mersenneforum.org  

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

Reply
 
Thread Tools
Old 2021-09-25, 16:19   #1
James Heinrich
 
James Heinrich's Avatar
 
"James Heinrich"
May 2004
ex-Northern Ontario

24×3×73 Posts
Default gpuowl for P-1 only

Can gpuowl be used to do only P-1? If so, how?
James Heinrich is offline   Reply With Quote
Old 2021-09-25, 16:53   #2
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

132478 Posts
Default

Yes. V6.11-380 or somewhat earlier. If you don't specify bounds somehow, it will default to B1=1M, B2=30M. Worktodo entry formats here. Speed comparisons here, and here.

Last fiddled with by kriesel on 2021-09-25 at 17:01
kriesel is online now   Reply With Quote
Old 2021-09-25, 17:31   #3
James Heinrich
 
James Heinrich's Avatar
 
"James Heinrich"
May 2004
ex-Northern Ontario

24×3×73 Posts
Default

Quote:
Originally Posted by kriesel View Post
Worktodo entry formats here
Great reference, thanks.
But it doesn't seem to work in the most recent version I found (v7.2.69)?
Using the sample worktodo entry from your link above:
Code:
U:\gpuowl>gpuowl-win
2021-09-25 13:29:23 GpuOwl VERSION v7.2-69-g23c14a1
2021-09-25 13:29:23 GpuOwl VERSION v7.2-69-g23c14a1
2021-09-25 13:29:23 config: -user JamesHeinrich -cpu RX480
2021-09-25 13:29:23 device 0, unique id ''
2021-09-25 13:29:23 RX480 worktodo.txt line ignored: "B1=790000,B2=16590000;PFactor=A125254BD75564243D4B73D4EC601234,1,2,91538501,-1,77,2"
2021-09-25 13:29:23 RX480 Bye
James Heinrich is offline   Reply With Quote
Old 2021-09-25, 18:13   #4
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

3·1,933 Posts
Default

Right. In V7.x you'd have to give bounds with a PRP worktodo line, then kill the run after P-1 stage 2 finishes. (Possibly you could use -iters if that worked. I vaguely recall trying that -iters method without success. The iters option value would be about 1.45-1.5 times the B1 bound.) And it would be slower than V6.11-380 or -364 etc. That's why I referred to V6.11-380 in post 2, and gave links to speed comparisons. Many of us are still running V6.11-380 or lower for some purposes. When I was running multiple RadeonVIIs as a P-1 factory, that was all with modified bounds prepended, on V6.11-380; sometimes ~40 wavefront P-1 results/day/GPU with GPU72 row bounds indicated at mersenne.ca.
If you reread the worktodo entry formats post carefully in the Gpuowl section, you'll see that PFactor= is not one of the accepted forms for V7.x. That's intentional, not an omission. V7.x does not support standalone PFactor, nor LL; it's combined P-1/PRP only. In some cases that's faster than V6.11-380 standalone P-1 followed by PRP, in most cases not. Also note that v7.2-69 was not the fastest 7.2 version in my experience; generally that was 7.2-53.

All that is part of why the Gpuowl intro post includes "In either case, note that the computation types, hardware supported, fft size limits, file formats, etc have varied greatly and rapidly over the course of the hundreds of versions. Choose version according to what you want to run and what each offers."

Preda never let the past get in the way of trying to make improvements. That seems to have worked well overall. And presents an occasional pitfall for the unwary.
kriesel is online now   Reply With Quote
Old 2021-09-25, 18:21   #5
James Heinrich
 
James Heinrich's Avatar
 
"James Heinrich"
May 2004
ex-Northern Ontario

DB016 Posts
Default

Seems to be working with v6.11-380, thanks.
James Heinrich is offline   Reply With Quote
Old 2021-09-25, 20:38   #6
kriesel
 
kriesel's Avatar
 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest

3×1,933 Posts
Default

Good.
One thing in favor of V7.x P-1 is superior error detection.
1) It saves powers of 3 produced during PRP/GEC, for use in generating the P-1 stage 1 power of 3. Generation of those individual powers if it errors will get flagged by a GEC.
2) It uses Jacobi check on the P-1 computations.

An extremely long P-1 run might be better on V7.2-53 for that error detection despite a small speed disadvantage.

https://mersenneforum.org/showpost.p...30&postcount=2
https://mersenneforum.org/showpost.p...2&postcount=82
Jacobi check on P-1 is something I had identified as possible by early 2019, and R. Gerbicz reviewed. It's not as productive as for LL, since unlike for the LL sequence, it's not known which value is correct, that must be computed.

Last fiddled with by kriesel on 2021-09-25 at 20:40
kriesel is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Help with setting up gpuowl asejevu GpuOwl 6 2021-09-05 16:03
GpuOwl 7.x preda GpuOwl 199 2021-03-26 00:00
GpuOwl PRP-Proof changes preda GpuOwl 20 2020-10-17 06:51
gpuOWL for Wagstaff GP2 GpuOwl 22 2020-06-13 16:57
gpuowl tuning M344587487 GpuOwl 14 2018-12-29 08:11

All times are UTC. The time now is 14:45.


Sat Oct 23 14:45:30 UTC 2021 up 92 days, 9:14, 0 users, load averages: 2.00, 1.49, 1.28

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.