Prime95 version 29.6/29.7/29.8

[Cruelty]

Originally Posted by Prime95

Are there any Windows system tools that tell you if a process' threads are bound to specific logical CPUs?

I use:

start /AFFINITY [n] [.exe]

[harlee]

Originally Posted by Madpoo

Fixed. The latest builds have the known-factors in an array in the JSON data... not sure how the previous P95 builds would have included the known-factors if there were more than one, maybe just as a comma-delimited single value.

Thanks. Just checked and noticed that all of the older PRP results are now showing the known factors.

[newalex]

I started to use Prime95 v29.6 on machine with AVX512F for ECM work yesterday.

Today it found first factor for https://www.mersenne.org/report_expo...=350249&full=1
But result is displayed weird in my account and history of exponent.
It tells only Factor: without anything else.

[lycorn]

And it doesn't show up in the Recent Cleared report, either. The corresponding PRP-C test is there, tho.

[Prime95]

Originally Posted by Cruelty

"Small FFTs" 12h stress test passed - I'd suggest lowering Max FFT size from 586K to maybe 128K,

Should be fixed in build 7.

No progress on your affinity issue -- no idea on how to proceed. Cannot reproduce the problem on my simple 4-core machine.

[kriesel]

Following is from prime95 v29.6b6, x64, on Win10 Pro, dual e5-2690. Benchmark output is truncated in the worker window, for high worker count on high core count systems. This has been reported also for previous prime95 versions. Scroll to far right for 16-worker line; of "throughput xxxx.xx iter/sec", only "through" appears.

[Mar 8 09:10] Worker starting
[Mar 8 09:10] Your timings will be written to the results.txt file.
[Mar 8 09:10] Compare your results to other computers at http://www.mersenne.org/report_benchmarks
[Mar 8 09:10] Benchmarking multiple workers to measure the impact of memory bandwidth
[Mar 8 09:10] Timing 1024K FFT, 16 cores, 1 worker.  Average times:  0.72 ms.  Total throughput: 1382.83 iter/sec.
[Mar 8 09:11] Timing 1024K FFT, 16 cores, 2 workers.  Average times:  1.01,  1.01 ms.  Total throughput: 1982.69 iter/sec.
[Mar 8 09:11] Timing 1024K FFT, 16 cores, 4 workers.  Average times:  1.96,  1.98,  1.95,  1.96 ms.  Total throughput: 2036.00 iter/sec.
[Mar 8 09:11] Timing 1024K FFT, 16 cores, 8 workers.  Average times:  4.54,  4.54,  4.54,  4.54,  4.13,  4.11,  4.11,  4.13 ms.  Total throughput: 1851.38 iter/sec.
[Mar 8 09:11] Timing 1024K FFT, 16 cores, 16 workers.  Average times:  9.80,  9.80,  9.82,  9.88,  9.86,  9.81,  9.84,  9.79,  8.48,  8.42,  8.55,  8.41,  8.41,  8.41,  8.42,  8.48 ms.  Total through
[Mar 8 09:12] Timing 1024K FFT, 16 cores hyperthreaded, 1 worker.  Average times:  1.00 ms.  Total throughput: 995.87 iter/sec.
[Mar 8 09:12] Timing 1024K FFT, 16 cores hyperthreaded, 2 workers.  Average times:  1.12,  1.11 ms.  Total throughput: 1798.63 iter/sec.
(etc)

[Cruelty]

Originally Posted by Prime95

Should be fixed in build 7.

OK, I'll give it a try.

[newalex]

The second ECM factor was found using prime 29.6. And it isn't displayed correctly in reports, too.

https://www.mersenne.org/report_expo...=351457&full=1

[pepi37]

Benchmark type (0 = Throughput, 1 = FFT timings, 2 = Trial factoring) (0): 1

FFTs to benchmark
Minimum FFT size (in K) (256):
Maximum FFT size (in K) (512):
Benchmark with round-off checking enabled (Y): n
Benchmark all-complex FFTs (for LLR,PFGW,PRP users) (N): y
Limit FFT sizes (mimic older benchmarking code) (N): y

CPU cores to benchmark
Number of CPU cores (comma separated list of ranges) (4):

Accept the answers above? (Y):
Main Menu

1. Test/Primenet
2. Test/Worker threads
3. Test/Status
4. Test/Continue
5. Test/Exit
6. Advanced/Test
7. Advanced/Time
8. Advanced/P-1
9. Advanced/ECM
10. Advanced/Manual Communication
11. Advanced/Unreserve Exponent
12. Advanced/Quit Gimps
13. Options/CPU
14. Options/Preferences
15. Options/Torture Test
16. Options/Benchmark
17. Help/About
18. Help/About PrimeNet Server
[Main thread Mar 10 18:29:53] Starting worker.
Your choice: [Work thread Mar 10 18:29:53] Worker starting
[Work thread Mar 10 18:29:53] Your timings will be written to the results.txt file.
[Work thread Mar 10 18:29:53] Compare your results to other computers at http://www.mersenne.org/report_benchmarks
[Work thread Mar 10 18:29:53] Timing FFTs using 4 cores.
[Work thread Mar 10 18:29:53] Timing 100 iterations of 256K all-complex FFT length. Best time: 0.000 sec., avg time: 0.000 sec.
[Work thread Mar 10 18:29:53] Timing 100 iterations of 320K all-complex FFT length. Best time: 0.000 sec., avg time: 0.000 sec.
[Work thread Mar 10 18:29:53] Timing 100 iterations of 384K all-complex FFT length. Best time: 0.000 sec., avg time: 0.000 sec.
[Work thread Mar 10 18:29:53] Timing 86 iterations of 512K all-complex FFT length. Best time: 0.001 sec., avg time: 0.001 sec.
[Work thread Mar 10 18:29:53] FFT timings benchmark complete.
[Work thread Mar 10 18:29:53] Worker stopped.

Linux x64 , Prime95 29.6 beta3
So I put FFT timings from 256K to 512K on 4 cores

[kriesel]

1) Prime95 v29.6b6 x64 on dual 8-core E5-2690, Win 10 Pro; throughput benchmark 1,2,4,8,16 workers, HT and not, 1M-32M, completed without incident;

2) Prime95 v29.6b7 x64 on 6-core i7-8750H, Win 10 Home; throughput benchmark 1-3,6 workers, HT and not, 1M-32M, completed without incident;

3) Prime95 v29.6b7 x64 on Core 2 Duo E8200, Vista; throughput benchmark 1-2 workers, 1M-32M, completed without incident;

4) Prime95 v29.6b7 x64 on dual 6-core Xeon X5650, Win 7 Pro; throughput benchmark 1,2,3,4,6,12 workers, HT and not, 1M-32M, completed without incident;

5) Prime95 v29.6b3 x32 on Pentium 750 M, Vista; throughput benchmark 1M-32M, completed without incident.

[pepi37]

So I put FFT timings from 256K to 512K on 4 cores, not throughput benchmark