Team CFC: Folding@Home project

Status
Not open for further replies.
ggganz, have you checked the expiration date on your WU ?
There's nothing more annoying than to lose a WU at 80-90% because of the damn expiration date.
Once it happened at 97% :mad:
Basically, that means a week of folding down the drain (4 days for the too late one, and another 4 for the next one).
In fact, the main gripe I have with the SMP is the fact that it needs an active connection to fold. If I lose my connection for even a short time, the program shutdowns, and if I'm away my computers stays idle for hours.
Sometimes it shuts down an hour after I leave for the week-end.
This is why I lose so many WUs.
 
Yeah, I should be fine. I don't have it folding 24/7, and right NOW it is 29 hours to go. But I should be fine... it isn't due for 905 hours from now. In order to miss that deadline, I would have to have my computer on less than 3% of the time.
 
Well, only 7 months for me, and we're tied for the 26th place.
Have you been slacking ? :p
;)
 
Just finished it!! :dance: I'll finally be in the top 100 on this team!

I just got a 292 point WU: 2% and 9h 09m to go. That's WAY faster than I usually do... I think they over-credited this WU.

EDIT: Okay, FahMon got some more sense... 3% and 9h 25m to go. I have it set to measure just the last frame, because I've been using Sleep much more. I guess I'll just see when I finish it!
 
Hi,
I have a quick question.
I'm using the SMP 591 beta 6 client.
Is there an easy way to dump the protein I'm working on and start a new one ?
The reason I'm asking is that it will take my PC at least 30 hours to complete the unit... But the deadline is in less than 24 hours. So, it's a waste of processing time.

Thanks in advance for the answer.
 
If I use the System Preferences panel on my Mac to stop F@H, it crashes it and fails to restart at the checkpoint. The client will retry a WU three times before giving up. I don't know whether the Windows client is as fragile as the Mac one, but this may be a way to force it to a move on to new unit.
 
I thought about deleting some files (the work folder for example), but I'm not sure which ones.
 
Thanks for the confirmation.
It might be useful in the future, as it is something that frequently happens.
Also when the client is folding and reaches the deadline it does not terminate properly (I have to do it manually and restart). Is that normal ?
 
Good lord, I'm now seeing much better just how Sahkuhnder is racking up that kind of points-per-day - I added the SMP client to a temporarily-out-of-production dual-processor dual-core 4GB RAM server a few days ago and it's cranking out a 1760pt work unit every 12-16 hours. :eek:
 
Wow! What's its clock speed, and how many clients are you running? My four core 2.66GHz system takes 20 hours for each p2605 unit (1760 points = 88 pph), running one SMP client. That's the highest points-per-hour it achieves.
 
I cannot get the SMP console client to work on any machine. :( Just have to be satisified with the 5.x console client.
 
Wow! What's its clock speed, and how many clients are you running? My four core 2.66GHz system takes 20 hours for each p2605 unit (1760 points = 88 pph), running one SMP client. That's the highest points-per-hour it achieves.

1333MHz bus speed. Just the one client, I thought that the SMP client makes use of all the processors in the box. I can't imagine getting more folding out of it. :eek: And it looks to be settling out to be about 16 hours per (1760pt) work unit.

On the flip side, I wasn't able to get the SMP client running as a service on the workstation I was trying it on, so I think I'll just go with what I've got for now. :D
 
Hmmm i've had my FH running constantly again but according to the table im inactive. :confused:

Anyways has there been any actual benefit come from us all doing this yet? Anything i can understand that is :)
 
My SMP client runs my four cores at about 78%, 65%, 65% and 50%. Overall, I'd say it was only using 2/3 of my CPU power, even when nothing else is active.
 
My SMP client runs my four cores at about 78%, 65%, 65% and 50%. Overall, I'd say it was only using 2/3 of my CPU power, even when nothing else is active.

I started a second client on my heavy-hitter server, and while there were four more fah processes started, overall it went from four processes at 25% each to eight at roughly half that each but still totalling 100%, and the single WU averaging 1% every 9-10min went to 1% every 20min and the second client's WU was going 1% every 20min, too. I did the -local flags on both, and did use a different client number, but I obviously didn't do something right. I might take another whack at it soon, but it looks like that server might be going into production in the next week or two, so I may not have a chance to troubleshoot before it stops mattering.
 
Looks like you're about to break the million point barrier today or tomorrow.
Congrats :goodjob:
 
Status
Not open for further replies.
Back
Top Bottom