mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Hardware

Reply
 
Thread Tools
Old 2005-04-29, 15:18   #177
db597
 
db597's Avatar
 
Jan 2003

3138 Posts
Default 23.8->24.11: Small speed up on Prescott

My benchmarks for primality tests show between 2% to 2.5% improvement in speed over v23.8. I've omitted trial factoring results - for speed reasons, these should obviously be done on an Athlon64 instead.

Intel(R) Pentium(R) 4 CPU 3.00GHz
CPU speed: 3606.52 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: unknown
L2 cache size: 1024 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: 11.277 ms.
Best time for 448K FFT length: 13.754 ms.
Best time for 512K FFT length: 15.516 ms.
Best time for 640K FFT length: 18.498 ms.
Best time for 768K FFT length: 22.251 ms.
Best time for 896K FFT length: 26.829 ms.
Best time for 1024K FFT length: 29.903 ms.
Best time for 1280K FFT length: 39.753 ms.
Best time for 1536K FFT length: 47.627 ms.
Best time for 1792K FFT length: 57.110 ms.
Best time for 2048K FFT length: 63.914 ms.


------------------------------------------------------------------


Intel(R) Pentium(R) 4 CPU 3.00GHz
CPU speed: 3606.36 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 15.204 ms.
Best time for 640K FFT length: 17.987 ms.
Best time for 768K FFT length: 21.828 ms.
Best time for 896K FFT length: 26.251 ms.
Best time for 1024K FFT length: 29.212 ms.
Best time for 1280K FFT length: 38.418 ms.
Best time for 1536K FFT length: 46.333 ms.
Best time for 1792K FFT length: 55.928 ms.
Best time for 2048K FFT length: 62.366 ms.
db597 is offline   Reply With Quote
Old 2005-05-02, 03:28   #178
Peter Nelson
 
Peter Nelson's Avatar
 
Oct 2004

232 Posts
Default INTEL 630 benchmark (in 32-bit mode)

I was able to benchmark prime95 on one of Intel's new processors.
Interesting to compare with standard Northwood/Prescott 3GHz processors.

Not massively fast, dual channel memory would help it, but at least it runs ok. There are big potential speedups if a 64 bit client can be made to work on this processor but a 64 bit OS would be required. Note the L2 cache is 2MB.

[INTEL 630 @ native 3GHz]
Primenet client version 24.11.2
HP/Compaq machine running SINGLE channel 512MB memory using XP HOME.


Intel(R) Pentium(R) 4 CPU 3.00GHz
CPU speed: 3000.78 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 2048 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
TLBS: 64
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 18.358 ms.
Best time for 640K FFT length: 21.796 ms.
Best time for 768K FFT length: 26.365 ms.
Best time for 896K FFT length: 31.650 ms.
Best time for 1024K FFT length: 35.298 ms.
Best time for 1280K FFT length: 46.438 ms.
Best time for 1536K FFT length: 55.953 ms.
Best time for 1792K FFT length: 67.795 ms.
Best time for 2048K FFT length: 75.294 ms.
Best time for 58 bit trial factors: 9.258 ms.
Best time for 59 bit trial factors: 9.919 ms.
Best time for 60 bit trial factors: 9.899 ms.
Best time for 61 bit trial factors: 9.947 ms.
Best time for 62 bit trial factors: 13.751 ms.
Best time for 63 bit trial factors: 13.761 ms.
Best time for 64 bit trial factors: 15.693 ms.
Best time for 65 bit trial factors: 15.589 ms.
Best time for 66 bit trial factors: 15.661 ms.
Best time for 67 bit trial factors: 15.525 ms.

Last fiddled with by Peter Nelson on 2005-05-02 at 03:32
Peter Nelson is offline   Reply With Quote
Old 2005-05-21, 06:07   #179
dans
 
dans's Avatar
 
Oct 2002

3×7 Posts
Default Windows x64 release version benchmanrks

Using the x64 version of 24.11 I benchmarked a cheap-o Frys (ECS motherboard, AMD Athlon x64 3000 combo) with 512mb SDRAM, using the release version of Microsofts Windows XP 2003 x64 SP1. Here are the three sequential benchmarks:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1998.94 MHz
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.234 ms.
Best time for 640K FFT length: 34.057 ms.
Best time for 768K FFT length: 40.837 ms.
Best time for 896K FFT length: 49.316 ms.
Best time for 1024K FFT length: 55.013 ms.
Best time for 1280K FFT length: 73.149 ms.
Best time for 1536K FFT length: 90.600 ms.
Best time for 1792K FFT length: 108.582 ms.
Best time for 2048K FFT length: 121.559 ms.

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1998.94 MHz
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.384 ms.
Best time for 640K FFT length: 34.024 ms.
Best time for 768K FFT length: 40.827 ms.
Best time for 896K FFT length: 49.260 ms.
Best time for 1024K FFT length: 55.037 ms.
Best time for 1280K FFT length: 73.128 ms.
Best time for 1536K FFT length: 90.858 ms.
Best time for 1792K FFT length: 108.570 ms.
Best time for 2048K FFT length: 121.433 ms.

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1998.94 MHz
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.387 ms.
Best time for 640K FFT length: 34.020 ms.
Best time for 768K FFT length: 40.843 ms.
Best time for 896K FFT length: 49.293 ms.
Best time for 1024K FFT length: 55.024 ms.
Best time for 1280K FFT length: 73.018 ms.
Best time for 1536K FFT length: 90.954 ms.
Best time for 1792K FFT length: 108.646 ms.
Best time for 2048K FFT length: 121.696 ms.
dans is offline   Reply With Quote
Old 2005-07-13, 04:26   #180
nomadicus
 
nomadicus's Avatar
 
Jan 2003
North Carolina

2·3·41 Posts
Default

AMD X2 4400+ dual-core, DFI mobo, OCZ memory

CPU0 - Benchmark
CPU1 - Idle
Code:
[Tue Jul 12 21:00:53 2005]
AMD Athlon(tm) 64 X2 Dual Core Processor 4400+
CPU speed: 2210.14 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, 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 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 21.049 ms.
Best time for 640K FFT length: 27.487 ms.
Best time for 768K FFT length: 33.360 ms.
Best time for 896K FFT length: 39.836 ms.
Best time for 1024K FFT length: 44.425 ms.
Best time for 1280K FFT length: 56.564 ms.
Best time for 1536K FFT length: 69.115 ms.
Best time for 1792K FFT length: 83.712 ms.
Best time for 2048K FFT length: 93.139 ms.
Best time for 2560K FFT length: 122.299 ms.
Best time for 3072K FFT length: 149.325 ms.
Best time for 3584K FFT length: 180.292 ms.
Best time for 4096K FFT length: 200.947 ms.
Best time for 58 bit trial factors: 5.387 ms.
Best time for 59 bit trial factors: 5.402 ms.
Best time for 60 bit trial factors: 5.381 ms.
Best time for 61 bit trial factors: 5.402 ms.
Best time for 62 bit trial factors: 9.684 ms.
Best time for 63 bit trial factors: 9.718 ms.
Best time for 64 bit trial factors: 12.549 ms.
Best time for 65 bit trial factors: 12.460 ms.
Best time for 66 bit trial factors: 12.471 ms.
Best time for 67 bit trial factors: 12.426 ms.
CPU0 - Benchmark
CPU1 - Torture test: Custom, 1024FFT, 512MB
Code:
AMD Athlon(tm) 64 X2 Dual Core Processor 4400+
CPU speed: 2210.14 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, 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 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 21.042 ms.
Best time for 640K FFT length: 27.546 ms.
Best time for 768K FFT length: 33.439 ms.
Best time for 896K FFT length: 39.926 ms.
Best time for 1024K FFT length: 44.548 ms.
Best time for 1280K FFT length: 56.819 ms.
Best time for 1536K FFT length: 69.385 ms.
Best time for 1792K FFT length: 83.998 ms.
Best time for 2048K FFT length: 93.520 ms.
Best time for 2560K FFT length: 122.645 ms.
Best time for 3072K FFT length: 149.550 ms.
Best time for 3584K FFT length: 180.499 ms.
Best time for 4096K FFT length: 201.461 ms.
Best time for 58 bit trial factors: 5.387 ms.
Best time for 59 bit trial factors: 5.402 ms.
Best time for 60 bit trial factors: 5.379 ms.
Best time for 61 bit trial factors: 5.407 ms.
Best time for 62 bit trial factors: 9.674 ms.
Best time for 63 bit trial factors: 9.712 ms.
Best time for 64 bit trial factors: 12.549 ms.
Best time for 65 bit trial factors: 12.452 ms.
Best time for 66 bit trial factors: 12.471 ms.
Best time for 67 bit trial factors: 12.417 ms.
CPU0 - Idle
CPU1 - Benchmark
Code:
[Tue Jul 12 21:02:35 2005]
AMD Athlon(tm) 64 X2 Dual Core Processor 4400+
CPU speed: 2210.14 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, 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 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 21.061 ms.
Best time for 640K FFT length: 27.479 ms.
Best time for 768K FFT length: 33.349 ms.
Best time for 896K FFT length: 39.856 ms.
Best time for 1024K FFT length: 44.431 ms.
Best time for 1280K FFT length: 56.533 ms.
Best time for 1536K FFT length: 69.104 ms.
Best time for 1792K FFT length: 83.691 ms.
Best time for 2048K FFT length: 93.151 ms.
Best time for 2560K FFT length: 122.160 ms.
Best time for 3072K FFT length: 149.143 ms.
Best time for 3584K FFT length: 180.023 ms.
Best time for 4096K FFT length: 200.761 ms.
Best time for 58 bit trial factors: 5.389 ms.
Best time for 59 bit trial factors: 5.400 ms.
Best time for 60 bit trial factors: 5.378 ms.
Best time for 61 bit trial factors: 5.405 ms.
Best time for 62 bit trial factors: 9.684 ms.
Best time for 63 bit trial factors: 9.712 ms.
Best time for 64 bit trial factors: 12.540 ms.
Best time for 65 bit trial factors: 12.454 ms.
Best time for 66 bit trial factors: 12.464 ms.
Best time for 67 bit trial factors: 12.417 ms.
CPU0 - Torture test: Custom, 1024FFT, 512MB
CPU1 - Benchmark results
Code:
AMD Athlon(tm) 64 X2 Dual Core Processor 4400+
CPU speed: 2210.15 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, 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 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 21.031 ms.
Best time for 640K FFT length: 27.525 ms.
Best time for 768K FFT length: 33.428 ms.
Best time for 896K FFT length: 39.945 ms.
Best time for 1024K FFT length: 44.504 ms.
Best time for 1280K FFT length: 56.806 ms.
Best time for 1536K FFT length: 69.468 ms.
Best time for 1792K FFT length: 84.002 ms.
Best time for 2048K FFT length: 93.488 ms.
Best time for 2560K FFT length: 122.551 ms.
Best time for 3072K FFT length: 149.565 ms.
Best time for 3584K FFT length: 180.501 ms.
Best time for 4096K FFT length: 201.201 ms.
Best time for 58 bit trial factors: 5.403 ms.
Best time for 59 bit trial factors: 5.401 ms.
Best time for 60 bit trial factors: 5.383 ms.
Best time for 61 bit trial factors: 5.405 ms.
Best time for 62 bit trial factors: 9.695 ms.
Best time for 63 bit trial factors: 9.716 ms.
Best time for 64 bit trial factors: 12.538 ms.
Best time for 65 bit trial factors: 12.493 ms.
Best time for 66 bit trial factors: 12.464 ms.
Best time for 67 bit trial factors: 12.421 ms.
HTT looks like it has eliminated almost all memory contention.
Nice coding George!
nomadicus is offline   Reply With Quote
Old 2005-07-13, 23:40   #181
wfgarnett3
 
wfgarnett3's Avatar
 
"William Garnett III"
Oct 2002
Bensalem, PA

2×43 Posts
Default

Can you have both "CPUs" run their own respective benchmarks simultaneously, and then post the results? Do 2 dual cores almost behave like 2 separate PCs (i.e. twice as much done)? Anyone have a dual core P4 do test on?
wfgarnett3 is offline   Reply With Quote
Old 2005-07-14, 06:43   #182
Dresdenboy
 
Dresdenboy's Avatar
 
Apr 2003
Berlin, Germany

192 Posts
Default

Quote:
Originally Posted by nomadicus
HTT looks like it has eliminated almost all memory contention.
Nice coding George!
Did you mean the Hypertransport link? If so, then your statement is not correct, because the link between the 2 cores is avoiding the HT controller. Communication happens via the SRQ ("System Request Queue"), which is the next thing after the L2 for both cores. IIRC bandwidth is 64bit@cpu clock in both directions (Hypertransport is 16bit@DDR2000 in both directions) and latency is also lower.

Quote:
Originally Posted by wfgarnett3
Can you have both "CPUs" run their own respective benchmarks simultaneously, and then post the results? Do 2 dual cores almost behave like 2 separate PCs (i.e. twice as much done)? Anyone have a dual core P4 do test on?
Some of the benchmarks done by nomadicus should already give you the desired numbers. I think, it doesn't matter, if the non-benched core runs a torture test with big FFTs or another benchmark, except if the second benchmark is running small FFTs while the other the big ones, which would corrupt the result.

What we see there is what I expected according to the architecture and the way, Prime95 works. Prime95 now needs low latency for memory accesses more than high bandwidth.

This can be seen in the development of the 24.xx versions on a Socket 754 (single channel) A64 vs. a Socket 939 (dual channel) A64. One observation is, that the improvements on S754 were greater than on S939 and the results came closer to the latter. This means, Prime95 became less dependent on memory bandwidth and maybe memory accesses at all (since George mainly optimized the use of the caches).

For results of 2 x Prime95 on a Smithfield, you could also take results of a dual Xeon system with 2 Prescotts and FSB800. If memory type and behaviour of the chipset are similar, the results should be too.
Dresdenboy is offline   Reply With Quote
Old 2005-07-15, 05:45   #183
Mike
 
May 2004
Vancouver, Canada

22×52 Posts
Default

Dell Latitude D610

Intel(R) Pentium(R) M processor 2.00GHz
CPU speed: 798.04 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 32 KB
L2 cache size: 2048 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
TLBS: 128
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 34.729 ms.
Best time for 640K FFT length: 45.092 ms.
Best time for 768K FFT length: 55.294 ms.
Best time for 896K FFT length: 66.351 ms.
Best time for 1024K FFT length: 76.485 ms.
Best time for 1280K FFT length: 93.653 ms.
Best time for 1536K FFT length: 114.354 ms.
Best time for 1792K FFT length: 137.284 ms.
Best time for 2048K FFT length: 154.242 ms.
Best time for 2560K FFT length: 197.927 ms.
Best time for 3072K FFT length: 241.244 ms.
Best time for 3584K FFT length: 288.680 ms.
Best time for 4096K FFT length: 323.769 ms.
Best time for 58 bit trial factors: 6.590 ms.
Best time for 59 bit trial factors: 6.617 ms.
Best time for 60 bit trial factors: 6.609 ms.
Best time for 61 bit trial factors: 6.599 ms.
Best time for 62 bit trial factors: 9.313 ms.
Best time for 63 bit trial factors: 9.359 ms.
Best time for 64 bit trial factors: 16.100 ms.
Best time for 65 bit trial factors: 15.991 ms.
Best time for 66 bit trial factors: 16.019 ms.
Best time for 67 bit trial factors: 15.981 ms.


Btw, it says it's running at about 800 MHz, but it's actually at 1.99 GHz (100% according to Windows). There's no hyperthreading either

Last fiddled with by Mike on 2005-07-15 at 05:48
Mike is offline   Reply With Quote
Old 2005-07-16, 05:52   #184
wfgarnett3
 
wfgarnett3's Avatar
 
"William Garnett III"
Oct 2002
Bensalem, PA

2×43 Posts
Default anybody have a dual core P4?

Does anyone have a dual core P4? If so, can you please post some benchmarks similar to nomadicus above. I am curious that when running 2 instances of Prime95 on a dual core P4, how close do you get to getting "twice" as much done? Nomadicus benchmarks seem to show the dual core X2 from AMD gets real close. Finally, even though for single cores, the Pentium 4 is faster, which architecture (at same clock speeds) gets 2 tests done quicker, a dual core X2 or a dual core P4? Is P4 overall still ahead, or will a dual core X2 be the new CPU of choice?
wfgarnett3 is offline   Reply With Quote
Old 2005-07-16, 10:00   #185
lunna
 
Mar 2003

208 Posts
Default

Pentium D 830 (dual-core 3.0 GHz, no hyperthreading)
ASUS P5WD2 Premium/WiFi-TV (Intel 955X chipset)
Kingston ValueRAM 2*1GB PC2-4300 ECC memory (DDR2-533 CL4)
Windows XP Professional SP2 (32-bit)

(The benchmarks in the same order as nomadicus to make comparisons. Also, benchmarks only run once so no "taking the best run".)

CPU0 - Benchmark
CPU1 - Idle
Code:
Intel(R) Pentium(R) D CPU 3.00GHz
CPU speed: 3010.65 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 15.576 ms.
Best time for 640K FFT length: 19.791 ms.
Best time for 768K FFT length: 24.072 ms.
Best time for 896K FFT length: 29.257 ms.
Best time for 1024K FFT length: 33.584 ms.
Best time for 1280K FFT length: 41.234 ms.
Best time for 1536K FFT length: 50.019 ms.
Best time for 1792K FFT length: 59.813 ms.
Best time for 2048K FFT length: 66.894 ms.
Best time for 2560K FFT length: 88.706 ms.
Best time for 3072K FFT length: 106.831 ms.
Best time for 3584K FFT length: 128.068 ms.
Best time for 4096K FFT length: 143.323 ms.
Best time for 58 bit trial factors: 9.224 ms.
Best time for 59 bit trial factors: 9.251 ms.
Best time for 60 bit trial factors: 9.165 ms.
Best time for 61 bit trial factors: 9.174 ms.
Best time for 62 bit trial factors: 12.597 ms.
Best time for 63 bit trial factors: 12.619 ms.
Best time for 64 bit trial factors: 14.757 ms.
Best time for 65 bit trial factors: 14.734 ms.
Best time for 66 bit trial factors: 14.706 ms.
Best time for 67 bit trial factors: 14.622 ms.
CPU0 - Benchmark
CPU1 - Torture test: Custom, 1024FFT, 512MB
Code:
Intel(R) Pentium(R) D CPU 3.00GHz
CPU speed: 3010.56 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 15.750 ms.
Best time for 640K FFT length: 20.044 ms.
Best time for 768K FFT length: 24.368 ms.
Best time for 896K FFT length: 29.629 ms.
Best time for 1024K FFT length: 34.155 ms.
Best time for 1280K FFT length: 41.928 ms.
Best time for 1536K FFT length: 50.902 ms.
Best time for 1792K FFT length: 60.814 ms.
Best time for 2048K FFT length: 67.936 ms.
Best time for 2560K FFT length: 89.980 ms.
Best time for 3072K FFT length: 108.772 ms.
Best time for 3584K FFT length: 130.326 ms.
Best time for 4096K FFT length: 146.301 ms.
Best time for 58 bit trial factors: 9.222 ms.
Best time for 59 bit trial factors: 9.208 ms.
Best time for 60 bit trial factors: 9.146 ms.
Best time for 61 bit trial factors: 9.201 ms.
Best time for 62 bit trial factors: 12.639 ms.
Best time for 63 bit trial factors: 12.623 ms.
Best time for 64 bit trial factors: 14.853 ms.
Best time for 65 bit trial factors: 14.795 ms.
Best time for 66 bit trial factors: 14.768 ms.
Best time for 67 bit trial factors: 14.704 ms.
CPU0 - Idle
CPU1 - Benchmark
Code:
Intel(R) Pentium(R) D CPU 3.00GHz
CPU speed: 3010.54 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 15.534 ms.
Best time for 640K FFT length: 19.746 ms.
Best time for 768K FFT length: 24.000 ms.
Best time for 896K FFT length: 29.139 ms.
Best time for 1024K FFT length: 33.482 ms.
Best time for 1280K FFT length: 41.112 ms.
Best time for 1536K FFT length: 49.799 ms.
Best time for 1792K FFT length: 59.478 ms.
Best time for 2048K FFT length: 66.362 ms.
Best time for 2560K FFT length: 88.267 ms.
Best time for 3072K FFT length: 106.235 ms.
Best time for 3584K FFT length: 127.332 ms.
Best time for 4096K FFT length: 142.559 ms.
Best time for 58 bit trial factors: 9.137 ms.
Best time for 59 bit trial factors: 9.166 ms.
Best time for 60 bit trial factors: 9.150 ms.
Best time for 61 bit trial factors: 9.158 ms.
Best time for 62 bit trial factors: 12.582 ms.
Best time for 63 bit trial factors: 12.583 ms.
Best time for 64 bit trial factors: 14.718 ms.
Best time for 65 bit trial factors: 14.657 ms.
Best time for 66 bit trial factors: 14.662 ms.
Best time for 67 bit trial factors: 14.593 ms.
CPU0 - Torture test: Custom, 1024FFT, 512MB
CPU1 - Benchmark results
Code:
Intel(R) Pentium(R) D CPU 3.00GHz
CPU speed: 3010.51 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 16 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 15.679 ms.
Best time for 640K FFT length: 19.797 ms.
Best time for 768K FFT length: 24.237 ms.
Best time for 896K FFT length: 29.451 ms.
Best time for 1024K FFT length: 33.796 ms.
Best time for 1280K FFT length: 41.628 ms.
Best time for 1536K FFT length: 50.561 ms.
Best time for 1792K FFT length: 60.404 ms.
Best time for 2048K FFT length: 67.470 ms.
Best time for 2560K FFT length: 89.416 ms.
Best time for 3072K FFT length: 107.962 ms.
Best time for 3584K FFT length: 129.294 ms.
Best time for 4096K FFT length: 144.779 ms.
Best time for 58 bit trial factors: 9.173 ms.
Best time for 59 bit trial factors: 9.169 ms.
Best time for 60 bit trial factors: 9.168 ms.
Best time for 61 bit trial factors: 9.191 ms.
Best time for 62 bit trial factors: 12.579 ms.
Best time for 63 bit trial factors: 12.570 ms.
Best time for 64 bit trial factors: 14.762 ms.
Best time for 65 bit trial factors: 14.714 ms.
Best time for 66 bit trial factors: 14.654 ms.
Best time for 67 bit trial factors: 14.651 ms.
When not doing these benchmarks I have it overclocked to 3.5 GHz and both cores do LL tests with the 2048K FFT size. The iteration times are in the 58-60 ms area for both cores.

Xyzzy: Do you still want "FullBench=1" in prime.ini when doing the benchmarks?
lunna is offline   Reply With Quote
Old 2005-07-17, 12:41   #186
lunna
 
Mar 2003

24 Posts
Default

Because dual-core machines are still new I made a screendump of what LL testing looks like. Upper left is core 0, lower right is core 1.

You can see the machine isn't used much by the consistant iteration times - it's still waiting for some hardware components before the real install is done.

Also, LL testing on core 0 is slightly slower which I believe is due to the Windows scheduler putting jobs on the first available processor, and since none is available, on the first one.
Attached Thumbnails
Click image for larger version

Name:	Prime95_dualcore.gif
Views:	477
Size:	22.1 KB
ID:	645  
lunna is offline   Reply With Quote
Old 2005-07-17, 16:48   #187
Dresdenboy
 
Dresdenboy's Avatar
 
Apr 2003
Berlin, Germany

192 Posts
Default

Quote:
Originally Posted by lunna
Because dual-core machines are still new I made a screendump of what LL testing looks like. Upper left is core 0, lower right is core 1.

You can see the machine isn't used much by the consistant iteration times - it's still waiting for some hardware components before the real install is done.

Also, LL testing on core 0 is slightly slower which I believe is due to the Windows scheduler putting jobs on the first available processor, and since none is available, on the first one.
This difference of only 1 ms is something everybody could experience when running Prime95, closing it, and running it again on a single core (+/- HT) CPU.
Dresdenboy is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Perpetual "interesting video" thread... Xyzzy Lounge 34 2021-02-08 23:42
LLR benchmark thread Oddball Riesel Prime Search 5 2010-08-02 00:11
Perpetual I'm pi**ed off thread rogue Soap Box 19 2009-10-28 19:17
Perpetual autostereogram thread... Xyzzy Lounge 10 2006-09-28 00:36
Perpetual ECM factoring challenge thread... Xyzzy Factoring 65 2005-09-05 08:16

All times are UTC. The time now is 16:58.

Thu Feb 25 16:58:45 UTC 2021 up 84 days, 13:10, 0 users, load averages: 2.01, 1.89, 1.88

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.