Folding@Home Update

Illustration for article titled Folding@Home Update

JKMiner just passed me on the Oppo F@H group. But let’s look at the metrics carefully:

He has completed 23 work units for a total of 487,069 which comes to an average of 21,176 credits per work unit.

My rig has done 13 work units for 465,676 credits for an average of 35,821 per work unit.

Objectively, if both users were only utilizing a single CPU and a single GPU for the work units, that would put jkminer ahead of me. But that isn’t true.



Advertisement
Illustration for article titled Folding@Home Update

Look at the “active clients” within 7 days. It counts to 5 separate work queues. So his set-up is either multiple computers, or multiple work units assigned to separate cores on his CPU (yes, it’s possible to split up your cores to handle more work units).

As it stands, his F@H performance is on par with mine, but I suspect that he is actually using multiple computers to try to get ahead. Meanwhile, on my stats, I’ve got only two active clients - or work queues. One for my CPU, and one for my GPU.



Advertisement
Illustration for article titled Folding@Home Update

So what does this mean? In racing terms, it means I’m running a hypercar like a McLaren or Koenigsegg, while the other guy is running a few Porsche Caymans.



Edit: I forgot to add that I was running F@H at Medium for those 13 work units. I’ve just cranked it to Full, and we’re gonna see what that does...not seeing too much of an increase in power consumption, though.