Perpetual benchmark thread...

[Jurzal]

AMD 5900X

Timings for 2048K FFT length (12 cores, 2 workers): 1.13, 1.21 ms. Throughput: 1710.28 iter/sec.
Timings for 2240K FFT length (12 cores, 2 workers): 1.22, 1.28 ms. Throughput: 1599.99 iter/sec.
Timings for 2304K FFT length (12 cores, 2 workers): 1.23, 1.31 ms. Throughput: 1580.18 iter/sec.
Timings for 2400K FFT length (12 cores, 2 workers): 1.29, 1.39 ms. Throughput: 1497.56 iter/sec.
Timings for 2560K FFT length (12 cores, 2 workers): 1.45, 1.48 ms. Throughput: 1365.51 iter/sec.
Timings for 2688K FFT length (12 cores, 2 workers): 1.48, 1.54 ms. Throughput: 1326.29 iter/sec.
Timings for 2800K FFT length (12 cores, 2 workers): 1.63, 1.73 ms. Throughput: 1194.21 iter/sec.
Timings for 2880K FFT length (12 cores, 2 workers): 1.57, 1.64 ms. Throughput: 1248.87 iter/sec.
Timings for 3072K FFT length (12 cores, 2 workers): 1.68, 1.70 ms. Throughput: 1183.92 iter/sec.
Timings for 3200K FFT length (12 cores, 2 workers): 1.79, 1.83 ms. Throughput: 1105.26 iter/sec.
Timings for 3360K FFT length (12 cores, 2 workers): 1.90, 1.94 ms. Throughput: 1041.47 iter/sec.
Timings for 3584K FFT length (12 cores, 2 workers): 2.03, 2.05 ms. Throughput: 981.40 iter/sec.
Timings for 3840K FFT length (12 cores, 2 workers): 2.21, 2.22 ms. Throughput: 904.69 iter/sec.
Timings for 4096K FFT length (12 cores, 2 workers): 2.35, 2.32 ms. Throughput: 856.28 iter/sec.
Timings for 4480K FFT length (12 cores, 2 workers): 3.02, 2.87 ms. Throughput: 679.31 iter/sec.
Timings for 4608K FFT length (12 cores, 2 workers): 2.77, 2.74 ms. Throughput: 726.38 iter/sec.
Timings for 4800K FFT length (12 cores, 2 workers): 2.91, 2.83 ms. Throughput: 696.74 iter/sec.
Timings for 5120K FFT length (12 cores, 2 workers): 3.34, 3.25 ms. Throughput: 607.15 iter/sec.
Timings for 5376K FFT length (12 cores, 2 workers): 3.80, 3.58 ms. Throughput: 542.10 iter/sec.
Timings for 5600K FFT length (12 cores, 2 workers): 4.37, 4.05 ms. Throughput: 476.00 iter/sec.
Timings for 5760K FFT length (12 cores, 2 workers): 5.61, 4.90 ms. Throughput: 382.12 iter/sec.
Timings for 6144K FFT length (12 cores, 2 workers): 5.04, 4.72 ms. Throughput: 410.16 iter/sec.
Timings for 6400K FFT length (12 cores, 2 workers): 5.52, 5.15 ms. Throughput: 375.20 iter/sec.
Timings for 6720K FFT length (12 cores, 2 workers): 6.36, 5.88 ms. Throughput: 327.19 iter/sec.
Timings for 7168K FFT length (12 cores, 2 workers): 6.87, 6.41 ms. Throughput: 301.54 iter/sec.
Timings for 7680K FFT length (12 cores, 2 workers): 9.50, 8.92 ms. Throughput: 217.36 iter/sec.
Timings for 8000K FFT length (12 cores, 2 workers): 8.34, 7.70 ms. Throughput: 249.84 iter/sec.
Timings for 8064K FFT length (12 cores, 2 workers): 8.38, 7.89 ms. Throughput: 246.15 iter/sec.
Timings for 8192K FFT length (12 cores, 2 workers): 8.71, 8.27 ms. Throughput: 235.79 iter/sec.

[Jurzal]

Hi, what's up with the benchmark table in the server? It just shows empty logs.
Am I doing something wrong?

[aperson1]

Originally Posted by Jurzal

Hi, what's up with the benchmark table in the server? It just shows empty logs.
Am I doing something wrong?

I'm encountering a similar error with my AMD Ryzen 9 7900X 12-core processor. I feel the answer is that I'm doing something comically wrong...

[Jurzal]

Originally Posted by aperson1

I'm encountering a similar error with my AMD Ryzen 9 7900X 12-core processor. I feel the answer is that I'm doing something comically wrong...

Do you mind sharing 2 worker report here in the thread, would be interesting to compare 5900X to 7900X ^^

[aperson1]

Originally Posted by Jurzal

Do you mind sharing 2 worker report here in the thread, would be interesting to compare 5900X to 7900X ^^

Sure, here they are:

Timings for 2048K FFT length (12 cores, 2 workers): 0.70, 0.76 ms. Throughput: 2741.08 iter/sec.
Timings for 2100K FFT length (12 cores, 2 workers): 0.79, 0.79 ms. Throughput: 2536.56 iter/sec.
Timings for 2160K FFT length (12 cores, 2 workers): 0.77, 0.80 ms. Throughput: 2548.07 iter/sec.
Timings for 2240K FFT length (12 cores, 2 workers): 0.83, 0.89 ms. Throughput: 2326.03 iter/sec.
Timings for 2304K FFT length (12 cores, 2 workers): 0.82, 0.90 ms. Throughput: 2319.11 iter/sec.
Timings for 2400K FFT length (12 cores, 2 workers): 0.86, 0.93 ms. Throughput: 2235.55 iter/sec.
Timings for 2520K FFT length (12 cores, 2 workers): 0.91, 0.95 ms. Throughput: 2147.25 iter/sec.
Timings for 2560K FFT length (12 cores, 2 workers): 0.94, 0.91 ms. Throughput: 2158.68 iter/sec.
Timings for 2592K FFT length (12 cores, 2 workers): 0.99, 0.95 ms. Throughput: 2064.67 iter/sec.
Timings for 2688K FFT length (12 cores, 2 workers): 1.01, 0.96 ms. Throughput: 2040.48 iter/sec.
Timings for 2880K FFT length (12 cores, 2 workers): 1.08, 1.10 ms. Throughput: 1828.71 iter/sec.
Timings for 2940K FFT length (12 cores, 2 workers): 1.14, 1.13 ms. Throughput: 1761.08 iter/sec.
Timings for 3000K FFT length (12 cores, 2 workers): 1.15, 1.15 ms. Throughput: 1745.86 iter/sec.
Timings for 3072K FFT length (12 cores, 2 workers): 1.09, 1.10 ms. Throughput: 1823.41 iter/sec.
Timings for 3136K FFT length (12 cores, 2 workers): 1.23, 1.29 ms. Throughput: 1590.74 iter/sec.
Timings for 3200K FFT length (12 cores, 2 workers): 1.21, 1.20 ms. Throughput: 1659.60 iter/sec.
Timings for 3360K FFT length (12 cores, 2 workers): 1.30, 1.30 ms. Throughput: 1537.20 iter/sec.
Timings for 3456K FFT length (12 cores, 2 workers): 1.30, 1.30 ms. Throughput: 1535.33 iter/sec.
Timings for 3600K FFT length (12 cores, 2 workers): 1.39, 1.38 ms. Throughput: 1444.47 iter/sec.
Timings for 3840K FFT length (12 cores, 2 workers): 1.49, 1.46 ms. Throughput: 1356.25 iter/sec.
Timings for 3920K FFT length (12 cores, 2 workers): 1.62, 1.55 ms. Throughput: 1262.92 iter/sec.
Timings for 4032K FFT length (12 cores, 2 workers): 1.52, 1.53 ms. Throughput: 1309.85 iter/sec.
Timings for 4200K FFT length (12 cores, 2 workers): 1.69, 1.65 ms. Throughput: 1197.77 iter/sec.
Timings for 4320K FFT length (12 cores, 2 workers): 1.76, 1.63 ms. Throughput: 1182.38 iter/sec.
Timings for 4480K FFT length (12 cores, 2 workers): 1.85, 1.81 ms. Throughput: 1093.87 iter/sec.
Timings for 4608K FFT length (12 cores, 2 workers): 2.20, 2.03 ms. Throughput: 948.57 iter/sec.
Timings for 4704K FFT length (12 cores, 2 workers): 2.28, 2.15 ms. Throughput: 904.16 iter/sec.
Timings for 4800K FFT length (12 cores, 2 workers): 2.47, 2.28 ms. Throughput: 843.64 iter/sec.
Timings for 5040K FFT length (12 cores, 2 workers): 2.35, 2.18 ms. Throughput: 885.84 iter/sec.
Timings for 5120K FFT length (12 cores, 2 workers): 2.57, 2.37 ms. Throughput: 811.50 iter/sec.
Timings for 5184K FFT length (12 cores, 2 workers): 2.51, 2.36 ms. Throughput: 822.58 iter/sec.
Timings for 5376K FFT length (12 cores, 2 workers): 2.87, 2.67 ms. Throughput: 722.36 iter/sec.
Timings for 5760K FFT length (12 cores, 2 workers): 3.68, 3.24 ms. Throughput: 580.13 iter/sec.
Timings for 6048K FFT length (12 cores, 2 workers): 3.76, 3.48 ms. Throughput: 552.97 iter/sec.
Timings for 6144K FFT length (12 cores, 2 workers): 3.98, 3.70 ms. Throughput: 521.45 iter/sec.
Timings for 6272K FFT length (12 cores, 2 workers): 3.92, 3.64 ms. Throughput: 529.57 iter/sec.
Timings for 6400K FFT length (12 cores, 2 workers): 4.42, 4.06 ms. Throughput: 472.51 iter/sec.
Timings for 6720K FFT length (12 cores, 2 workers): 4.74, 4.36 ms. Throughput: 440.39 iter/sec.
Timings for 7056K FFT length (12 cores, 2 workers): 5.04, 4.73 ms. Throughput: 409.55 iter/sec.
Timings for 7168K FFT length (12 cores, 2 workers): 5.15, 4.82 ms. Throughput: 401.51 iter/sec.
Timings for 7200K FFT length (12 cores, 2 workers): 5.65, 5.07 ms. Throughput: 374.21 iter/sec.
Timings for 7680K FFT length (12 cores, 2 workers): 5.94, 5.84 ms. Throughput: 339.41 iter/sec.
Timings for 8064K FFT length (12 cores, 2 workers): 7.12, 6.73 ms. Throughput: 288.97 iter/sec.

[Jurzal]

Interesting, so at wave front DC 3360K 7900X is 1537.20 iter/sec, but my 5900X is 1041.47 iter/sec, which is 47.5% improvement, bigger than I thought.

[Mark Rose]

Originally Posted by Jurzal

Interesting, so at wave front DC 3360K 7900X is 1537.20 iter/sec, but my 5900X is 1041.47 iter/sec, which is 47.5% improvement, bigger than I thought.

I bet you have faster memory in your 7900X system. AVX512 in the 7900X also has more throughput.

[Jurzal]

Originally Posted by Mark Rose

I bet you have faster memory in your 7900X system. AVX512 in the 7900X also has more throughput.

Yes, 7900X should be paired with 6000-6200 MT DDR5 memory. I have 3800 MT CL16 dual rank with tuned sub timings. While running PRP on 1 worker 6M length and LL on 2nd worker at 3360K length, my system reports 24-31 Gbps read bandwidth.

[Alisha78]

Originally Posted by guido72

Motherboard is Asus P4T533-C with 512MB RD1066 Memory.

Hi! I know that curiosity killed the cat, but here in Italy is virtually impossible to find some sticks of PC1066 memory: so, where didi you find 'em? Thanks!


Daffy: I do notice this. Sometimes notice that Prime95 is running slower than it should be. No idea how it happens it just happens.

The same is happening to me as well and no ideas too :-((
regards[/QUOTE]

The ASUS P4T533-C incorporates the Intel 850E chipset with high-bandwidth 1066 MHz RDRAM support to deliver the best performance for 533MHz FSB Pentium 4 processors. A maximum 2GB of RDRAM is supported with a peak bandwidth reaching up to 4.2GB/s.

[lycorn]

I couldn´t find any compared performance figures of these two CPUs running Prime95.
Does anybody care to share some, if available?
I´m curious whether the v-cache will make any difference for GIMPS work.

[Mark Rose]

Originally Posted by lycorn

I couldn´t find any compared performance figures of these two CPUs running Prime95.
Does anybody care to share some, if available?
I´m curious whether the v-cache will make any difference for GIMPS work.

With the 5800X and 5800X3D, the extra v-cache did help running PRP/LL on larger exponents, using FFT 7680K or greater.

Memory bandwidth will also have an impact, of course.