• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

PrimeGrid Winter Solstice Challenge December 19-22, with GPUs!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Halfway stats:

Rank___Credits____Username
5______11511965___10esseeTony
85_____1655161____GLeeM
149____849492_____zzuupp
152____825895_____Orange Kid
211____549473_____biodoc
316____283164_____wayliff
341____229228_____SlangNRox
342____229228_____geecee
343____222486_____Ken_g6
356____205631_____waffleironhead
504____57307______Drsignguy
574____26968______mk
623____13484______Ryan
727____3371_______Neurodog

Rank__Credits____Team
3_____52294323___SETI.Germany
4_____42494826___Czech National Team
5_____32317777___Aggie The Pew
6_____16659482___TeAm AnandTech
7_____14906562___Team 2ch
8_____11690628___SETI.USA
9_____10291663___Dutch Power Cows

The dual tasks per GPU is the real booster here though
Huh. That should never be necessary. :\ It should be all about adjusting the -m values.

Well, whatever works.
 
<cmdline>-m64</cmdline>

Ah, I see. And what does this do exactly? I'm now (possibly) gaining on the 4th place competitor, any further optimization is appreciated! 😀
 
I had not added an app_config to WUSS, due to the triple cards being packed so tightly and running near max temps already. But, Ken, I'm sorry to say, I added your config, told BOINC to read the config, waited...and nothing happened really. Temps stayed the same, still took 8m 51s per task, and the Utilization continued to jump up and down (avg unknown).

I then replaced the app_config with the dual task one, and the utilization flatlined at 99% used, the temps rose a bit, but holy crap, the time to complete, oh wow!

Single task again was 8m51s each. I was expecting to get just a bit less than double that once I was running 2 concurrently, so maybe 17m15seconds. But instead, they are completing in as little as 14m30s! (With an average of ~16m at the moment)

The Titans have far surpassed the other cards in regards to improved output. I'm impressed and a little excited and am hoping I'm not aggravating anyone by posting so much today and am hungry and am going to go eat now and.... 😛

5oemq.png
 
Last edited:
Enjoying the posts and learning.
I've come to the conclusion that the best way to run DC nowadays is to have just enough cpu power to help gpus work efficiently. Spend the $$$ on gpus, mother boards and power supplies and keep the rest at a minimum.
 
Really depends on what you want to run though. Relatively few projects support GPU crunching, while ALL support CPUs. For example, how long has Rosetta been around, still with no GPU tasks? Worse, if a project does support GPU crunching, it's often AMD or Nvidia cards, not both (or one runs great and the other sucks). And then they often run out of GPU tasks, while having plenty of CPU tasks, lol.


_
 
Day 2 stats:

Rank___Credits____Username
5______15698747___10esseeTony
89_____2194521____GLeeM
146____1200076____zzuupp
162____1132656____Orange Kid
202____825895_____biodoc
319____411262_____wayliff
346____326987_____geecee
356____306761_____SlangNRox
361____296648_____Ken_g6
371____273051_____waffleironhead
462____128098_____Neurodog
554____57307______Drsignguy
615____30339______mk
656____20226______Ryan

Rank__Credits____Team
3_____71718025___SETI.Germany
4_____58527302___Czech National Team
5_____44173584___Aggie The Pew
6_____22902574___TeAm AnandTech
7_____20539503___Team 2ch
8_____16565094___SETI.USA
9_____14458219___Canada

We're doing pretty well in 6th; Tony's barely hanging on to his 5th place.
 
We're doing pretty well in 6th; Tony's barely hanging on to his 5th place.

Thanks for the stats. I've lost BumbleBee and his 2 R9-280X's. Whether under load or not, he runs for about 5 minutes then the display goes out (tried both cards individually).

I don't have any spare equipment to troubleshoot with at the moment. He'll just have to enjoy his sabbatical.

So, I'll probably drop in the ranks. Fortunately there's not much time left for the others to catch up. 🙂

I'm shooting for 8 million on this day #2.

I needed 15.3M to do that, mission accomplished. I was hoping to make the claim that on the last day I'd pull 9M points, to continue the trend, but BumbleBee has prevented that claim. 🙁
 
Last edited:
I've lost BumbleBee and his 2 R9-280X's. Whether under load or not, he runs for about 5 minutes then the display goes out (tried both cards individually).

So far so good, regarding BulmbleBee's resurrection, but not with any double slot cards. I've managed to revive him partially, using an antiquated pair of Quadro 4000's. Output is abysmal, but it's all I've got for the moment.
 
The race is over! Mostly-final stats:

Rank___Credits____Username
5______23573403___10esseeTony
95_____3242902____GLeeM
155____1833824____zzuupp
168____1732694____Orange Kid
194____1361884____biodoc
322____653974_____wayliff
354____519134_____geecee
377____458456_____Ken_g6
383____431488_____Neurodog
393____411262_____waffleironhead
397____401149_____SlangNRox
635____57307______Drsignguy
652____50565______mk
706____37081______Ryan

Rank__Credits____Team
3_____109314788___SETI.Germany
4_____89082046___Czech National Team
5_____68023409___Aggie The Pew
6_____34765123___TeAm AnandTech
7_____31596383___Team 2ch
8_____25670165___SETI.USA
9_____24048714___Gridcoin

Tony hung on to 5th place; the TeAm onto 6th. 🙂 That might be our best finish ever! 😱

Tony did more work this year than the entire TeAm did last year!

I say mostly-final because double-checks are technically needed; but errors are very rare. But you should abort any work you haven't finished, to speed up the cleanup. Then you should go join the F@H race. 😉
 
Back
Top