Beta version 24.12 available - Page 2

Cruelty · 2005-06-09, 17:39

This is a CG (Newcastle) core with 512 kB L2 cache.

Cruelty · 2005-06-09, 17:44

This is new E3 (Venice) core with 512 kB L2 cache, overclocked to 2520 MHz.

It looks like improvement over 24.11 is slightly smaller than with older A64 core...

moo · 2005-06-09, 18:09

When running I had my music players closed and im clients off so total processes running was 40 there was still 500 megs of ram left to use.

System=
Intel pentium 4 proc clocked at 1.8 ghz northwood core socted 478 fsb=400mhz Not oced.
768 ddr 2100 ram
Intel motherboard d845dbv

Here are my benchmarks

Quote:

Originally Posted by Version 23.8.1

Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.

Intel(R) Pentium(R) 4 CPU 1.80GHz
CPU speed: 1799.83 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 8 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 20.219 ms.
Best time for 448K FFT length: 24.022 ms.
Best time for 512K FFT length: 27.484 ms.
Best time for 640K FFT length: 32.964 ms.
Best time for 768K FFT length: 40.294 ms.
Best time for 896K FFT length: 47.616 ms.
Best time for 1024K FFT length: 53.336 ms.
Best time for 1280K FFT length: 69.812 ms.
Best time for 1536K FFT length: 85.951 ms.
Best time for 1792K FFT length: 102.178 ms.
Best time for 2048K FFT length: 115.703 ms.

Quote:

Originally Posted by Version 24.12

[Thu Jun 09 13:04:18 2005]
Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.

Intel(R) Pentium(R) 4 CPU 1.80GHz
CPU speed: 1799.67 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 8 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.12, RdtscTiming=1
Best time for 512K FFT length: 24.564 ms.
Best time for 640K FFT length: 31.503 ms.
Best time for 768K FFT length: 38.994 ms.
Best time for 896K FFT length: 46.187 ms.
Best time for 1024K FFT length: 51.946 ms.
Best time for 1280K FFT length: 66.591 ms.
Best time for 1536K FFT length: 82.198 ms.
Best time for 1792K FFT length: 96.708 ms.
Best time for 2048K FFT length: 108.096 ms.
Best time for 2560K FFT length: 162.177 ms.
Best time for 3072K FFT length: 177.361 ms.
Best time for 3584K FFT length: 212.283 ms.
Best time for 4096K FFT length: 246.490 ms.
Best time for 58 bit trial factors: 17.951 ms.
Best time for 59 bit trial factors: 18.123 ms.
Best time for 60 bit trial factors: 26.249 ms.
Best time for 61 bit trial factors: 18.163 ms.
Best time for 62 bit trial factors: 17.699 ms.
Best time for 63 bit trial factors: 17.686 ms.
Best time for 64 bit trial factors: 21.947 ms.
Best time for 65 bit trial factors: 21.390 ms.
Best time for 66 bit trial factors: 21.520 ms.
Best time for 67 bit trial factors: 21.636 ms.

Just a little edit
Seams like my system doesnt even stack up to the newer faster systems. ;)
George if you need testing on intel celerions or semprons or pentium 3s tell me.

Glenn Leider · 2005-06-10, 05:57

Here are my results.

[Thu Jun 09 06:31:46 2005]
Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.

Intel(R) Celeron(R) CPU 2.93GHz
CPU speed: 2926.18 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: unknown
L2 cache size: 256 KB
L1 cache line size: unknown
L2 cache line size: 128 bytes
TLBS: 64
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 15.201 ms.
Best time for 448K FFT length: 18.425 ms.
Best time for 512K FFT length: 20.765 ms.
Best time for 640K FFT length: 25.404 ms.
Best time for 768K FFT length: 30.606 ms.
Best time for 896K FFT length: 37.832 ms.
Best time for 1024K FFT length: 41.821 ms.
Best time for 1280K FFT length: 56.317 ms.
Best time for 1536K FFT length: 69.028 ms.
Best time for 1792K FFT length: 88.251 ms.
Best time for 2048K FFT length: 94.696 ms.

One question: Why am I told on the Benchmark Page my current 10-million-digit test will take 108+ days when Prime95 is telling me it'll take 32 1/2 (or 108 x log 2)?

xenon · 2005-06-10, 12:10

Quote:

Originally Posted by Cruelty

There is a significant gain (>20%) by going from 23.8 to 24.11

Didn't work for my Pentium III 1000MHz 256K L2 cache. But my speed is better than yours.

TheJudger · 2005-06-10, 12:58

*hehe* well done

improvements are a bit less when i run 2 clients on my (memory bandwidth-limited) dual Xeon 2.2GHz 400MHz FSB.

Speedup from 24.11 to 24.12 for 768K FFT (this machine does double-checks
so 768K is the size which cares right now):
1 client, other CPU idle: + ~2.3%
2 clients: a bit less than 2% faster than before

George: can you change the output of "Per iteration time" of mprime
from "x.xxx sec" to e.g. "xxx.xxx msecs"?

db597 · 2005-06-10, 14:42

Quote:

Originally Posted by Glenn Leider

Prime95 version 23.8, RdtscTiming=1

Glenn, you are still running 23.8!

lycorn · 2005-06-10, 19:41

The full benchies on a PIV Northwood. A gain of roughly 4-5% from 23.8

.

Intel(R) Pentium(R) 4 CPU 3.20GHz
CPU speed: 3503.12 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 8 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.12, RdtscTiming=1
Best time for 4K FFT length: 0.064 ms.
Best time for 5K FFT length: 0.083 ms.
Best time for 6K FFT length: 0.101 ms.
Best time for 7K FFT length: 0.132 ms.
Best time for 8K FFT length: 0.133 ms.
Best time for 10K FFT length: 0.192 ms.
Best time for 12K FFT length: 0.237 ms.
Best time for 14K FFT length: 0.296 ms.
Best time for 16K FFT length: 0.308 ms.
Best time for 20K FFT length: 0.390 ms.
Best time for 24K FFT length: 0.479 ms.
Best time for 28K FFT length: 0.611 ms.
Best time for 32K FFT length: 0.636 ms.
Best time for 40K FFT length: 0.813 ms.
Best time for 48K FFT length: 1.013 ms.
Best time for 56K FFT length: 1.234 ms.
Best time for 64K FFT length: 1.377 ms.
Best time for 80K FFT length: 1.887 ms.
Best time for 96K FFT length: 2.334 ms.
Best time for 112K FFT length: 2.877 ms.
Best time for 128K FFT length: 3.058 ms.
Best time for 160K FFT length: 3.706 ms.
Best time for 192K FFT length: 4.614 ms.
Best time for 224K FFT length: 5.424 ms.
Best time for 256K FFT length: 6.172 ms.
Best time for 320K FFT length: 7.722 ms.
Best time for 384K FFT length: 9.551 ms.
Best time for 448K FFT length: 11.261 ms.
Best time for 512K FFT length: 12.625 ms.
Best time for 640K FFT length: 16.279 ms.
Best time for 768K FFT length: 20.138 ms.
Best time for 896K FFT length: 23.665 ms.
Best time for 1024K FFT length: 26.488 ms.
Best time for 1280K FFT length: 33.920 ms.
Best time for 1536K FFT length: 41.896 ms.
Best time for 1792K FFT length: 49.281 ms.
Best time for 2048K FFT length: 55.321 ms.
Best time for 2560K FFT length: 73.093 ms.
Best time for 3072K FFT length: 92.949 ms.
Best time for 3584K FFT length: 108.441 ms.
Best time for 4096K FFT length: 125.086 ms.
Best time for 5120K FFT length: 153.703 ms.
Best time for 6144K FFT length: 197.016 ms.
Best time for 7168K FFT length: 234.491 ms.
Best time for 8192K FFT length: 271.404 ms.
Best time for 10240K FFT length: 355.877 ms.
Best time for 12288K FFT length: 469.791 ms.
Best time for 14336K FFT length: 603.695 ms.
Best time for 16384K FFT length: 746.396 ms.
Best time for 20480K FFT length: 990.379 ms.
[Fri Jun 10 20:32:27 2005]
Best time for 24576K FFT length: 1250.233 ms.
Best time for 28672K FFT length: 1580.596 ms.
Best time for 32768K FFT length: 1846.726 ms.
Best time for 58 bit trial factors: 8.224 ms.
Best time for 59 bit trial factors: 8.264 ms.
Best time for 60 bit trial factors: 8.200 ms.
Best time for 61 bit trial factors: 8.262 ms.
Best time for 62 bit trial factors: 9.033 ms.
Best time for 63 bit trial factors: 9.055 ms.
Best time for 64 bit trial factors: 11.011 ms.
Best time for 65 bit trial factors: 10.962 ms.
Best time for 66 bit trial factors: 10.967 ms.
Best time for 67 bit trial factors: 10.941 ms.

lycorn · 2005-06-10, 19:44

Just out of curiosity:
What will be the intended use for the large FFT sizes? We are still far from the 4096K limit... Is it for added precision or just a future proofing operation?

penguin22 · 2005-06-10, 20:01

As Kaboom had mentioned earlier in the thread, will there be a 64 bit client for Windows XP x64 Edition? I have been using 24.11 x64, but switched to this version and have noticed a slight speed increase even though it is 32... not sure about factoring though.

Curious as to why the exe is 4MB now as well...

Dave - penguin22

moo · 2005-06-10, 20:30

there is 64 bit code running around someware i forgot were it should be in a post. Also its larger becase of extra new code for new processers i belive.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
LLR beta Version 3.8.13 (deprecated)	Jean Penné	Software	111	2015-01-26 21:41
Prime95 beta version 28.4	Prime95	Software	20	2014-03-02 02:51
Prime95 beta version 28.3	Prime95	Software	68	2014-02-23 05:42
Early Beta of version 24.11	Prime95	Software	113	2005-05-24 17:05
Beta version of PRP	Prime95	PSearch	15	2004-09-17 19:21

2005-06-10, 05:57	#15
Glenn Leider Aug 2002 Carlsbad, Calif. 45₈ Posts	Here are my results. [Thu Jun 09 06:31:46 2005] Compare your results to other computers at http://www.mersenne.org/bench.htm That web page also contains instructions on how your results can be included. Intel(R) Celeron(R) CPU 2.93GHz CPU speed: 2926.18 MHz CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2 L1 cache size: unknown L2 cache size: 256 KB L1 cache line size: unknown L2 cache line size: 128 bytes TLBS: 64 Prime95 version 23.8, RdtscTiming=1 Best time for 384K FFT length: 15.201 ms. Best time for 448K FFT length: 18.425 ms. Best time for 512K FFT length: 20.765 ms. Best time for 640K FFT length: 25.404 ms. Best time for 768K FFT length: 30.606 ms. Best time for 896K FFT length: 37.832 ms. Best time for 1024K FFT length: 41.821 ms. Best time for 1280K FFT length: 56.317 ms. Best time for 1536K FFT length: 69.028 ms. Best time for 1792K FFT length: 88.251 ms. Best time for 2048K FFT length: 94.696 ms. One question: Why am I told on the Benchmark Page my current 10-million-digit test will take 108+ days when Prime95 is telling me it'll take 32 1/2 (or 108 x log 2)?

2005-06-10, 12:58	#17
TheJudger "Oliver" Mar 2005 Germany 1111₁₀ Posts	hehe well done improvements are a bit less when i run 2 clients on my (memory bandwidth-limited) dual Xeon 2.2GHz 400MHz FSB. Speedup from 24.11 to 24.12 for 768K FFT (this machine does double-checks so 768K is the size which cares right now): 1 client, other CPU idle: + ~2.3% 2 clients: a bit less than 2% faster than before George: can you change the output of "Per iteration time" of mprime from "x.xxx sec" to e.g. "xxx.xxx msecs"?

2005-06-10, 19:41	#19
lycorn Sep 2002 Oeiras, Portugal 2⁶·23 Posts	The full benchies on a PIV Northwood. A gain of roughly 4-5% from 23.8 . Intel(R) Pentium(R) 4 CPU 3.20GHz CPU speed: 3503.12 MHz CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2 L1 cache size: 8 KB L2 cache size: 512 KB L1 cache line size: 64 bytes L2 cache line size: 128 bytes TLBS: 64 Prime95 32-bit version 24.12, RdtscTiming=1 Best time for 4K FFT length: 0.064 ms. Best time for 5K FFT length: 0.083 ms. Best time for 6K FFT length: 0.101 ms. Best time for 7K FFT length: 0.132 ms. Best time for 8K FFT length: 0.133 ms. Best time for 10K FFT length: 0.192 ms. Best time for 12K FFT length: 0.237 ms. Best time for 14K FFT length: 0.296 ms. Best time for 16K FFT length: 0.308 ms. Best time for 20K FFT length: 0.390 ms. Best time for 24K FFT length: 0.479 ms. Best time for 28K FFT length: 0.611 ms. Best time for 32K FFT length: 0.636 ms. Best time for 40K FFT length: 0.813 ms. Best time for 48K FFT length: 1.013 ms. Best time for 56K FFT length: 1.234 ms. Best time for 64K FFT length: 1.377 ms. Best time for 80K FFT length: 1.887 ms. Best time for 96K FFT length: 2.334 ms. Best time for 112K FFT length: 2.877 ms. Best time for 128K FFT length: 3.058 ms. Best time for 160K FFT length: 3.706 ms. Best time for 192K FFT length: 4.614 ms. Best time for 224K FFT length: 5.424 ms. Best time for 256K FFT length: 6.172 ms. Best time for 320K FFT length: 7.722 ms. Best time for 384K FFT length: 9.551 ms. Best time for 448K FFT length: 11.261 ms. Best time for 512K FFT length: 12.625 ms. Best time for 640K FFT length: 16.279 ms. Best time for 768K FFT length: 20.138 ms. Best time for 896K FFT length: 23.665 ms. Best time for 1024K FFT length: 26.488 ms. Best time for 1280K FFT length: 33.920 ms. Best time for 1536K FFT length: 41.896 ms. Best time for 1792K FFT length: 49.281 ms. Best time for 2048K FFT length: 55.321 ms. Best time for 2560K FFT length: 73.093 ms. Best time for 3072K FFT length: 92.949 ms. Best time for 3584K FFT length: 108.441 ms. Best time for 4096K FFT length: 125.086 ms. Best time for 5120K FFT length: 153.703 ms. Best time for 6144K FFT length: 197.016 ms. Best time for 7168K FFT length: 234.491 ms. Best time for 8192K FFT length: 271.404 ms. Best time for 10240K FFT length: 355.877 ms. Best time for 12288K FFT length: 469.791 ms. Best time for 14336K FFT length: 603.695 ms. Best time for 16384K FFT length: 746.396 ms. Best time for 20480K FFT length: 990.379 ms. [Fri Jun 10 20:32:27 2005] Best time for 24576K FFT length: 1250.233 ms. Best time for 28672K FFT length: 1580.596 ms. Best time for 32768K FFT length: 1846.726 ms. Best time for 58 bit trial factors: 8.224 ms. Best time for 59 bit trial factors: 8.264 ms. Best time for 60 bit trial factors: 8.200 ms. Best time for 61 bit trial factors: 8.262 ms. Best time for 62 bit trial factors: 9.033 ms. Best time for 63 bit trial factors: 9.055 ms. Best time for 64 bit trial factors: 11.011 ms. Best time for 65 bit trial factors: 10.962 ms. Best time for 66 bit trial factors: 10.967 ms. Best time for 67 bit trial factors: 10.941 ms.

2005-06-10, 19:44	#20
lycorn Sep 2002 Oeiras, Portugal 2⁶×23 Posts	Just out of curiosity: What will be the intended use for the large FFT sizes? We are still far from the 4096K limit... Is it for added precision or just a future proofing operation?

2005-06-10, 20:01	#21
penguin22 Sep 2002 2²·3 Posts	As Kaboom had mentioned earlier in the thread, will there be a 64 bit client for Windows XP x64 Edition? I have been using 24.11 x64, but switched to this version and have noticed a slight speed increase even though it is 32... not sure about factoring though. Curious as to why the exe is 4MB now as well... Dave - penguin22

2005-06-10, 20:30	#22
moo Jul 2004 Nowhere 809 Posts	there is 64 bit code running around someware i forgot were it should be in a post. Also its larger becase of extra new code for new processers i belive.