mersenneforum.org  

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

Reply
 
Thread Tools
Old 2005-02-19, 17:30   #166
patrik
 
patrik's Avatar
 
"Patrik Johansson"
Aug 2002
Uppsala, Sweden

52×17 Posts
Default Prescotts almost matching Northwoods?

Following the discussion in this thread (about running one or two instances of Prime95) I did some benchmarks. The interesting ones are from my new 3.2 GHz Prescott. I also downclocked an overclocked Northwood 3.0 GHz to about 3.2 GHz for easy comparison.

The very interensting result is that for a hyperthreaded Prescott you get an overall performance gain by running two instances of Prime95, even when doing two Lucas-Lehmer tests. Furthermore, the benchmark then almost matches the Northwood.

The benchmarks below are the best times for three or four benchmarks.

1. Benchmark while DoubleCheck=13201703,64,1 running in the background.

Intel(R) Pentium(R) 4 CPU 3.20GHz
CPU speed: 3206.73 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 23.9, RdtscTiming=1
Best time for 384K FFT length: 22.869 ms.
Best time for 448K FFT length: 27.621 ms.
Best time for 512K FFT length: 30.467 ms.
Best time for 640K FFT length: 37.389 ms.
Best time for 768K FFT length: 45.457 ms.
Best time for 896K FFT length: 54.429 ms.
Best time for 1024K FFT length: 60.250 ms.
Best time for 1280K FFT length: 79.546 ms.
Best time for 1536K FFT length: 97.146 ms.
Best time for 1792K FFT length: 117.733 ms.
Best time for 2048K FFT length: 131.689 ms.
Best time for 2560K FFT length: 166.244 ms.
Best time for 3072K FFT length: 204.503 ms.
Best time for 3584K FFT length: 247.794 ms.
Best time for 4096K FFT length: 278.716 ms.

2. Benchmark while Test=27013621,67,1 running in the background.

Intel(R) Pentium(R) 4 CPU 3.20GHz
CPU speed: 3207.83 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 23.9, RdtscTiming=1
Best time for 384K FFT length: 22.850 ms.
Best time for 448K FFT length: 27.544 ms.
Best time for 512K FFT length: 30.459 ms.
Best time for 640K FFT length: 37.607 ms.
Best time for 768K FFT length: 45.587 ms.
Best time for 896K FFT length: 54.744 ms.
Best time for 1024K FFT length: 60.321 ms.
Best time for 1280K FFT length: 80.209 ms.
Best time for 1536K FFT length: 98.357 ms.
Best time for 1792K FFT length: 118.863 ms.
Best time for 2048K FFT length: 133.361 ms.
Best time for 2560K FFT length: 168.011 ms.
Best time for 3072K FFT length: 208.208 ms.
Best time for 3584K FFT length: 253.071 ms.
Best time for 4096K FFT length: 285.369 ms.

3. Benchmark when only one mprime client running.

Intel(R) Pentium(R) 4 CPU 3.20GHz
CPU speed: 3207.83 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 23.9, RdtscTiming=1
Best time for 384K FFT length: 12.799 ms.
Best time for 448K FFT length: 15.582 ms.
Best time for 512K FFT length: 17.519 ms.
Best time for 640K FFT length: 20.947 ms.
Best time for 768K FFT length: 25.195 ms.
Best time for 896K FFT length: 30.331 ms.
Best time for 1024K FFT length: 33.974 ms.
Best time for 1280K FFT length: 45.011 ms.
Best time for 1536K FFT length: 54.056 ms.
Best time for 1792K FFT length: 65.079 ms.
Best time for 2048K FFT length: 72.909 ms.
Best time for 2560K FFT length: 93.961 ms.
Best time for 3072K FFT length: 115.239 ms.
Best time for 3584K FFT length: 138.523 ms.
Best time for 4096K FFT length: 153.291 ms.

4. Reference Northwood benchmark (no speed improvement was made from v23.5 to v23.9 AFAIK)

Intel(R) Pentium(R) 4 CPU 3.00GHz
CPU speed: 3208.66 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 23.5, RdtscTiming=1
Best time for 384K FFT length: 11.321 ms.
Best time for 448K FFT length: 13.444 ms.
Best time for 512K FFT length: 15.293 ms.
Best time for 640K FFT length: 18.250 ms.
Best time for 768K FFT length: 22.253 ms.
Best time for 896K FFT length: 26.336 ms.
Best time for 1024K FFT length: 29.525 ms.
Best time for 1280K FFT length: 38.728 ms.
Best time for 1536K FFT length: 47.640 ms.
Best time for 1792K FFT length: 56.561 ms.
Best time for 2048K FFT length: 64.160 ms.
Best time for 2560K FFT length: 81.034 ms.
Best time for 3072K FFT length: 101.501 ms.
Best time for 3584K FFT length: 123.872 ms.
Best time for 4096K FFT length: 138.102 ms.

Hardware:
Pentium 4 3.20E GHz (Prescott core) 800 MHz system bus
ASUS P4P800S-X
Kingston KVR400X64C25/256 (single channel 400 MHz DDR)

Pentium 4 3.0 GHz (Northwood core) 800 MHz system bus
ASUS P4P800-E Deluxe
2xKingston KVR400X64C25/256 (dual channel 400 MHz DDR)
overclocked to 3.2 GHz (FSB214)

Edit: OS: Linux kernel 2.4.20-8smp, Red Hat Linux release 9 (Shrike)

Last fiddled with by patrik on 2005-02-19 at 17:33
patrik is offline   Reply With Quote
Old 2005-02-20, 12:25   #167
ET_
Banned
 
ET_'s Avatar
 
"Luigi"
Aug 2002
Team Italia

10010110111102 Posts
Default

Quote:
Originally Posted by patrik
Following the discussion in this thread (about running one or two instances of Prime95) I did some benchmarks. The interesting ones are from my new 3.2 GHz Prescott. I also downclocked an overclocked Northwood 3.0 GHz to about 3.2 GHz for easy comparison.
Did you try a comparison during factoring?

I've a 3.2 GHz Prescott where I run Prime95 and LLR.
I noticed a sudden slow of LLR as soon as I switched from double-check to factoring in Prime95.

The switching was motivated by another thread, where it was said that hyperthreading was useful if factoring and ll-testing were done together.

Luigi
ET_ is offline   Reply With Quote
Old 2005-02-20, 14:19   #168
geoff
 
geoff's Avatar
 
Mar 2003
New Zealand

13·89 Posts
Default

Quote:
Originally Posted by patrik
1. Benchmark while DoubleCheck=13201703,64,1 running in the background.
If you run two instances of mprime doing different tasks or even doing the same task using different FFT sizes, then one will take more of the total processor time than the other, so I don't think it will be possible to get an accurate benchmark this way. On one machine the mprime running the benchmark task might be faster just because the LL testing task is running slower.
geoff is offline   Reply With Quote
Old 2005-02-20, 16:46   #169
Danath
 
Danath's Avatar
 
May 2003
Republic of Moldova

23·5 Posts
Default

So it would be the best to run two regular LL tests simultaneously for some time (rather than one LL test and a benchmark), and then watch the average timings for this period of time (you need to set the "time between screen outputs" to some value, say 10-30 minutes).
Danath is offline   Reply With Quote
Old 2005-02-20, 21:14   #170
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

11101110101102 Posts
Default

Quote:
Originally Posted by Danath
So it would be the best to run two regular LL tests simultaneously for some time (rather than one LL test and a benchmark).
I agree. The benchmark is not good because it returns the BEST time, where as a second LL test returns an AVERAGE time.

The danger with looking at the best time is that it can be easily skewed. If the OS swaps out the first LL test to run a thread that does some non-FPU work, then all of a sudden the benchmark gets a speed boost that will not be indicative of your average throughput in an LL test.
Prime95 is online now   Reply With Quote
Old 2005-02-20, 21:33   #171
patrik
 
patrik's Avatar
 
"Patrik Johansson"
Aug 2002
Uppsala, Sweden

1A916 Posts
Default

Quote:
Originally Posted by ET_
Did you try a comparison during factoring?

I've a 3.2 GHz Prescott where I run Prime95 and LLR.
I noticed a sudden slow of LLR as soon as I switched from double-check to factoring in Prime95.

The switching was motivated by another thread, where it was said that hyperthreading was useful if factoring and ll-testing were done together.

Luigi
No, not for the Prescott, but I run 1 factoring + 1 LL-test on my two HT-capable Northwoods. It has been known for some time to give higher throughput than running them on separate computers.

The interesting point is that for Prescott two LL tests also give higher throughput. That is not the case for my Northwoods. My guess is that this applies to all Northwoods and Prescotts. Since most GIMPSters buy Northwoods we have believed that running two LL tests in parallel was always a bad thing. Just like I did until yesterday.

My timings seem to be roughly accurate for the 768K FFT length my LL-tests are running at for the moment. Then I get 0.026 s and 0.046 s iteration times for one or two clients running. (That's why I decided to do some benchmarks.) But one or two FFT levels might of course still be a bit off.

Last fiddled with by patrik on 2005-02-20 at 21:42
patrik is offline   Reply With Quote
Old 2005-04-01, 06:05   #172
TauCeti
 
TauCeti's Avatar
 
Mar 2003
Braunschweig, Germany

E216 Posts
Default New Beta 24.11 with Athlon64 (W32/32 bit)

Mainboard ECS-nForce4-A939, DDR-400:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1809.08 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 29.930 ms.
Best time for 640K FFT length: 37.258 ms.
Best time for 768K FFT length: 45.091 ms.
Best time for 896K FFT length: 54.571 ms.
Best time for 1024K FFT length: 60.653 ms.
Best time for 1280K FFT length: 80.412 ms.
Best time for 1536K FFT length: 98.028 ms.
Best time for 1792K FFT length: 118.198 ms.
Best time for 2048K FFT length: 131.602 ms.
[Fri Apr 01 07:57:33 2005]


Results from 23.5 Version:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1809.25 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 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: 23.853 ms.
Best time for 448K FFT length: 28.699 ms.
Best time for 512K FFT length: 32.122 ms.
Best time for 640K FFT length: 40.456 ms.
Best time for 768K FFT length: 49.149 ms.
Best time for 896K FFT length: 59.255 ms.
Best time for 1024K FFT length: 66.528 ms.
Best time for 1280K FFT length: 88.512 ms.
Best time for 1536K FFT length: 106.524 ms.
Best time for 1792K FFT length: 128.484 ms.
Best time for 2048K FFT length: 144.814 ms.
TauCeti is offline   Reply With Quote
Old 2005-04-01, 17:52   #173
Jeff Gilchrist
 
Jeff Gilchrist's Avatar
 
Jun 2003
Ottawa, Canada

49516 Posts
Default

TauCeti, how does 24.6 compare as well in speed?
Jeff Gilchrist is offline   Reply With Quote
Old 2005-04-01, 18:29   #174
TauCeti
 
TauCeti's Avatar
 
Mar 2003
Braunschweig, Germany

22610 Posts
Default 23.8 / 24.6 / 24.11 on my A64-3000+ Winnie / DDR400

I do only TF on my A64, so i never used 24.6 before. If you need more/other benchmarks, just drop a line...

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1809.32 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 23.796 ms.
Best time for 448K FFT length: 28.631 ms.
Best time for 512K FFT length: 31.960 ms.
Best time for 640K FFT length: 40.412 ms.
Best time for 768K FFT length: 48.982 ms.
Best time for 896K FFT length: 58.961 ms.
Best time for 1024K FFT length: 66.275 ms.
Best time for 1280K FFT length: 88.329 ms.
Best time for 1536K FFT length: 105.977 ms.
Best time for 1792K FFT length: 128.341 ms.
Best time for 2048K FFT length: 144.325 ms.


AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1809.32 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.6, RdtscTiming=1
Best time for 512K FFT length: 32.344 ms.
Best time for 640K FFT length: 40.449 ms.
Best time for 768K FFT length: 49.196 ms.
Best time for 896K FFT length: 59.287 ms.
Best time for 1024K FFT length: 66.484 ms.
Best time for 1280K FFT length: 88.567 ms.
Best time for 1536K FFT length: 106.343 ms.
Best time for 1792K FFT length: 128.518 ms.
Best time for 2048K FFT length: 144.893 ms.
[Fri Apr 01 20:05:29 2005]

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1809.14 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.6, RdtscTiming=1
Best time for 512K FFT length: 33.003 ms.
Best time for 640K FFT length: 44.065 ms.
Best time for 768K FFT length: 53.294 ms.
Best time for 896K FFT length: 64.741 ms.
Best time for 1024K FFT length: 72.098 ms.
Best time for 1280K FFT length: 94.982 ms.
Best time for 1536K FFT length: 113.497 ms.
Best time for 1792K FFT length: 136.471 ms.
Best time for 2048K FFT length: 151.474 ms.

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1808.97 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 29.906 ms.
Best time for 640K FFT length: 37.193 ms.
Best time for 768K FFT length: 45.105 ms.
Best time for 896K FFT length: 54.600 ms.
Best time for 1024K FFT length: 60.659 ms.
Best time for 1280K FFT length: 80.405 ms.
Best time for 1536K FFT length: 98.117 ms.
Best time for 1792K FFT length: 118.071 ms.
Best time for 2048K FFT length: 131.648 ms.
TauCeti is offline   Reply With Quote
Old 2005-04-04, 20:13   #175
Dresdenboy
 
Dresdenboy's Avatar
 
Apr 2003
Berlin, Germany

192 Posts
Default 23.8 / 24.6 / 24.11 (32/64bit) on my A64-3000+ (orig. Sledgehammer core)/DDR400CL2.5

I have an A64 3000+ (2Ghz, 512kB L2, single channel RAM) with C0 stepping (unfortunately not the new E0-Stepping - several percent faster than 130nm steppings thanks to the core improvements).

Now the numbers:

Prime95 version 23.8

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 2002.45 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 23.219 ms.
Best time for 448K FFT length: 27.941 ms.
Best time for 512K FFT length: 31.232 ms.
Best time for 640K FFT length: 39.588 ms.
Best time for 768K FFT length: 48.090 ms.
Best time for 896K FFT length: 58.028 ms.
Best time for 1024K FFT length: 65.605 ms.
Best time for 1280K FFT length: 87.466 ms.
Best time for 1536K FFT length: 104.623 ms.
Best time for 1792K FFT length: 127.425 ms.
Best time for 2048K FFT length: 143.992 ms.



Prime95 version 24.6:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 2002.55 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.6, RdtscTiming=1
Best time for 512K FFT length: 31.568 ms.
Best time for 640K FFT length: 39.566 ms.
Best time for 768K FFT length: 48.178 ms.
Best time for 896K FFT length: 58.333 ms.
Best time for 1024K FFT length: 65.933 ms.
Best time for 1280K FFT length: 87.748 ms.
Best time for 1536K FFT length: 105.167 ms.
Best time for 1792K FFT length: 127.511 ms.
Best time for 2048K FFT length: 144.549 ms.


Prime95 version 24.11 (32 bit):

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 2002.89 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.314 ms.
Best time for 640K FFT length: 33.947 ms.
Best time for 768K FFT length: 40.932 ms.
Best time for 896K FFT length: 49.756 ms.
Best time for 1024K FFT length: 55.855 ms.
Best time for 1280K FFT length: 73.371 ms.
Best time for 1536K FFT length: 90.066 ms.
Best time for 1792K FFT length: 108.263 ms.
Best time for 2048K FFT length: 121.138 ms.
Best time for 58 bit trial factors: 6.010 ms.
Best time for 59 bit trial factors: 6.032 ms.
Best time for 60 bit trial factors: 6.003 ms.
Best time for 61 bit trial factors: 6.029 ms.
Best time for 62 bit trial factors: 10.961 ms.
Best time for 63 bit trial factors: 10.963 ms.
Best time for 64 bit trial factors: 13.932 ms.
Best time for 65 bit trial factors: 13.836 ms.
Best time for 66 bit trial factors: 13.849 ms.
Best time for 67 bit trial factors: 13.831 ms.


Prime95 version 24.11 (32 bit - no prefetching):

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 2002.59 MHz
CPU features: RDTSC, CMOV, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.202 ms.
Best time for 640K FFT length: 33.791 ms.
Best time for 768K FFT length: 40.883 ms.
Best time for 896K FFT length: 49.690 ms.
Best time for 1024K FFT length: 55.278 ms.
Best time for 1280K FFT length: 73.254 ms.
Best time for 1536K FFT length: 89.828 ms.
Best time for 1792K FFT length: 107.797 ms.
Best time for 2048K FFT length: 120.921 ms.
Best time for 58 bit trial factors: 6.025 ms.
Best time for 59 bit trial factors: 6.031 ms.
Best time for 60 bit trial factors: 6.022 ms.
Best time for 61 bit trial factors: 6.034 ms.
Best time for 62 bit trial factors: 10.988 ms.
Best time for 63 bit trial factors: 11.008 ms.
Best time for 64 bit trial factors: 13.936 ms.
Best time for 65 bit trial factors: 13.850 ms.
Best time for 66 bit trial factors: 13.884 ms.
Best time for 67 bit trial factors: 13.805 ms.


Prime95 version 24.11 (64 bit):

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 2002.99 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 28.344 ms.
Best time for 640K FFT length: 34.008 ms.
Best time for 768K FFT length: 41.064 ms.
Best time for 896K FFT length: 49.705 ms.
Best time for 1024K FFT length: 55.251 ms.
Best time for 1280K FFT length: 73.524 ms.
Best time for 1536K FFT length: 90.159 ms.
Best time for 1792K FFT length: 108.715 ms.
Best time for 2048K FFT length: 121.390 ms.
Best time for 58 bit trial factors: 3.786 ms.
Best time for 59 bit trial factors: 3.782 ms.
Best time for 60 bit trial factors: 4.032 ms.
Best time for 61 bit trial factors: 4.140 ms.
Best time for 62 bit trial factors: 4.843 ms.
Best time for 63 bit trial factors: 5.643 ms.
Best time for 64 bit trial factors: 6.816 ms.
Best time for 65 bit trial factors: 8.007 ms.
Best time for 66 bit trial factors: 7.951 ms.
Best time for 67 bit trial factors: 7.934 ms.

Last fiddled with by Dresdenboy on 2005-04-04 at 20:22
Dresdenboy is offline   Reply With Quote
Old 2005-04-04, 20:29   #176
Dresdenboy
 
Dresdenboy's Avatar
 
Apr 2003
Berlin, Germany

5518 Posts
Default

And now the results at nearly the same clock as TauCeti's results (so it's a single channel C0 stepping vs. dual channel D0 stepping comparison):

Prime95 version 23.8:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1802.17 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 25.667 ms.
Best time for 448K FFT length: 30.738 ms.
Best time for 512K FFT length: 34.357 ms.
Best time for 640K FFT length: 43.645 ms.
Best time for 768K FFT length: 53.144 ms.
Best time for 896K FFT length: 64.121 ms.
Best time for 1024K FFT length: 72.369 ms.
Best time for 1280K FFT length: 96.437 ms.
Best time for 1536K FFT length: 115.376 ms.
Best time for 1792K FFT length: 140.484 ms.
Best time for 2048K FFT length: 158.688 ms.


Prime95 version 24.6:

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1802.60 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.6, RdtscTiming=1
Best time for 512K FFT length: 34.759 ms.
Best time for 640K FFT length: 43.714 ms.
Best time for 768K FFT length: 53.068 ms.
Best time for 896K FFT length: 64.443 ms.
Best time for 1024K FFT length: 72.695 ms.
Best time for 1280K FFT length: 97.047 ms.
Best time for 1536K FFT length: 115.727 ms.
Best time for 1792K FFT length: 140.678 ms.
Best time for 2048K FFT length: 159.164 ms.


Prime95 version 24.11 (32 bit):

AMD Athlon(tm) 64 Processor 3000+
CPU speed: 1802.18 MHz
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 version 24.11, RdtscTiming=1
Best time for 512K FFT length: 30.952 ms.
Best time for 640K FFT length: 37.624 ms.
Best time for 768K FFT length: 45.487 ms.
Best time for 896K FFT length: 55.276 ms.
Best time for 1024K FFT length: 61.360 ms.
Best time for 1280K FFT length: 81.566 ms.
Best time for 1536K FFT length: 99.931 ms.
Best time for 1792K FFT length: 119.917 ms.
Best time for 2048K FFT length: 134.330 ms.


(I didn't repeat the 64bit test at 1.8 GHz and also removed the TF results)
Dresdenboy is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Perpetual "interesting video" thread... Xyzzy Lounge 43 2021-07-17 00:00
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 19:40.


Wed Oct 27 19:40:24 UTC 2021 up 96 days, 14:09, 0 users, load averages: 2.25, 1.49, 1.41

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.