back in the P3/Xeon days. The GrandChampion (GC) series for P4/Xeon was also powerful until Intel came out with the E7500 series based on it, and then the more entry-level E7200 series after that.
The absolute best server "chips" (since AMD HyperTransport is no longer a "fixed" chipset design), are the AMD8000 series -- especially the dual-channel AMD8131/8132 PCI-X 1.0/2.0 HyperTransport tunnels. The AMD8131/8132 paired with newer logic like the nForce Pro series for PCIe and peripherals, it's a very powerful workstation and/or server combination for Opteron.
Say no to VIA. (warning: biased opinion from a Chinese guy who has been burnt too many times by chipsets from said Taiwanese
company
in both consumer and server boards and therefore has not
tried
any of the latest chipsets from said company)
ViA is great for ViA C3/Eden platforms. They typically lag in features, so by the time the leading-edge desktop ViA chipsets get peripheral support in Linux, they are adopted by the low-power C3/Eden platforms.
So yes, for desktop, ViA changes their peripheral logic way too much. That keeps the kernel developers adding PCI IDs, tracking little variants in their ATA and other logic, etc... ViA has _not_ switched to native HyperTransport on AMD, and are still using their VLink PCI-based interconnect.
But ViA has _never_ designed a server chipset either.
I was very impressed with nVidia's ability to keep PCI IDs and other peripherals consistent from the nForce2/MCP-02 through the single-chip nForce4 (integrated MCP-04). Unfortunately, that seems to have ended with the new nForce4x0/GeForce61x0 (C51/NV44), it uses new PCI IDs and other things so you need a recent kernel.
E.g., FC4's installer 2.6.11 didn't cut it -- the updated 2.6.14 did, however. Although I have to had it to nVidia, they at least give you an installable driver set that you can do on a minimal install. I.e., I installed FC4 with kernel 2.6.11, installed the nForce platform driver with its "nvnet" for 10/100[/1000] NIC, ran yum update, then switched back to the GPL "forcedeth" after the reboot into kernel 2.6.14 (did not have to re-install the nForce platform drivers).
Bryan J. Smith wrote:
Feizhou feizhou@graffiti.net wrote:
Well, the OP probably is stuck with a processor capable of frying eggs...which is not possible with your suggestion. My colleagues take their sweaters/warmers off due to the
Dell
cum heater box besides their feet in the office.
Socket-478/LGA-775 Netburst architecture (Pentium 4) is the absolute worst in heat generation, even at 90nm. They've brought it down some with the dual-core solutions, but it's still way too high.
Heh, that is exactly what they have.
Newer Socket-754/939/940 Athlon 3000-3500+ and Opteron "HE" (or 150+) only generate 31-55W heat. Dual-core versions are 70-110W.
Still below the 120W+ of the P4.
The best is the newer, but little known Socket-479 Pentium Pro-III architecture (Pentium M), and uses as little as 21W. The 2.0-2.26GHz versions will typically best all but the highest clock Pentium 4. They even offer it with the PCIe/DDR2 i915 chipset, although Socket-479 is a major mark-up (but far better than it was just a little bit ago).
Yeah, read about it quite a while ago when it knocked out the P4 2.8 processor.
an Intel chipset motherboard seems to be the safest bet.
Not always. But for the most part, the new ICH7 peripherals on the i9x5 are fairly well supported now.
Thanks to their providing docs.
Ironically enough, Intel does _not_ make good server chipsets, they _never_ have. The only Intel chipsets for servers that are worthy are the E7200/7500 series -- designed by ServerWorks (now owned by Broadcom). The last server chipset Intel designed was the NX450 -- well over 5 years ago.
Yes, these are the chums in use in the newer boxes I used to admin. I loved the 3ware + riser card fiasco though.
ViA is great for ViA C3/Eden platforms. They typically lag in features, so by the time the leading-edge desktop ViA chipsets get peripheral support in Linux, they are adopted by the low-power C3/Eden platforms.
So yes, for desktop, ViA changes their peripheral logic way too much. That keeps the kernel developers adding PCI IDs, tracking little variants in their ATA and other logic, etc... ViA has _not_ switched to native HyperTransport on AMD, and are still using their VLink PCI-based interconnect.
The problems I have are related to their hardware, not whether there are good drivers or not. The poor latencies just won't let me use a Pinnacle DC10 board without crashing.
But ViA has _never_ designed a server chipset either.
Yeah, whatever. Tyan come out with a board for servers based on a VIA chipset for Pentium III cpus and so I got to deal with them.
I was very impressed with nVidia's ability to keep PCI IDs and other peripherals consistent from the nForce2/MCP-02 through the single-chip nForce4 (integrated MCP-04). Unfortunately, that seems to have ended with the new nForce4x0/GeForce61x0 (C51/NV44), it uses new PCI IDs and other things so you need a recent kernel.
E.g., FC4's installer 2.6.11 didn't cut it -- the updated 2.6.14 did, however. Although I have to had it to nVidia, they at least give you an installable driver set that you can do on a minimal install. I.e., I installed FC4 with kernel 2.6.11, installed the nForce platform driver with its "nvnet" for 10/100[/1000] NIC, ran yum update, then switched back to the GPL "forcedeth" after the reboot into kernel 2.6.14 (did not have to re-install the nForce platform drivers).
I cannot wait for a promise by a Nvidia rep about their future chipsets using SATA NCP technology that will allow an open source driver to be written to be acted on.
Feizhou feizhou@graffiti.net wrote:
Yeah, read about it quite a while ago when it knocked out the P4 2.8 processor.
Intel is returning to the architecture. The original i686 was very efficient, it just didn't scale beyond 1.0GHz initially -- 1.5GHz with some mid-release asynchronous hacks, now possible 2.5GHz with more timing/async changes.
Intel hasn't do a full i686 redesign since the original 1994 Pentium Pro. Understand that's because Intel thought everything would be IA-64 now. You have to plan architectures 5+ years in advance, because it takes at least 12 months to design, and another 24-36 months of timing closure.
So when EPIC/Predication didn't deliver in the first IA-64 Itanium-Merced, even for IA-64's native ISA (not even looking at x86 comaptibility), that's when the Netburst team was formed. They did the design in only 18 months -- by completely by-passing timing closure, using very long pipes that have many stages doing absolutely nothing.
That's where HyperTransport comes in -- it offers two schedulers in the hope that two virtualized cores can put more stages to use. It only works on the horribly inefficient Netburst architctures -- you will _never_ see HyperTransport on the Pentium-M or Intel's newer processors. The concept of multi-threading on a single core lives and dies with NetBurst.
The next evolution is multi-threading across multi-core.
Thanks to their providing docs.
AMD, Intel and nVidia are pretty open with specifications, except when legally bound otherwise. E.g., the biggest and ironic reason why nVidia couldn't share its AGPgart interface until about 18 months ago was a NDA with Intel (long story).
Yes, these are the chums in use in the newer boxes I used to admin. I loved the 3ware + riser card fiasco though.
Well, when you're pushing 200+ traces at 66MHz, there tends to be EMF/EMI issues. 3Ware isn't the only one that has had issues with traces. Remember, 3Ware 7000/8000/9000S (not 9550SX) use 0 wait state, 64-bit ASIC+SRAM devices. Trace length and timing is everything, and heavily affected by EMF/EMI.
Again, I refer back to the i865 v. i875 issues. The traces of a PCB designed for the i875 -- such as the Asus P4C800 -- didn't necessarly work for the exact same chip in the i865, because it tested to lower tolerances.
The problems I have are related to their hardware, not whether there are good drivers or not.
Actually, the firmware has always been the issue. The ASIC+SRAM design was always sound. They've done some stupid things, like the RAID-5 firmware update for the 6000 series (which was _never_ designed for RAID-5). But other than that, it's always been a
The poor latencies just won't let me use a Pinnacle DC10 board without crashing.
??? Let me guess, RAID-5 on a 9500S? ;->
Yeah, whatever. Tyan come out with a board for servers based on a VIA chipset for Pentium III cpus and so I got to deal with them.
The Tyan "Tiger" series is _not_ a workstation/server platform, it's the _desktop_ platform. That's a very common misnomer. The "Thunder" is the workstation/server series. ;->
I cannot wait for a promise by a Nvidia rep about their future chipsets using SATA NCP technology that will allow an open source driver to be written to be acted on.
Do you mean NCQ?
Understand that nVidia is _totally_open_ with their designs right now, including the SATA. But the SATA hardware just doesn't do NCQ at all.
But yes, nVidia has been extremely open.
On Thu, 15 Dec 2005 at 8:40am, Bryan J. Smith wrote
That's where HyperTransport comes in -- it offers two schedulers in the hope that two virtualized cores can put more stages to use. It only works on the horribly inefficient Netburst architctures -- you will _never_ see HyperTransport on the Pentium-M or Intel's newer processors. The concept of multi-threading on a single core lives and dies with NetBurst.
That would be 'HyperThreading' above, not 'HyperTransport', and I know you know that. ;)
That's where HyperTransport comes in -- it offers two schedulers in the hope that two virtualized cores can put more stages to use. It only works on the horribly inefficient Netburst architctures -- you will _never_ see HyperTransport on the Pentium-M or Intel's newer processors. The concept of multi-threading on a single core lives and dies with NetBurst.
I think you meant HyperThreading.
The next evolution is multi-threading across multi-core.
Yes, these are the chums in use in the newer boxes I used to admin. I loved the 3ware + riser card fiasco though.
Well, when you're pushing 200+ traces at 66MHz, there tends to be EMF/EMI issues. 3Ware isn't the only one that has had issues with traces. Remember, 3Ware 7000/8000/9000S (not 9550SX) use 0 wait state, 64-bit ASIC+SRAM devices. Trace length and timing is everything, and heavily affected by EMF/EMI.
Again, I refer back to the i865 v. i875 issues. The traces of a PCB designed for the i875 -- such as the Asus P4C800 -- didn't necessarly work for the exact same chip in the i865, because it tested to lower tolerances.
Ok, thank you for your explanation. I guess that is why we had to get one particular rise from one particular manufacturer.
The problems I have are related to their hardware, not whether there are good drivers or not.
Actually, the firmware has always been the issue. The ASIC+SRAM design was always sound. They've done some stupid things, like the RAID-5 firmware update for the 6000 series (which was _never_ designed for RAID-5). But other than that, it's always been a
I am sorry Bryan but we seem to have some misunderstanding.
3ware is on Intel 7500 motherboards.
The VIA issue is something else entirely.
The poor latencies just won't let me use a Pinnacle DC10 board without crashing.
??? Let me guess, RAID-5 on a 9500S? ;->
This is on a KT400 VIA chipset. Nothing to do with 3ware. This is purely a dumb VIA chipset problem.
The Tyan "Tiger" series is _not_ a workstation/server platform, it's the _desktop_ platform. That's a very common misnomer. The "Thunder" is the workstation/server series. ;->
Tell that to the one who picked the board.
I cannot wait for a promise by a Nvidia rep about their future chipsets using SATA NCP technology that will allow an open source driver to be written to be acted on.
Do you mean NCQ?
Yes :)
Understand that nVidia is _totally_open_ with their designs right now, including the SATA. But the SATA hardware just doesn't do NCQ at all.
Nforce4 Ultra and above support command queueing according to them.
But yes, nVidia has been extremely open.
Yes...where possible. Their SATA/NCQ implementation apparently does not allow them to provide specs or something.
Feizhou feizhou@graffiti.net wrote:
Ok, thank you for your explanation. I guess that is why we had to get one particular rise from one particular manufacturer.
Yep.
I am sorry Bryan but we seem to have some misunderstanding. 3ware is on Intel 7500 motherboards.
Really? You can get an Intel E7500 chipset with 3Ware? Cool! Which models?
The VIA issue is something else entirely.
Oh, I know.
Tell that to the one who picked the board.
Oh, I've seen all sorts of such "selections" in my time.
Yes :)
Seems like the typo bug is going around. ;-> [ Okay, I admit, I hold the record -- by a wide margin ]
Nforce4 Ultra and above support command queueing according to them.
They wish. ;->
Then again, the ATA "controller" is rather "dumb" -- the NCQ is on the IDE of the end-device. It's not like SCSI where there's an intelligent, hardware host.
The AHCI is supposed to be the software/host equivalent of controlling up to 32 ATA devices with commands, etc... But I haven't seen much work well in that regard on even Windows.
Yes...where possible. Their SATA/NCQ implementation apparently does not allow them to provide specs or
something.
Really? Hmmm, I thought the libata guys were pretty happy with them. I could be wrong, as it's been 2nd hand commentary.
I am sorry Bryan but we seem to have some misunderstanding. 3ware is on Intel 7500 motherboards.
Really? You can get an Intel E7500 chipset with 3Ware?
Argh...why would I need a riser if the 3ware were onboard? The 3ware was attached to the board via a PCI-X riser card.
I meant the 3ware + Intel 7501 board + PCI-X riser card fiasco.
Cool! Which models?
Sorry, misunderstanding.
Seems like the typo bug is going around. ;-> [ Okay, I admit, I hold the record -- by a wide margin ]
Nforce4 Ultra and above support command queueing according to them.
They wish. ;->
Eh? http://www.nvidia.com/object/IO_16450.html
They specifically highlight NCQ on page 4 for their SATA implementation.
Then again, the ATA "controller" is rather "dumb" -- the NCQ is on the IDE of the end-device. It's not like SCSI where there's an intelligent, hardware host.
Well, they are not talking about ATA...they are talking about SCSI like tag command queueing on their SATA side of things, you know, Native Command Queueing. No, not NForce 2 or 3 or 4 but Nforce 4 Ultra and above (Nforce4 SLI, Nforce Pro 2xxx)
The AHCI is supposed to be the software/host equivalent of controlling up to 32 ATA devices with commands, etc... But I haven't seen much work well in that regard on even Windows.
Yes...where possible. Their SATA/NCQ implementation apparently does not allow them to provide specs or
something.
Really? Hmmm, I thought the libata guys were pretty happy with them. I could be wrong, as it's been 2nd hand commentary.
Jeff does not get anything. He does not list Nvidia implementations of their SATA controllers as 'open' like the Intel AHCI and Silicon Image 3112 among others.
Feizhou feizhou@graffiti.net wrote:
Eh? http://www.nvidia.com/object/IO_16450.html They specifically highlight NCQ on page 4 for their SATA implementation.
According to the SATA/libata status page: http://linux.yyz.us/sata/sata-status.html#nvidia
I guess they were talking about the MCP-03, instead of the newer MCP-04? In any case, I'm not getting NCQ support on my nForce4 serieschipsets under Windows. I guess some just don't include it?
Well, they are not talking about ATA...
ATA = SATA You can have NCQ on parallel ATA too. It's just a physical/datalink layer difference.
they are talking about SCSI like tag command queueing on their SATA side of things,
Between the drive and [software] host, individually. Then you use [software] host AHCI to control and schedule up to 32 drives.
My point is that it's still not a hardware-based host adapter. It's only the end-device component, with a software host.
you know, Native Command Queueing. No, not NForce 2 or 3 or 4 but Nforce 4 Ultra and above (Nforce4 SLI, Nforce Pro 2xxx)
Yeah, it seems the regular nForce 4 and at least the nForce 410 (not sure about the 430) don't have it.
Jeff does not get anything. He does not list Nvidia implementations of their SATA controllers as 'open' like
the
Intel AHCI and Silicon Image 3112 among others.
I stand corrected then. As I said, I heard 2nd hand.
Bryan J. Smith wrote:
Feizhou feizhou@graffiti.net wrote:
Eh? http://www.nvidia.com/object/IO_16450.html They specifically highlight NCQ on page 4 for their SATA implementation.
According to the SATA/libata status page: http://linux.yyz.us/sata/sata-status.html#nvidia
I guess they were talking about the MCP-03, instead of the newer MCP-04? In any case, I'm not getting NCQ support on my nForce4 serieschipsets under Windows. I guess some just don't include it?
I built a box with the Tomcat K8E which does have NCQ supported on the SATA host controller but it also requires SATA disks that support NCQ...which I did not get. So no NCQ enabled there.
My point is that it's still not a hardware-based host adapter. It's only the end-device component, with a software host.
Yeah, it seems the regular nForce 4 and at least the nForce 410 (not sure about the 430) don't have it.
Hardware support there in the SATA controller. Don't which particular chip.
[ NOTE TO SELF: Proofread dammit! "Bryan J. Smith" thebs413@earthlink.net wrote:
That's where HyperTransport comes in
Er, Hyper_Threading_! Doh!
It only works on the horribly inefficient Netburst architctures -- you will _never_ see HyperTransport on the Pentium-M or Intel's newer processors.
Again, Hyper_Threading_. Doh!
Bryan J. Smith wrote:
[ NOTE TO SELF: Proofread dammit! "Bryan J. Smith" thebs413@earthlink.net wrote:
That's where HyperTransport comes in
Er, Hyper_Threading_! Doh!
It only works on the horribly inefficient Netburst architctures -- you will _never_ see HyperTransport on the Pentium-M or Intel's newer processors.
Again, Hyper_Threading_. Doh!
Opteron-on-the-brain :-) ?
I've only put CentOS onto a fanless mini ATX board with an Eden chip on it, which works fine. But I have installed Fedora onto three home built units with genuine Intel motherboards, model D865PERL, with Seagate SATA drives, which have given me no trouble at all.
You won't be able to buy this motherboard any more I don't think although there might still be some ten packs available.
My suggestion would be to look for something similar to this board but with the newer processor socket. If you go to intel.com and enter the model number you will get the full spec. there with BIOS updates and everything else you need to know.
Hope I've helped.
Dave Fletcher
On Thursday 15 Dec 2005 14:43, Todd Cary wrote:
Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-ASUS&... epts=BUNDLE2&DNAME=Motherboard+Bundles%2D+By+MB
Any suggestions will be appreciated....
Todd
David -
The suggestion of the D865PERL is the kind of information that is most helpful since someone (you) are using it with Centos. Now to find a current board that is in use with Centos and some ATA drives.
Thank you!
Todd
David Fletcher wrote:
I've only put CentOS onto a fanless mini ATX board with an Eden chip on it, which works fine. But I have installed Fedora onto three home built units with genuine Intel motherboards, model D865PERL, with Seagate SATA drives, which have given me no trouble at all.
You won't be able to buy this motherboard any more I don't think although there might still be some ten packs available.
My suggestion would be to look for something similar to this board but with the newer processor socket. If you go to intel.com and enter the model number you will get the full spec. there with BIOS updates and everything else you need to know.
Hope I've helped.
Dave Fletcher
On Thursday 15 Dec 2005 14:43, Todd Cary wrote:
Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-ASUS&... epts=BUNDLE2&DNAME=Motherboard+Bundles%2D+By+MB
Any suggestions will be appreciated....
Todd
I forgot to add that performance is not an issue since this server is used for testing PHP code, FTPing applications via DSL to my clients and viewing photos.
http://209.204.172.137/rotary/
Todd
Todd Cary wrote:
David -
The suggestion of the D865PERL is the kind of information that is most helpful since someone (you) are using it with Centos. Now to find a current board that is in use with Centos and some ATA drives.
Thank you!
Todd
David Fletcher wrote:
I've only put CentOS onto a fanless mini ATX board with an Eden chip on it, which works fine. But I have installed Fedora onto three home built units with genuine Intel motherboards, model D865PERL, with Seagate SATA drives, which have given me no trouble at all.
You won't be able to buy this motherboard any more I don't think although there might still be some ten packs available.
My suggestion would be to look for something similar to this board but with the newer processor socket. If you go to intel.com and enter the model number you will get the full spec. there with BIOS updates and everything else you need to know.
Hope I've helped.
Dave Fletcher
On Thursday 15 Dec 2005 14:43, Todd Cary wrote:
Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
http://www.mwave.com/mwave/ViewProducts.hmx?PID=MOTHERBOARDBUNDLES-ASUS&... epts=BUNDLE2&DNAME=Motherboard+Bundles%2D+By+MB
Any suggestions will be appreciated....
Todd
-- Ariste Software 2200 D Street Ext Petaluma, CA 94952 (707) 773-4523
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Todd Cary todd@aristesoftware.com wrote:
I forgot to add that performance is not an issue since this server is used for testing PHP code, FTPing applications
via
DSL to my clients and viewing photos. http://209.204.172.137/rotary/
Oh, in that case, nVidia C51/NV44 is excellent in the entry Socket-754 platform.
I've just assembled in some $57 ASRock K8NF4G-SATA2 mainboards with $75 Sempron 64 2800+ (latest Rev. E/SSE3) CUPs. You can't beat the combo for the price, especially since you don't have to add a thing (NV44 integrated video, 2xSATA, 2xATA, 8USB2.0, ALC850 7.1, etc...).
Just note the need to update the kernel to get newer PCI ID support -- they changed a few from the rest of the nForce series: http://thebs413.blogspot.com/2005/12/linux-on-nvidia-c51nv44-nforce.html
I haven't tested if the RHEL 4 kernels have backported the newer 2.6.13/14 PCI ID support and updated GPL forcedeth driver. If not, you can always load the nForce platform driver that nvnet (for the newer MAC+PHY IDs) and nvsound (not needed). On Fedora Core 4, the 2.6.14 kernel handles it fine. I need to check the last 2.6.12 kernel for Fedora Core 3 to see if it is the same.
Ironically enough, the NV44 _does_ work with the X11R6.8.2 NV4x code, so that works for 2D. You don't have to use VESA. nVidia just released their first ForceWare 80 driver (1.0-8174) for Linux about 10 days ago, if you want 3D acceleration.
Bryan J. Smith wrote:
Todd Cary todd@aristesoftware.com wrote:
I forgot to add that performance is not an issue since this server is used for testing PHP code, FTPing applications
via
DSL to my clients and viewing photos. http://209.204.172.137/rotary/
Oh, in that case, nVidia C51/NV44 is excellent in the entry Socket-754 platform.
I've just assembled in some $57 ASRock K8NF4G-SATA2 mainboards with $75 Sempron 64 2800+ (latest Rev. E/SSE3) CUPs. You can't beat the combo for the price, especially since you don't have to add a thing (NV44 integrated video, 2xSATA, 2xATA, 8USB2.0, ALC850 7.1, etc...).
Where did you get the Sempron-64's ? I thought only HP/IBM/etc had access to those ....
socket 754 is phasing out...
William A. Mahaffey III wrote:
Bryan J. Smith wrote:
Todd Cary todd@aristesoftware.com mailto:todd@aristesoftware.com wrote:
I forgot to add that performance is not an issue since this server is used for testing PHP code, FTPing applications
via
DSL to my clients and viewing photos. http://209.204.172.137/rotary/
Oh, in that case, nVidia C51/NV44 is excellent in the entry Socket-754 platform.
I've just assembled in some $57 ASRock K8NF4G-SATA2 mainboards with $75 Sempron 64 2800+ (latest Rev. E/SSE3) CUPs. You can't beat the combo for the price, especially since you don't have to add a thing (NV44 integrated video, 2xSATA, 2xATA, 8USB2.0, ALC850 7.1, etc...).
Where did you get the Sempron-64's ? I thought only HP/IBM/etc had access to those ....
-- William A. Mahaffey III
Remember, ignorance is bliss, but willful ignorance is LIBERALISM !!!!
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
"William A. Mahaffey III" wam@HiWAAY.net wrote:
Where did you get the Sempron-64's ? I thought only HP/IBM/etc had access to those ....
Heck, you can't get anything but the Sempron 64 for Socket-754 now. Once Intel started selling the Celeron with EM64T, AMD stopped crippling the Socket-754 Semprons.
Even better is that the two (2) I just got are the full Rev. E units with SSE3 and everything. Not bad for $75!
Maybe you're thinking of the Socket-939 Semprons? I haven't seen those other than that one, new HP Sempron S939 3200+. Remember, Socket-939 *AND* Socket-754 are Athlon 64 capable. It's only the old Socket-A/462 that is not. ;->
Peter Farrow peter@farrows.org wrote:
socket 754 is phasing out...
Everything's being "phased out" next year, although I thought the Socket-754 was still going to be used as a "value" solution for remaining DDR Hammer x86-64 cores/stock(?). Others have stated that Socket-939 is going to be the new Sempron, but I've seen info that Socket-754 is where the value will stay. So who knows?
BTW, as Peter Arremann posted over on the AMD64 list: https://www.redhat.com/archives/amd64-list/2005-December/msg00103.html
XbitLabs just posted AMD's 2006Q2 plans scheduled for April 6th, 2006: http://www.xbitlabs.com/news/cpu/display/20051214062233.html
Socket-754 1xDDR -> Socket-"S1"/638 1xDDR2** Socket-939 2xDDR -> Socket-"M2"/940 2xDDR2 Socket-940 2xDDR -> Socket-"F"/1207 2xDDR2
**NOTE: There were mentions of the Socket-S1/638 being marketed as dual-DDR2, despite being only a single 240-trace DDR2 channel. I think it's still single DDR, but uses dual-DDR interleaving over a single channel like Socket-423/478 and Socket-A/462 were.
In any case, this is probably a discussion left for the Red Hat AMD64 list instead of here. Feel free to join ... (I did back when must of the existing Alpha/AXP list migrated over in 2003): https://www.redhat.com/archives/amd64-list/index.html
Bryan J. Smith wrote:
"William A. Mahaffey III" wam@HiWAAY.net wrote:
Where did you get the Sempron-64's ? I thought only HP/IBM/etc had access to those ....
Heck, you can't get anything but the Sempron 64 for Socket-754 now. Once Intel started selling the Celeron with EM64T, AMD stopped crippling the Socket-754 Semprons.
Even better is that the two (2) I just got are the full Rev. E units with SSE3 and everything. Not bad for $75!
Maybe you're thinking of the Socket-939 Semprons? I haven't seen those other than that one, new HP Sempron S939 3200+. Remember, Socket-939 *AND* Socket-754 are Athlon 64 capable. It's only the old Socket-A/462 that is not. ;->
Yep (939), I stand corrected :-).
On Thu, 2005-12-15 at 17:56 -0600, William A. Mahaffey III wrote:
Yep (939), I stand corrected :-).
Well, I was already ignorant of the fact that they existed in our prior exchange on another list (as you caught me on). Boy did I look like a fool.
On 12/16/05, Bryan J. Smith thebs413@earthlink.net wrote:
On Thu, 2005-12-15 at 17:56 -0600, William A. Mahaffey III wrote:
Yep (939), I stand corrected :-).
Well, I was already ignorant of the fact that they existed in our prior exchange on another list (as you caught me on). Boy did I look like a fool.
No you look like anyone else that makes a mistake, or was not aware of some thing, and admits it.
-- Leonard Isham, CISSP Ostendo non ostento.
David Fletcher centos@fletchersweb.net wrote:
I've only put CentOS onto a fanless mini ATX board with an Eden chip on it, which works fine.
As I noted in another e-mail, ViA's C3/Eden platforms typically lag their AMD/Intel chipset offerings. So by the time the C3/Eden version of logic comes out, Linux already has support.
The CLE266 works flawlessly for me -- and I can drive 480p HDTV video with it (although not 720p). I haven't tried the new CN400 yet, let alone the nano-BGA dual-processor C3 option in Mini-ITX.
But I have installed Fedora onto three home built units
with
genuine Intel motherboards, model D865PERL, with Seagate
SATA
drives, which have given me no trouble at all.
Just FYI, if you're interested in stability, choose the Intel i875. The i865 roll of the same line as the i875, but do not test to the tolerances required for the i875.
This pin compatibility led several vendors to introduce an i865 version of their i875 mainboard design, without retesting the PCB for the lesser i865. The most infamous was the Asus P4P800 using the i865, which used the exact same PCB as its premium P4C800 with the i875 -- but wasn't nearly as stable/reliable.
For workstations or servers, spend the extra $100 or 200, respectively, and get the Intel E7200 or E7500. World of difference in I/O. The latest E7221 for LGA-775 has all the goodies a single processor server would want -- PCI-X, PCIe x8, DDR2, etc...
David Fletcher wrote:
I've only put CentOS onto a fanless mini ATX board with an Eden chip on it, which works fine. But I have installed Fedora onto three home built units with genuine Intel motherboards, model D865PERL, with Seagate SATA drives, which have given me no trouble at all.
I was unlucky to have several of those (D865PERL). The motherboard is real horror. Sure, it works in simple setups, and that's about it. As soon as I attempted to put three dual-port ethernet cards (also Intel brand) into it, it just froze. Intel support refused to solve the problem although both the ethernet PCI cards and the motherboard were manufactured by Intel (they were claiming that I was mixing server and desktop components, blah). There is ton of BIOS updates for it. Some of them issued to solve similar problems (BIOS freezing with various VGA and ethernet cards). However, the problems were never solved completely. I guess after some time Intel just gave up of fixing it. Oh, and also, when connected to some types of KVM switches, for whatever strange reason BIOS would freeze on powerup if machine is not selected on KVM swtich...
Long story short, if you intend to keep all those PCI slots (5 or 6, don't remember how many it has) empty, and you don't intend to connect it to KVM swtich, you'll be fine. If you plan on actually using those PCI slots, you better skip it.
Advices... If you don't have all the other components (processor, memory, etc), look at something into what you can plug 64bit processor (preferably AMD64, Intel EM64T only if you really want/need/must have Intel processor. If the price isn't too big issue, you might want to look into bying motherboard with support for dual core CPU.
If you need/want server grade motherboard, and decide to go with Intel processor (although, IMO, AMD would be smarter choice), look at www.supermicro.com. Note that those are *server* class motherboards, and hence the price. They also have some desktop/workstation stuff, which I guess should be a bit cheaper. The only thing to be carefull is to avoid SuperMicro motherboards with Marvell SATA chipset. Marvell isn't supporte by Linux out-of-the-box (yet). However, there's only one or two SuperMicro motherboards with that chip (and they all seem to also have an Intel SATA controller onboard, which is supported by Linux). I've used many of them (from P3 to Xeons and dual-core Pentium D with EM64T), and they all worked (and still work) rock solid stable. Oh, and also, with SuperMicro make sure you check the form factor (size) of the motherboard. There's couple of them (designed for 1U cases) that are non-standard size and would fit only in SuperMicro 1U case. Unfortunately, they don't have any AMD based motherboards.
On 15/12/05, Todd Cary todd@aristesoftware.com wrote:
Not being familiar with chip sets and other factors that "hardware experts" understand, I am perplexed with the multitude of available motherboards. Is there a Web site that can be of help or a "strategy"?
My goal is to replace my motherboard with one that is Centos compatible and uses a Pentium 4 in the 2 MHz range. As example, when I go to Mwave, I am presented with this list, but I really do not know what I should be looking for in the specs.
You have a couple of choices. Educate yourself a little on the state of current hardware by reading up for a few days on, for example:
http://www.anandtech.com/ http://www.tomshardware.com/ http://arstechnica.com/index.ars http://www.sharkyextreme.com/ etc. etc.
Bearing in mind your end use of the system. If it's going to be a desktop or workstation then focus on articles pertaining to those, if you're looking for server hardware concentrate on those.
You could also have a browse of the RHEL Hardware Compatibility List:
https://hardware.redhat.com/hwcert/index.cgi
Or you can wait for the suggestions from list members. :)
I'm not bang up-to-date with what's going on in hardware at the moment because I haven't had to purchase in quantity recently but FWIW I like Supermicro boards. You might want to try and avoid anything with Adaptec SCSI controller chipsets, I seem to recall reading they're dropping linux support.
Will.
-
Aleksandar Milivojevic -
Bryan J. Smith -
David Fletcher -
Feizhou -
Joshua Baker-LePain -
Leonard Isham -
Maciej Żenczykowski -
Peter Farrow -
Todd Cary -
Will McDonald -
William A. Mahaffey III