New Posts  All Forums:

Posts by tictoc

Congrats.  The OP has been updated.  If anyone sees anything that needs to be added or changed let me know.  If I missed it earlier in the thread, congrats to the following new team project leaders:@BeerCan - POEM@Home @magic8192 - World Community Grid @nova4005 - theSkyNet POGS
POTM Update   Van Der Waerden Numbers Start: 52 - Goal: 30 - Currently: 23   |  This goal was crushed, and we might grab one more spot before the month is over.    Enigma@Home Start: 47 - Goal: 44 - Currently: 40  |  We blew past our goal and a few more spots might be within our reach.    GPUGrid Start: 24  - Goal: 23 - Currently: 24   |  We have given a nice little boost, to what can be a frustrating project to run. 
 Congrats on all the milestones.
I am offline now, and I don't think I'll be up and running for another 12 hours or so.
Looks like my 290 subs are skewing the results a bit. I might get a chance to throw up some new submissions with a 7970, 290x and an RX 480 this weekend. Not sure if I'll have time, but I'll see what I can do.
I would imagine the points will be the same. I think Universe is staying with a fixed point system, regardless of the length of the tasks. Either way, as soon as I get my 2nd ES chip I'll be joining the party with 10 cores/20 threads. I really like the way this project has progressed since the beta app, and the research is interesting.
 There are only two projects that utilize fp64 instructions.  All of the GPU apps for MilkyWay require double precision for computations.  That is why 7970s and 7990s are still the best bang for your buck on MilkyWay.  The other project that uses fp64 is PrimeGrid.  For PrimeGrid the only application that requires double precision is the Generalized Fermat Prime Search. It looks like the peak fp64 performance for that Tesla is 665 Gflops which is roughly the same as an R9...
 I just had the same thing happen a few minutes ago.  I was working on one of my other rigs, came back to my main rig, and it was super laggy.  Once I finally got BOINCTasks up front, I saw that I had 8 DENIS tasks crunching.   It is good to see the project back up and running.  It seems to be a pretty worthwhile medical project, and when it was up and running tasks the admins were very responsive.
I would have to agree. I've been rocking Arch for quite some time, and the biggest thing that I don't miss with Arch, is all the conflicting apps and drivers I would get with Mint. It makes it much easier to trouble shoot when there are very few apps installed.I don't have any new suggestions for what to do with the Teslas. On my testing server, that had some Firepros, the quickest way that I got it running, after pulling the 7970s, was to do a fresh install with the...
With the number of tasks you are running, the tasks waiting to be validated may take days to level off. It is highly dependent on how many other people are crunching the project. I currently have 2800+ tasks waiting to validate on VDW.
New Posts  All Forums: