I see you guys are 400k-1million+ ppd, my 32kppd seems like a waste of electricity.
As others noted, after your host completed 10 WUs successfully, credits will go up because the project servers will start giving "quick return bonus", as it's then assumed that your host is reliable.
Re "waste of electricity": The actual amount of protein folding simulation done per kWh is about the same within a given GPU generation, regardless if the GPU is small or big. But it's true that credits/kWh to some degree gets better with larger GPUs because of Folding@Home's quick return bonus.
Re others' PPD: Currently, EOC's
Team AnandTech stats and
Tom's Hardware stats are showing the "average" PPD per user to be 390 k and 380 k, respectively. But keep in mind that this number is a little bit skewed: Before this race, this average is driven up because few but very dedicated folders are at work. Within the race,
hopefully many contributors with more normal (and typically not as optimized) gear will show up.
Our performance will not just depend on the few guys who have several and big GPUs, but it will also depend very much on how many TeAm mates with just one and more commonly sized GPU will join!
PS,
folding on CPUs however is a different matter. The folding performance gap between CPUs and GPUs is so large that folding on CPUs would often be a waste indeed. Depending on a host's CPU(s)+GPU(s) combination, it may in the end be more effective to let the CPUs just serve by supporting the GPU slot. The GPU folding application generally profits from having one spare CPU core all to itself, from the CPU running at turbo, and from the CPU
not having an AVX- and memory-intensive workload (like CPU folding) in parallel. Hosts with a Threadripper or dual-socket Xeons or the like might still do well at CPU folding though, during a short race like this at least.