• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

GPU Folding

T

The Goose

Gawd
Joined
Oct 29, 2006
Messages
780
Currently I own a machine with a P4 3GHz w/ HT (two instances of folding) and a video card in it that cannot fold. Soon I'm getting a X1950Pro to put in there, and obviously, I am wondering about GPU folding. I have a few questions about it and I wonder if you guys know the answers.

1. I read on Stanford's site that you should leave 25% of the CPU free for the GPU client. How would I set this up, would I just have one instance running, or what?

2. Depending on how much CPU power I'd be losing, how much of an increase in points or productivity (folding-wise) would I see with this?

3. I also read that only the Catalyst 6.5, 6.6, 6.10, and maybe 6.11 drivers are supported. Do these recognize the X1950Pro and work with it?

4. During normal usage I would just leave the folding running and I would just turn it off during games, right?

Thanks in advance for helping.
 
OK, I looked on that thread and a couple of the links in it and apparently it's most productive to run one instance of F@H and leave one "core" for the GPU. I also found gpu.fahinfo.org, which has info on an owner of a X1950Pro using the 6.11s getting about 600 PPD, which would really improve my average. Thanks for the link.
 
I have a X1950Pro and am getting 600PPD from the GPU and about 175PPD from the other core (X2 3800 O/C to 2.65ghz)
 
You might want to run only one instance instead of two since HT doesn't help the cause very much.

Especially if you start running a GPU client.
 
Why in the hell did they make ATi's cards fold before nvidia's? Isn't nvidia the bigger of the two? Valve's hardware surveys show it anyway...
 
The architecture used on ATI cards (the X1XXX series) is better suited for folding.
 
GPU foldig started with trying to get nVidia card to fold.
Stanford found that they did not do i very well due to their arcitecture.
I think its because you have seperate vertex & pixcel shaders, it harder to program them to do a different job.

With ATI the shaders are either so are easier to program to do a different job.
Hence ATI GPU folding started first with the X1*** series cards and hopefully nVidia will soon start with the 8*** series cards.

So nVidia cards do fold but just so slow that its not worth releasing a client for them yet.

Luck .............. :D
 
You must log in or register to reply here.
Back
Top