Let's buy GIMPS an Opteron! - Page 13

E_tron · 2003-09-03, 03:24

Quote:

Originally Posted by Prime Monster

Quote:

Originally Posted by clowns789

Xyzzy,

When you get the computer to George you should ask Prime Monster if he needs any more motherboards. I noticed he has a server connected to a hub and then ten P4 motherboards and one AMD. I think it might be interesting to contribute if he wants. I'll PM him too.

I actually have 24 mobos (11 P4s, 7 AMD XPs and 6 PIIIs) running headless/diskless/etc :D and yes I am looking for some new boards, but not immediately. Will probably go for a P4 combo again, as I feel quite comfortable with them.

PM

Are you waiting for the release of the Athlon64 and P5, so you can pick up some P4's on the cheap side :) ?

Xyzzy · 2003-09-03, 04:27

It appears that the Redhat preview does not support 32-bit mode, so tonight I'm going to put SuSE on... I'll email updated login info to everyone ASAP...

Xyzzy · 2003-09-03, 06:14

Here is a benchmark from SuSE... I'm pretty sure it is in 32-bit mode...

Code:

AMD Opteron(tm) Processor 140
CPU speed: 1395.95 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 23.5, RdtscTiming=1
Best time for 384K FFT length: 32.590 ms.
Best time for 448K FFT length: 38.989 ms.
Best time for 512K FFT length: 44.109 ms.
Best time for 640K FFT length: 54.951 ms.
Best time for 768K FFT length: 66.901 ms.
Best time for 896K FFT length: 80.841 ms.
Best time for 1024K FFT length: 91.071 ms.
Best time for 1280K FFT length: 121.700 ms.
Best time for 1536K FFT length: 148.412 ms.
Best time for 1792K FFT length: 178.064 ms.
Best time for 2048K FFT length: 200.169 ms.

Xyzzy · 2003-09-03, 06:20

More stuff...

[code:1]opteron:/ # uname -a
Linux opteron 2.4.19 #1 Wed Jun 25 21:34:14 UTC 2003 x86_64 unknown unknown GNU/Linux[/code:1]

[code:1]opteron:/ # hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.21 seconds =609.52 MB/sec
Timing buffered disk reads: 64 MB in 1.39 seconds = 46.04 MB/sec[/code:1]

[code:1]opteron:/ # gcc -v
Reading specs from /usr/lib64/gcc-lib/x86_64-suse-linux/3.3/specs
Configured with: ../configure --enable-threads=posix --prefix=/usr --with-local-prefix=/usr/local --infodir=/usr/share/info --mandir=/usr/share/man --libdir=/usr/lib64 --enable-languages=c,c++,f77,objc,java,ada --disable-checking --enable-libgcj --with-gxx-include-dir=/usr/include/g++ --with-slibdir=/lib64 --with-system-zlib --enable-shared --enable-__cxa_atexit x86_64-suse-linux
Thread model: posix
gcc version 3.3 (SuSE Linux)[/code:1]

Erix · 2003-09-03, 06:30

Quote:

Originally Posted by Xyzzy

No... The 501 just happens to have a router built in... No other PIX (AFAIK) does... It is geared more towards the SOHO market...

Thanks Xyzzy.

Dresdenboy · 2003-09-03, 08:49

Quote:

Originally Posted by Xyzzy

Here is a benchmark from SuSE... I'm pretty sure it is in 32-bit mode...

Could you make a benchmark with disabled SSE2 please? If there are really harming SSE2 issues then x87 code could be a little faster.

Xyzzy · 2003-09-03, 09:37

Quote:

Originally Posted by Dresdenboy

Could you make a benchmark with disabled SSE2 please? If there are really harming SSE2 issues then x87 code could be a little faster.

Code:

AMD Opteron(tm) Processor 140
CPU speed: 1396.08 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE
L1 cache size: 64 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 23.5, RdtscTiming=1
Best time for 384K FFT length: 36.584 ms.
Best time for 448K FFT length: 41.377 ms.
Best time for 512K FFT length: 45.026 ms.
Best time for 640K FFT length: 60.961 ms.
Best time for 768K FFT length: 72.321 ms.
Best time for 896K FFT length: 87.900 ms.
Best time for 1024K FFT length: 96.657 ms.
Best time for 1280K FFT length: 124.453 ms.
Best time for 1536K FFT length: 151.075 ms.
Best time for 1792K FFT length: 180.411 ms.
Best time for 2048K FFT length: 201.764 ms.

Dresdenboy · 2003-09-03, 09:58

Thank you.

Only little differences between x87 and SSE2 here... very interesting. It looks like the current code is limited by one of the architectural units (FPU, cache, decoders, whatever).

Xyzzy · 2003-09-03, 16:26

I just ran the TLC SETI benchmark work unit... It took 4 hours 24 minutes and 18.61 seconds...

You can read how the benchmark works here...

http://www.teamlambchop.com/bench/benchfile.htm

And here are the results...

http://cox-internet.com/setispy/tlctop200.htm

I honestly expected it to be much faster... :(

Xyzzy · 2003-09-04, 07:48

They say Linux is free, if your time is worth nothing...

I spent a whole day tracking down a mysterious bug that was preventing sftp from working... Eventually I found the answer here...

http://lists.suse.com/archive/suse-amd64/2003-Aug/0036.html

I'm only posting this so if I have this problem again I have the solution documented!

dsouza123 · 2003-09-04, 13:40

I am just curious, looking at the TLC SETI benchmark page using the link provided, it has one mysterious colum CpF with values ranging from 1.29 to 11.12 .

What is it and how is it calculated ? Is it some sort of efficiency calc ?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Opteron is Hyperthreaded ?	bgbeuning	Information & Answers	3	2016-01-10 08:26
Opteron web server...	Xyzzy	Lounge	14	2003-11-05 23:07
Opteron Bottleneck??	Prime95	Hardware	31	2003-09-17 06:54
AMD Opteron	naclosagc	Software	27	2003-08-10 19:14
What will an AMD Opteron be classified as ?	dsouza123	Software	4	2003-08-02 14:29

2003-09-03, 04:27	#134
Xyzzy "Mike" Aug 2002 5·17·97 Posts	It appears that the Redhat preview does not support 32-bit mode, so tonight I'm going to put SuSE on... I'll email updated login info to everyone ASAP...

2003-09-03, 06:20	#136
Xyzzy "Mike" Aug 2002 10000000110101₂ Posts	More stuff... [code:1]opteron:/ # uname -a Linux opteron 2.4.19 #1 Wed Jun 25 21:34:14 UTC 2003 x86_64 unknown unknown GNU/Linux[/code:1] [code:1]opteron:/ # hdparm -Tt /dev/hda /dev/hda: Timing buffer-cache reads: 128 MB in 0.21 seconds =609.52 MB/sec Timing buffered disk reads: 64 MB in 1.39 seconds = 46.04 MB/sec[/code:1] [code:1]opteron:/ # gcc -v Reading specs from /usr/lib64/gcc-lib/x86_64-suse-linux/3.3/specs Configured with: ../configure --enable-threads=posix --prefix=/usr --with-local-prefix=/usr/local --infodir=/usr/share/info --mandir=/usr/share/man --libdir=/usr/lib64 --enable-languages=c,c++,f77,objc,java,ada --disable-checking --enable-libgcj --with-gxx-include-dir=/usr/include/g++ --with-slibdir=/lib64 --with-system-zlib --enable-shared --enable-__cxa_atexit x86_64-suse-linux Thread model: posix gcc version 3.3 (SuSE Linux)[/code:1]

2003-09-03, 09:58	#140
Dresdenboy Apr 2003 Berlin, Germany 19² Posts	Thank you. Only little differences between x87 and SSE2 here... very interesting. It looks like the current code is limited by one of the architectural units (FPU, cache, decoders, whatever).

2003-09-03, 16:26	#141
Xyzzy "Mike" Aug 2002 5·17·97 Posts	I just ran the TLC SETI benchmark work unit... It took 4 hours 24 minutes and 18.61 seconds... You can read how the benchmark works here... http://www.teamlambchop.com/bench/benchfile.htm And here are the results... http://cox-internet.com/setispy/tlctop200.htm I honestly expected it to be much faster... :(

2003-09-04, 07:48	#142
Xyzzy "Mike" Aug 2002 8245₁₀ Posts	They say Linux is free, if your time is worth nothing... I spent a whole day tracking down a mysterious bug that was preventing sftp from working... Eventually I found the answer here... http://lists.suse.com/archive/suse-amd64/2003-Aug/0036.html I'm only posting this so if I have this problem again I have the solution documented!

2003-09-04, 13:40	#143
dsouza123 Sep 2002 2×331 Posts	I am just curious, looking at the TLC SETI benchmark page using the link provided, it has one mysterious colum CpF with values ranging from 1.29 to 11.12 . What is it and how is it calculated ? Is it some sort of efficiency calc ?