PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
1) Message boards : Aggie The Pew message board (Message 123461)
Posted 3 days ago by Iain BethuneProject donor
No idea about the rat-master, but I'm still lurking around too...
2) Message boards : News : Genefer 3.3.4 now live (Message 122551)
Posted 29 days ago by Iain BethuneProject donor
The apps were kept in sync across all of the sub-projects, but the performance improvement and multithreading only works for n=21.

- Iain
3) Message boards : Aggie The Pew message board (Message 121111)
Posted 61 days ago by Iain BethuneProject donor
I spotted that there is a chap called Deano Montreuil from Suwanee, GA entered in this weekend's "Big Dog's Backyard Ultra". I wonder if it's the very same Deano who is a member of AtP?
4) Message boards : Number crunching : Source code repositories (Message 121016)
Posted 66 days ago by Iain BethuneProject donor
Hi,

The latest version of the gcwsieve source code is available here: https://github.com/ibethune/gcwsieve

Cheers

- Iain
5) Message boards : Generalized Fermat Prime Search : High CPU usage again: Genefer 21 3.19 GPU (OCLcudaGFN) (Message 120956)
Posted 70 days ago by Iain BethuneProject donor
I have confirmed the LD_PRELOAD hack still fixes it. This seems to affect all my machines (but I set up all of my machines very similarly, they're all running close to the same versions of everything.)


Can you let me know specifically what libraries you linked via LD_PRELOAD to resolve this?

Cheers

- Iain
6) Message boards : Aggie The Pew message board (Message 120955)
Posted 70 days ago by Iain BethuneProject donor
I also want to go to the Lagavulin distillery. Iain will give me crap about that. Got to love that smoky whisky.


What a waste of a trip ;)
7) Message boards : Generalized Fermat Prime Search : Comprehensive GFN prime list for n=15 and above (Message 120889)
Posted 73 days ago by Iain BethuneProject donor

Yes but fortunately the error is a typo on the page http://www.primegrid.com/gfn_history.php#n15
As of 27th July 2018, there are 1114 known primes of the form b^2^15+1 = b^32768+1.
is not correct, there are 1014 known primes.


Sorry, will fix that shortly!
8) Message boards : News : Genefer 3.3.4 now live (Message 120797)
Posted 79 days ago by Iain BethuneProject donor
The latest release of genefer (3.3.4) is now released on BOINC labelled as app version 3.20. The client will download the new app automatically, no action is required on your part unless you are using app_info.xml (if you don't know what this means, don't worry).

The new version provides a speed-up of around 10% for AVX and FMA3 transforms for n=21, as well as providing support for multithreading. To enable it, create an app_config.xml file in your BOINC/projects/www.primegrid.com directory containing (for 2 threads per task)

<app_config> <app_version> <app_name>genefer</app_name> <cmdline>-nt 2</cmdline> <avg_ncpus>2</avg_ncpus> <plan_class>cpuGFN21</plan_class> </app_version> </app_config>


Any problems or questions, let us know! Please note that as the new release does not affect the OCL code, the GPU app versions remain unchanged.
9) Message boards : Generalized Fermat Prime Search : genefer 3.3.4 (Message 120796)
Posted 79 days ago by Iain BethuneProject donor
Thanks everyone for helping out with testing! Manual credit has now been applied to your account!

Cheers

- Iain
10) Message boards : Generalized Fermat Prime Search : genefer 3.3.4 (Message 120509)
Posted 92 days ago by Iain BethuneProject donor
The spreadsheet tells me it should complete with SSE2, but it MaxErr and falls back to x87. It still reaches the correct result, but I figure I should mention it anyway.


Yes, I also observed this with the 32-bit build. If I remember correctly, some transforms have a slightly different implementation (there are some x86_64-specific optimisiations) that also happens to make them more accurate i.e. higher b limit. As a result, the 64-bit test runs fine, and the 32-bit drops back and forward between x87 and the faster transforms. Nothing to worry about.

- Iain


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2018 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.65, 0.73, 0.79
Generated 16 Dec 2018 | 8:42:05 UTC