Question about gpu computational power/ppd

A forum just for SPCR's folding team... by request.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
chahahc
Posts: 75
Joined: Tue Oct 23, 2007 7:52 pm
Location: The Most Mexican Place in the U.S.A

Question about gpu computational power/ppd

Post by chahahc » Sun Oct 12, 2008 6:46 pm

I have a question about the computational power of the ATI 3000/4000 series gpus and the nvidia gpus. From what I've researched, ATI gpus have theoretically far more computational power than the nvidia line (in general). But average ppd for ATI gpus are around 1500-2500 ppd while nvidia gpus are anywhere from 4000-7000 ppd. I'm confused on why this is so. If I was to make a gpu folding rig,my main goal would be to contribute the most calculations regardless of ppd. What is the cause of this large gap. Is the nvidia line producing more calculations than the ati line or is there just a weird optimization issue...or something else....I'd much appreciate it if someone enlightened me on this issue.i found

Edit: Protein size! (After a more in dept research on my part, I have found the answer to my question.) :D

Post Reply