LinkBack URL
About LinkBacksIncidentally 1stRaven, you realise that you accidentally folded for team 67996? http://fah-web.stanford.edu/cgi-bin/...type=userstats if that's still going that could partially explain your lower than expected ppd.
Incidentally 1stRaven, you realise that you accidentally folded for team 67996? http://fah-web.stanford.edu/cgi-bin/...type=userstats if that's still going that could partially explain your lower than expected ppd.
Stat update is in process at the moment so cant check but if its 1 wu then it was from a while back, I just let it ran trhough before changing the team back.Originally Posted by nibbler
as for the 85k, your right, it seems to be well over considering it was at 45k ppd yet I was only getting around 22k.
stats update done and no change so it will have just been an accidental entry from a while ago.
I've just been looking at the stats - I'm no longer our highest producing contributor. 1stRaven has taken the lead!
I tried to overclock my server a bit more so that it could get through the BIGADV units within the alotted time, and it's been crashing after a couple of hours, which I've not picked up until after work/when I've gotten up in the morning, losing me (and therefore us) points while the computer has sat there following a reboot, waiting to be logged on. Think I've nailed the top 24/7 O/C for folding now though, so we'll see. I'm only running my 920 i7 @ 3.8Ghz which i'm a bit disappointed with to be honest.
I'm hoping to get my main rig assembled this weekened, and get 2xGTX 480's plus my i7@4.2Ghz all contributing again so that this can run while I upgrade my server.
Wish I had access to our exchange server...Nothing wrong with a bit of healthy competition I suppose.....
I hate to do this to you but it'll only be a short time. That said, I am currently adding another two servers into the mix, though both only have single 2.57 quad cores so I'm only expecting a 1.3 bigadv units a day out of them. That'll be it for the weekend.I've just been looking at the stats - I'm no longer our highest producing contributor. 1stRaven has taken the lead!
I tried to overclock my server a bit more so that it could get through the BIGADV units within the alotted time, and it's been crashing after a couple of hours, which I've not picked up until after work/when I've gotten up in the morning, losing me (and therefore us) points while the computer has sat there following a reboot, waiting to be logged on. Think I've nailed the top 24/7 O/C for folding now though, so we'll see. I'm only running my 920 i7 @ 3.8Ghz which i'm a bit disappointed with to be honest.
I'm hoping to get my main rig assembled this weekened, and get 2xGTX 480's plus my i7@4.2Ghz all contributing again so that this can run while I upgrade my server.
Wish I had access to our exchange server...
On monday, one of the currently folding servers will go so thats a single 2.67 quad core out but I will have another 24 core server to come into the folding mix for a while as well as few other quad core boxes. I'm wondering if I can break the 100kppd mark.
Do it! Shouldn't an i7 at 3.8Ghz easily manage to do a bigadv on time? DDY does that iirc. Although I suppose if 30% of cpu usage is tied up in the gpu client it could be a lot slower. Still, should be doable.
The one it's processing at the moment, the ETA is 4 hours longer than the deadline, so I guess I've got a particularly processor intensive work unit or something. SASPRO on the bit-tech forums seems to think that you need to be at over 4Ghz to get through each unit while running dual GTX 295's. I was getting through some stuff ok for a while, just seem to have suddenly hit a wall metaphodrically peaking. I've tried deleting the work and getting it to start again, so maybe I have to get through this unit and submit it to not receive one this big again.
Any ideas peeps? I've got 3 days plus to get through this current one...
I have always found that trying to do smp folding with the gpu client always crashed my machine or at least, caused the wu to fail.
How long is it taking you to fold each % and on what size of wu?
It's taking nealry an hour per percent.
The work unit is 250001 steps. Where do I find the size?
In HFM.NET - on the left, below "status" there's "credit". It correlates with the size of a WU (481 credit is 500K, takes not much, 921 is 2M, takes a while, 1395 is 250K, takes forever - these aren't bigadv, of course).
The bigadv units are giving about 7k points per unit but take a while. Looks like my servers have finally kicked into gear as they folded over 77k points yesterday. I even overtook someone lol.
that's awesome man. The bonus points on a big adv units is huge though I think, DDY says he get's about 65k points including bonus points iirc.
I had a little break whilst I moved into my uni halls and sorted everything out and I've decided to rejoin the cause (most probably part time). So I installed the latest GPU client for Fermi cards and it's failed about 10 WUs so far giving this error message: "
mdrun_gpu returned
[16:41:35] NANs detected on GPU
I tried turning down the overclock on the GPU but that hasn't solved the problem at all... has anyone else encountered such problems before?
Try running in xp compatibility mode if you have win7 pro? What client are you running?
I've set the compatibility mode to XP SP 3 and the client i 6.30r2 - I'll report back on the stability issues when I get back this evening!
EDIT: it's running at 51s per %, let's hope it lasts
GPU3 client? I just did a quick google, I have no idea if that will work, let's hope so!
There are currently 2 users browsing this thread. (0 members and 2 guests)
Posting Permissions
Reply With Quote
