mersenneforum.org  

Go Back   mersenneforum.org > Search Forums

Showing results 1 to 25 of 1000
Search took 0.15 seconds.
Search: Posts Made By: preda
Forum: GpuOwl 2020-10-18, 23:15
Replies: 108
Views: 3,616
Posted By preda
I don't know, maybe only works for PRP. Anyway I...

I don't know, maybe only works for PRP. Anyway I consider that (not rejecting factors at the end) a bug.
Forum: GpuOwl 2020-10-18, 22:49
Replies: 108
Views: 3,616
Posted By preda
I think any v6.x accepts factors at the end of...

I think any v6.x accepts factors at the end of the assignment, and completely *ignores* them. Does not read them, does not handle them in any way, as if they weren't there. Does not continue P-1 past...
Forum: GpuOwl 2020-10-18, 21:49
Replies: 108
Views: 3,616
Posted By preda
No plans for PRP-CF. I don't have a good...

No plans for PRP-CF. I don't have a good understanding of that need myself, and I think that need is adequately covered by mprime.

Could you explain a bit more about what you're trying to do by...
Forum: GpuOwl 2020-10-18, 10:01
Replies: 108
Views: 3,616
Posted By preda
Thanks! I pushed a fix, should be fine now. The...

Thanks! I pushed a fix, should be fine now. The problem was affecting P2 where the number of buffers was > 1024, which was the case for this small exponent.
Forum: Number Theory Discussion Group 2020-10-17, 05:11
Replies: 3
Views: 115
Posted By preda
Apparently the complexity of jocobi symbol is the...

Apparently the complexity of jocobi symbol is the same as the GCD. For 100M-bits residues, Gnu-MP takes about 30s on CPU. (which is abouth the same time as a GCD).
Forum: GpuOwl 2020-10-16, 12:42
Replies: 108
Views: 3,616
Posted By preda
P2 hardening

Let me start with a brief description of how P2 works:

There is a large set of precomputed buffers, as many as the GPU memory allows (a 16GB GPU at the vawefront accommodates a bit over 300...
Forum: GpuOwl 2020-10-16, 06:59
Replies: 108
Views: 3,616
Posted By preda
P2 ETA added. Should be quite accurate, and P2...

P2 ETA added. Should be quite accurate, and P2 progress % is v. accurate too.
Forum: GpuOwl 2020-10-16, 06:56
Replies: 108
Views: 3,616
Posted By preda
Should be possible to port it over, yes. Simply...

Should be possible to port it over, yes. Simply take the output of stage1 and plug it into the new stage2. Of course things are a bit hairy, so that would still require some significant work I...
Forum: GpuOwl 2020-10-16, 04:52
Replies: 108
Views: 3,616
Posted By preda
P2 changes

A new prime-pairing algorithm for P2 has been implemented, P2 is now (significantly) faster.

This required a bump of the P2 savefile version, so don't switch version in the middle of P2, wait till...
Forum: GpuOwl 2020-10-16, 04:48
Replies: 108
Views: 3,616
Posted By preda
Some of these should be fixed now: - consistent...

Some of these should be fixed now:
- consistent % progress across restart.

Ctrl-C during P2 has this behavior:
- first Ctrl-C triggers GCD, which is followed by a save and exit 30s later
-...
Forum: GpuOwl 2020-10-14, 11:36
Replies: 108
Views: 3,616
Posted By preda
Worktype for PRP + P-1 in primenet.py : PRP_P1

The script gpuowl/tools/primenet.py has a new work-type for "PRP that needs P-1", which is what you want for the merged PRP + P-1; it's called "PRP_P1":

primenet.py -w PRP_P1 etc.

If you run it...
Forum: GpuOwl 2020-10-14, 11:28
Replies: 108
Views: 3,616
Posted By preda
Thanks. I don't know yet the reason for this....

Thanks. I don't know yet the reason for this. Also I can't seem to reproduce.
Does that GPU produce any errors during the PRP? (EE) (i.e. is the PRP 100% reliable, or sometimes there are errors and...
Forum: GpuOwl 2020-10-14, 01:38
Replies: 108
Views: 3,616
Posted By preda
New P2

A new implementation of P2 has been merged, which has two main changes in behavior:

1. it uses all the memory you can throw at it (up to -maxAlloc, of course) to speed up,
2. is significantly...
Forum: GpuOwl 2020-10-14, 01:33
Replies: 108
Views: 3,616
Posted By preda
I don't think this concerns the format change to...

I don't think this concerns the format change to -maxAlloc, which didn't change much and works in the same way as in v6.x.

The problem I see is that, even when the software (GpuOwl) stays nicely...
Forum: Hardware 2020-10-13, 00:51
Replies: 7
Views: 160
Posted By preda
Evil lair located in UK? In my case: I'm...

Evil lair located in UK?

In my case: I'm already prime member so the marginal cost of that is zero, but otherwise it's a few bucks per month (again, it depends). May even be free for a trial. And...
Forum: Hardware 2020-10-13, 00:25
Replies: 7
Views: 160
Posted By preda
I don't know the price as it depends on which...

I don't know the price as it depends on which amazon storefront you buy it from (US, AU, etc).

But doing the straight currency conversion, it comes to 240 USD to me. It's also 300 AUD in AU. And...
Forum: Hardware 2020-10-13, 00:06
Replies: 7
Views: 160
Posted By preda
Targeted to all those complaining "I don't have...

Targeted to all those complaining "I don't have 10GB free on my SSD to store all those proofs", 14TB should take care of that space problem.
Forum: Hardware 2020-10-13, 00:04
Replies: 7
Views: 160
Posted By preda
Cheap 14TB 3.5'' HDD

Amazon.uk has a rather good deal for the 14TB HDD, (prime day deal):

https://www.amazon.co.uk/WD-Elements-Desktop-External-Drive/dp/B07Y3KDVZH

The HDD is shuckable (can be extracted and used...
Forum: GpuOwl 2020-10-12, 07:04
Replies: 108
Views: 3,616
Posted By preda
And a 116bits factor: ?...

And a 116bits factor:

? factor(88196068168080719419045580163465679 - 1)
%28 =
[ 2 1]
[ 3 1]
[ 3119 1]
[ 3257 1]
[ 106781 1]
[ 2006659 1]
Forum: GpuOwl 2020-10-11, 22:30
Replies: 108
Views: 3,616
Posted By preda
One thing to check is whether the GPU memory is...

One thing to check is whether the GPU memory is exausted (this results in a terrible slowdown of the GPU).

On smaller exponents, a "residue" is smaller. It appears that for each buffer allocated...
Forum: GpuOwl 2020-10-11, 08:28
Replies: 20
Views: 1,404
Posted By preda
Laur, you'll be pleased (I hope) to find that in...

Laur, you'll be pleased (I hope) to find that in the latest GpuOwl 7.x I switched to using savefiles with explicit iteration numbers in filenames. They still don't have the res64 as part of the...
Forum: GpuOwl 2020-10-11, 08:20
Replies: 108
Views: 3,616
Posted By preda
I've been blessed with a flaky GPU

My broken GPU generates about 20 errors per PRP test. Those are handled just fine by the error check for PRP, but the situation is different for P-1.

I tried to harden the P1 (first stage). I...
Forum: GpuOwl 2020-10-10, 09:13
Replies: 108
Views: 3,616
Posted By preda
The most intensive CPU-side operation is the GCD....

The most intensive CPU-side operation is the GCD. After that, building the P2 plan (once at the start of P2) can also use more CPU. After that, maybe the P1 CPU use increased a (tiny) bit. Let's keep...
Forum: GpuOwl 2020-10-08, 21:43
Replies: 108
Views: 3,616
Posted By preda
And a factor: ?...

And a factor:

? factor(183358639670121039844487732119-1)
%14 =
[ 2 1]
[ 3 1]
[ 7 1]
[ 11 1]
[ 739 1]
[ 1301 1]
Forum: GpuOwl 2020-10-08, 19:07
Replies: 108
Views: 3,616
Posted By preda
Hi Ken, could you please check, is there a...

Hi Ken, could you please check, is there a genuine permission issue preventing the file rename?

PS: there may have been an issue with the worktodo.txt rename affecting some windows. I commited an...
Showing results 1 to 25 of 1000

 
All times are UTC. The time now is 02:58.

Wed Oct 21 02:58:10 UTC 2020 up 41 days, 9 mins, 0 users, load averages: 1.11, 1.59, 1.63

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, 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.