Flathead knows exactly what he's talking about. 150-200PPD is what fast systems get when running the uniprocessor client with standard units. Just because you chose to run the SMP client doesn't make him wrong...dark41 wrote:You're right, it is a waste of time because you have no clue as to what you're talking about.
And FWIW, my quad (Q9450@3.2GHz + 9600GT) got 6500PPD from 170W the other day, and I have seen 7700 from 180W (4 uniprocessor clients running the fastest units going at 1000PPD apiece and 3700PPD from the 9600GT). So I make that 3x the work done for 1.7x the power. Not as impressive as 30:1 but still an improvement.
I would put it that you don't know what you're talking about: SLI is irrelevant to F@H, and on the units released before today GTX280s could pull 7000PPD on their own, which is why if you look on the folding farm thread several people have duallies and quads feeding four GTX280s and running SMP for ~30,000PPD from one box. Someone said that their box pulled 900W at the wall. 9 of your 100W systems would produce maybe 25,000PPD absolute maximum, so that's 5000 "free" PPD.