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

Toggle Menu

Join PrimeGrid

Returning Participants


Leader Boards



11) Message boards : Generalized Fermat Prime Search : DO YOU FEEL LUCKY? (Message 131626)
Posted 1219 days ago by dukebg
how much faster GPU would have been counting … if my ask/favor wolud have been set

Next to none at all, I believe. Because, as dthonon wrote:

I believe that geneferocl will use as much CPU power as needed to get the GPU active. And its runs with a higher priority than llr tasks, so it is not slowed down by CPU only projects.
12) Message boards : Generalized Fermat Prime Search : DO YOU FEEL LUCKY? (Message 131620)
Posted 1219 days ago by dukebg
In BOINC it is possible to specify how many CPU cores does a GPU task take using app_config.xml. This is something the user does himself.

See here, the option is called cpu_usage
13) Message boards : Problems and Help : Suddenly "error while computing" (Message 131542)
Posted 1222 days ago by dukebg
After running PPS Sieve and AP 27 for many wus with rarely ever having an error I switched last night and started running GFN 15, 16 and 17 low.

All 28 wus failed.

My desktop is running Win 7 and has dual GeForce GTX 970 cards.

The only recent change made was an update to driver on May 5 2019.

Help please.

AT Hiker

Most likely you've installed a driver that doesn't have OpenCL libraries. You need a driver that has OpenCL libraries for GFN projects (but not PPS Sieve, it's CUDA).
14) Message boards : Problems and Help : How to only download one WU at a time? (Message 131539)
Posted 1222 days ago by dukebg

|---| |---| |---| |---| |---| |---|


|----------------------| |----------------------| |----------------------| |----------------------| |----------------------| |----------------------|

I too would like to find a way for the client to download only one task when in MT mode with zero resource share, so that each task us returned as fast as possible, and the best possibility of being first. As far as I can tell, this is not possible.

|---| |---| |---| |---| |---| |---|

Calculations on the same cores cannot physically run at the same time, they are just going to be waiting for their time on the CPU core (AVX unit or elsewhere) in a queue (CPU time scheduling + fighting for CPU cache space). So yes, this is impossible and kinda weird why would you imagine it being possible. Multithreading is using 6x calculations speed.

Here's a revision of your diagram

|666| |666| |666| |666| |666| |666|


|1111111111111111111111| |1111111111111111111111| |1111111111111111111111| |1111111111111111111111| |1111111111111111111111| |1111111111111111111111|

You cannot go more than 6 at the same time verically.
15) Message boards : AP26 - AP27 Search : 400 series Nvidia drivers slow down AP search (Message 131447)
Posted 1226 days ago by dukebg
To be more specific, it's fixed with the new apps that are live since 30 Jan 2019, they are also faster and have a cmdline flag to speed up even more
16) Message boards : Generalized Fermat Prime Search : GFN 19 seems not to continue (Message 131425)
Posted 1227 days ago by dukebg
Oh. That didn't even occur to me.

Well... my advice still remains looking at stuff with process explorer
17) Message boards : General discussion : Can Anyone Recommend A Good Book To Understand This Prime Math (Message 131424)
Posted 1227 days ago by dukebg
A good place to start might be

Not specifically about prime finding, but related math – "The Joy of Factoring" by Samuel Wagstaff

Then there's "Prime Numbers: A Computational Perspective" by Carl Pomerance and Richard Crandall

there's a lot of math textbooks and notes online:
18) Message boards : Generalized Fermat Prime Search : GFN 19 seems not to continue (Message 131421)
Posted 1227 days ago by dukebg
Multiple GFN workunits generally don't share the same GPU well for 17 and larger. They run best exclusively. Multiple running at the same time would mean some of them are running and the other are stalled

Look at what processes are runnning with the task manager or process explorer. With the last one you can see the GPU history (graph) of each process and assess what's going on.
19) Message boards : Generalized Fermat Prime Search : 271643232^131072 + 1 ? (Message 130713)
Posted 1254 days ago by dukebg
I can always sent residue to Primegrid, but 15000 candidates I processed is drop of water in number of Primegrid processed workunits :)

Just send them. If PG admins decide to use those residues or not would be their decision. Shouldn't be your decision.

As far as I'm aware, the work generator doesn't generate units for known primes (i.e. primes coming from T5K) even if they are not PG primes. Although I know that only about the LLR work generator, not sure about genefer.
20) Message boards : Generalized Fermat Prime Search : GFN-16 MEGA Prime Search starts March, 14, 2019 (Message 130664)
Posted 1255 days ago by dukebg
Hey Stream,

I dont get it, on gfn16 mega, the leading edge hasn't moved in sometime, yet the total tasks number is increasing.

Just an estimate?

22k done, 15k to go fingers crossed :)

Two months maybe?

In the beginning we've been running first tasks only without double checks. Now we're running pretty much mostly double checks for those first tasks without new first tasks. That's why the leading edge is not increasing.

Next 10 posts
[Return to PrimeGrid main page]
Copyright © 2005 - 2022 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 1.06, 1.07, 1.20
Generated 3 Dec 2022 | 6:19:06 UTC