AMD Phenom: phenomenally slow

[Clint Whaley]

Guys,

I just got my hands on a Phenom X4.  My understanding was this was the desktop
version of the 3rd gen opteron (K10h).  I expected it to have the same FPU
peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2 flops/cycle x87.

However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS as well,
and they also run at 1/2 speed compared to opteron).  The Athlon-64 was 2/4/2,
so it appears to be much crappier than the previous generation hardware.
Does anyone have any insight here?  The AMD techdocs all lead me to believe
4/8/x (no mention of x87, so crippling that is not as suprising).

?
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[M. Edward (Ed) Borasky]

Clint Whaley wrote:

> Guys,
>
> I just got my hands on a Phenom X4.  My understanding was this was the desktop
> version of the 3rd gen opteron (K10h).  I expected it to have the same FPU
> peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2 flops/cycle x87.
>
> However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS as well,
> and they also run at 1/2 speed compared to opteron).  The Athlon-64 was 2/4/2,
> so it appears to be much crappier than the previous generation hardware.
> Does anyone have any insight here?  The AMD techdocs all lead me to believe
> 4/8/x (no mention of x87, so crippling that is not as suprising).
>
> ?
> Clint
>
> **************************************************************************
> ** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
> **************************************************************************
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> Math-atlas-devel mailing list
> Math-atlas-devel@...
> https://lists.sourceforge.net/lists/listinfo/math-atlas-devel
>

BIOS? Motherboard? RAM? Or do you just need to go ask them for a
different one?

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Jeff Hammond-2]

There's quite a bit of discussion online about "erratum" for the
Phenom.  The fix is just to disable to TLB
[http://en.wikipedia.org/wiki/Phenom_(processor)] and it causes
significantly a performance hit for some applications
[http://techreport.com/articles.x/13741,http://www.legitreviews.com/article/618/1/].

I'm not an expert, but
http://www.hardcoreware.net/amd-phenom-is-broken-costs-a-14-performance-hit-to-fix/
contains this gem: "The results were even worse than expected. When
looking directly at cache memory performance, bandwidth dropped by as
much as 38.7%, and latency slowed down by over 50%!"

Since BLAS utilizes cache more than the desktop applications I found
results for, it could be that this is sufficient to explain the
problems you see.

Best,

Jeff

On Thu, Jul 3, 2008 at 11:28 AM, M. Edward (Ed) Borasky
<znmeb@...> wrote:

> Clint Whaley wrote:
>> Guys,
>>
>> I just got my hands on a Phenom X4.  My understanding was this was the desktop
>> version of the 3rd gen opteron (K10h).  I expected it to have the same FPU
>> peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2 flops/cycle x87.
>>
>> However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS as well,
>> and they also run at 1/2 speed compared to opteron).  The Athlon-64 was 2/4/2,
>> so it appears to be much crappier than the previous generation hardware.
>> Does anyone have any insight here?  The AMD techdocs all lead me to believe
>> 4/8/x (no mention of x87, so crippling that is not as suprising).
>>
>> ?
>> Clint
>>
>> **************************************************************************
>> ** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
>> **************************************************************************
>>
>> -------------------------------------------------------------------------
>> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
>> Studies have shown that voting for your favorite open source project,
>> along with a healthy diet, reduces your potential for chronic lameness
>> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
>> _______________________________________________
>> Math-atlas-devel mailing list
>> Math-atlas-devel@...
>> https://lists.sourceforge.net/lists/listinfo/math-atlas-devel
>>
>
> BIOS? Motherboard? RAM? Or do you just need to go ask them for a
> different one?
>
> -------------------------------------------------------------------------
> Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
> Studies have shown that voting for your favorite open source project,
> along with a healthy diet, reduces your potential for chronic lameness
> and boredom. Vote Now at http://www.sourceforge.net/community/cca08
> _______________________________________________
> Math-atlas-devel mailing list
> Math-atlas-devel@...
> https://lists.sourceforge.net/lists/listinfo/math-atlas-devel
>

--
Jeff Hammond
The University of Chicago
http://home.uchicago.edu/~jhammond/

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Mikhail Kuzminsky]

In message from Clint Whaley <whaley@...> (Thu, 03 Jul 2008
10:16:08 -0500):
>Guys,
>
>I just got my hands on a Phenom X4.  My understanding was this was
>the desktop
>version of the 3rd gen opteron (K10h).  I expected it to have the
>same FPU
>peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2
>flops/cycle x87.

AMD says about "Wide FP Accelerator" for both Phenom X3/X4 and
Opteron.
For Opteron it's written directly - about 4 FLOP/cycle. For Phenom I
didn't find this data explicitly.

"Software Optimization Guide for AMD Family 10h Processors" says:
"Previous AMD64 processors supported 64-bit floating-point execution
units. The new AMD Family 10h processors add support for 128-bit
floating-point execution units. As a result, the throughput of both
single-precision and double-precision floating-point SSEx vector
operations has improved by 2X over the previous generation of AMD
processors."

Is Phenom a 10h Processor ? (May be 10h is a part of CPUID result ?).

>
>However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS
>as well,
>and they also run at 1/2 speed compared to opteron).  The Athlon-64
>was 2/4/2,
>so it appears to be much crappier than the previous generation
>hardware.
>Does anyone have any insight here?  The AMD techdocs all lead me to
>believe
>4/8/x (no mention of x87, so crippling that is not as suprising).
>
>?

Atlas (make time) says me about 323.6% (max) on some dgemm on Opteron
2350/2 GHz, this corresponds to 6472 MFLOPS and can't be 2 DP FLOP per
cycle it's clear that chip has 8 GFLOPS.

Mikhail Kuzminsky,
Computer Assistance to Chemical Research Center
Zelinsky Institute of Organic Chemistry,
Moscow

>Clint
>
>**************************************************************************
>** R. Clint Whaley, PhD ** Assist Prof, UTSA **
>www.cs.utsa.edu/~whaley **
>**************************************************************************
>
>-------------------------------------------------------------------------
>Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
>Studies have shown that voting for your favorite open source project,
>along with a healthy diet, reduces your potential for chronic
>lameness
>and boredom. Vote Now at http://www.sourceforge.net/community/cca08
>_______________________________________________
>Math-atlas-devel mailing list
>Math-atlas-devel@...
>https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley]

Guys,

>BIOS? Motherboard? RAM? Or do you just need to go ask them for a
>different one?

I'm trying to get some understanding before attempting to isolate such a
problem.  I.e, I don't want to try swapping out all these components, only
to find that the Phenom really has a 2/4/1 peak.  I was hoping someone
had actually used a Phenom, and if they got 4/8/2, then I would know
something is wrong.  The only thing even possible is BIOS, it seems to me:
there's no new instructions that need to be enabled, etc.  It is difficult
to see how even BIOS could cause half the FPU to be unused, once the mobo
recognizes the chip at all (as mine does) . . .  I get these number with
in-cache timings, so it is not a memory issue . . .

>There's quite a bit of discussion online about "erratum" for the
>Phenom.  The fix is just to disable to TLB
>[http://en.wikipedia.org/wiki/Phenom_(processor)] and it causes
>significantly a performance hit for some applications
>[http://techreport.com/articles.x/13741,http://www.legitreviews.com/article/618/1/].
>
>I'm not an expert, but
>http://www.hardcoreware.net/amd-phenom-is-broken-costs-a-14-performance-hit-to-fix/
>contains this gem: "The results were even worse than expected. When
>looking directly at cache memory performance, bandwidth dropped by as
>much as 38.7%, and latency slowed down by over 50%!"
>
>Since BLAS utilizes cache more than the desktop applications I found
>results for, it could be that this is sufficient to explain the
>problems you see.

Nope, the problem occurs on in-cache data.  My K10h Opteron had that bug,
and I applied the BIOS fix, and it didn't change my peak numbers, which
are 4/8/2 for Opteron K10h.  Supposedly, Phenom with numbers ending in 50
don't have this bug, and mine is a 9750 . . .

>AMD says about "Wide FP Accelerator" for both Phenom X3/X4 and
>Opteron.
>For Opteron it's written directly - about 4 FLOP/cycle. For Phenom I
>didn't find this data explicitly.

This is the same experience I'm having: no confirmation on Phenom FPU peak.
I have a help request out to AMD, but no response so far (only been 1 day).

>Is Phenom a 10h Processor ? (May be 10h is a part of CPUID result ?).

ATLAS & Goto think so :)  Here's the snippit from /proc/cpuinfo:
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 2
model name      : AMD Phenom(tm) 9750 Quad-Core Processor
stepping        : 3
cpu MHz         : 2400.955
cache size      : 512 KB
physical id     : 0
siblings        : 4
core id         : 3
cpu cores       : 4
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good pni cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs


Notice it has the misalignsse, which AFAIK is a k10h feature . . .

>Atlas (make time) says me about 323.6% (max) on some dgemm on Opteron
>2350/2 GHz, this corresponds to 6472 MFLOPS and can't be 2 DP FLOP per
>cycle it's clear that chip has 8 GFLOPS.

Yeah, my K10h Opteron has 4/8/2 as well, it is my Phenom that is showing
2/4/1 . . .

Thanks,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[dean gaudet-2]

On Thu, 3 Jul 2008, Clint Whaley wrote:

> Guys,
>
> I just got my hands on a Phenom X4.  My understanding was this was the desktop
> version of the 3rd gen opteron (K10h).  I expected it to have the same FPU
> peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2 flops/cycle x87.
>
> However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS as well,
> and they also run at 1/2 speed compared to opteron).  The Athlon-64 was 2/4/2,
> so it appears to be much crappier than the previous generation hardware.
> Does anyone have any insight here?  The AMD techdocs all lead me to believe
> 4/8/x (no mention of x87, so crippling that is not as suprising).

in your notation i'm getting 4/8/2 from my 9600 ... that's a pretty weird
problem you've got there!

-dean

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley]

Dean,

>> Guys,
>>
>> I just got my hands on a Phenom X4.  My understanding was this was the desktop
>> version of the 3rd gen opteron (K10h).  I expected it to have the same FPU
>> peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2 flops/cycle x87.
>>
>> However, all my tests indicate it is 2/4/1 (I installed the Goto BLAS as well,
>> and they also run at 1/2 speed compared to opteron).  The Athlon-64 was 2/4/2,
>> so it appears to be much crappier than the previous generation hardware.
>> Does anyone have any insight here?  The AMD techdocs all lead me to believe
>> 4/8/x (no mention of x87, so crippling that is not as suprising).
>
>in your notation i'm getting 4/8/2 from my 9600 ... that's a pretty weird
>problem you've got there!

OK, making progress then!  So, your Phenom 9600 can do more than 2 flops/cycle
in double precision?  Have you installed ATLAS on the machine, and if so
can you post your "make time"?

On my machine, I have installed ATLAS, ACML, and Goto BLAS, and all post the
2/4/x numbers.

I was beginning to believe the machine really had 2/4/1, based on the fact
I sent in an AMD tech help request two days ago and still had not recieved
an answer . . .

I guess I will try updating my BIOS, as its about the only thing I can imagine
having this weird affect.

Thanks,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Mikhail Kuzminsky]

In message from Clint Whaley <whaley@...> (Fri, 04 Jul 2008
08:21:15 -0500):

>Dean,
>
>>> Guys,
>>>
>>> I just got my hands on a Phenom X4.  My understanding was this was
>>>the desktop
>>> version of the 3rd gen opteron (K10h).  I expected it to have the
>>>same FPU
>>> peak: 4 flops/cycle double SSE, 8 flops/cycle single SSE, 2
>>>flops/cycle x87.
>>>
>>> However, all my tests indicate it is 2/4/1 (I installed the Goto
>>>BLAS as well,
>>> and they also run at 1/2 speed compared to opteron).  The Athlon-64
>>>was 2/4/2,
>>> so it appears to be much crappier than the previous generation
>>>hardware.
>>> Does anyone have any insight here?  The AMD techdocs all lead me to
>>>believe
>>> 4/8/x (no mention of x87, so crippling that is not as suprising).
>>
>>in your notation i'm getting 4/8/2 from my 9600 ... that's a pretty
>>weird
>>problem you've got there!
>
>OK, making progress then!  So, your Phenom 9600 can do more than 2
>flops/cycle
>in double precision?  Have you installed ATLAS on the machine, and if
>so
>can you post your "make time"?
>
>On my machine, I have installed ATLAS, ACML, and Goto BLAS, and all
>post the
>2/4/x numbers.
>
>I was beginning to believe the machine really had 2/4/1, based on the
>fact
>I sent in an AMD tech help request two days ago and still had not
>recieved
>an answer . . .
>
>I guess I will try updating my BIOS, as its about the only thing I
>can imagine
>having this weird affect.

Taking into account that you have extremally strange result - pls
don't be offended for stupid's question - may be you have running
things like powersaved daemon etc - and therefore at your tests the
CPU frequency was twice decreased ?

Yours
Mikhail Kuzminsky
Computer Assistance to Chemical Research Center
Zelinsky Institute of Organic Chemistry
Moscow

>
>Thanks,
>Clint
>
>**************************************************************************
>** R. Clint Whaley, PhD ** Assist Prof, UTSA **
>www.cs.utsa.edu/~whaley **
>**************************************************************************
>
>-------------------------------------------------------------------------
>Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
>Studies have shown that voting for your favorite open source project,
>along with a healthy diet, reduces your potential for chronic
>lameness
>and boredom. Vote Now at http://www.sourceforge.net/community/cca08
>_______________________________________________
>Math-atlas-devel mailing list
>Math-atlas-devel@...
>https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley]

Guys,

>Taking into account that you have extremally strange result - pls
>don't be offended for stupid's question - may be you have running
>things like powersaved daemon etc - and therefore at your tests the
>CPU frequency was twice decreased ?

Apparently, the real problem is that the question was not stupid enough!
I just fixed the problem via the complicated technique of: rebooting the
machine :)

I have no idea what was causing the problem; my only guess involves the fact
that I had applied a boatload of updates that might have included kernel
patches (I don't remember what-all updates I'd installed), and this was
first reboot since then . . .

Anyway, I'm getting the 4/8/2 now.

Thanks,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[M. Edward (Ed) Borasky]

Clint Whaley wrote:

> Guys,
>
>> Taking into account that you have extremally strange result - pls
>> don't be offended for stupid's question - may be you have running
>> things like powersaved daemon etc - and therefore at your tests the
>> CPU frequency was twice decreased ?
>
> Apparently, the real problem is that the question was not stupid enough!
> I just fixed the problem via the complicated technique of: rebooting the
> machine :)
>
> I have no idea what was causing the problem; my only guess involves the fact
> that I had applied a boatload of updates that might have included kernel
> patches (I don't remember what-all updates I'd installed), and this was
> first reboot since then . . .
>
> Anyway, I'm getting the 4/8/2 now.

What OS/kernel version?

--
M. Edward (Ed) Borasky
http://ruby-perspectives.blogspot.com/

"A mathematician is a machine for turning coffee into theorems." --
Alfréd Rényi via Paul Erdős

[znmeb.vcf]

---

begin:vcard
fn:M. Edward (Ed) Borasky
n:;M. Edward (Ed) Borasky
email;internet:znmeb@...
x-mozilla-html:FALSE
version:2.1
end:vcard

---

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley]

Ed,

>What OS/kernel version?

Kubuntu 8.04:
>drteeth>uname -a
>Linux drteeth 2.6.24-19-generic #1 SMP Wed Jun 18 14:15:37 UTC 2008 x86_64 GNU/Linux

Cheers,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley]

Guys,

> I have no idea what was causing the problem; my only guess involves the fact
> that I had applied a boatload of updates that might have included kernel
> patches (I don't remember what-all updates I'd installed), and this was
> first reboot since then . . .
>
> Anyway, I'm getting the 4/8/2 now.

Well, not really fixed after all.  It now looks a lot like thermal throttling,
in that if I run a kernel test once, I get >8Gflop, but if I run it enough
times without letup, it eventually drops to 4Gflop.  However, as far as I
can tell my CPU temperature is not going up, so I'm not sure what is going on.
I have "cool & quiet" turned off, and according to cpuinfo, my Mhz never
changes, and according to lmsensors my CPU temp is constant at 43-44C (not
sure I trust lmsensors, since it says it doesn't really support k10, but
reboots into bios also show roughly this temp), but this looks a lot
like thermal throttling in behavior.  I've updated the BIOS, so it looks
more and more like I've got a problem in the mobo or processor
(i.e. one or the other of these triggers thermal throttling even when
temp is fine), or something else I'm not even seeing.

Anyway, the problem is definitely with my setup, so now I guess I'll see
if AMD ever gets back to me on the help requests.  Just thought you'd like
to know that the reboot wasn't the problem . . .

Cheers,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

[Clint Whaley-2]

Guys,

This issue is finally resolved: I had a bad motherboard.  I replaced my
ASUS M3A78-EMH mobo with a TYAN K9A2 CF, and my performance numbers
now hold steady.  Not sure if it is the individual board that's the
problem or all boards of this type . . .

Cheers,
Clint

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel