Folding@Home needs a reality check, possible a less=more plan.
Posted: Sat Jan 31, 2026 5:25 pm
Howdy,
Im been participating with folding@home since 2003, and i donated 813 WU's or 125,732,757 Points just over the last couple of months, but i don't think i can donate any more, its because of implications of the cost of electricity for domestic users.
I came back in to this hobby with some enthusiasm and joking maybe i could run more GPU's, and its true i could run 4 maybe, but as electronics engineer i should have remembered the power consumption implications that hide away even for just 1 GPU.
The folding@home community had a boost in recruitment over covid, but has dropped to dedicated donators who are made up supporters who use it to either show off there processing power while donating, take advantage of systems idling away, or users that use cpu's to donate micro amounts and have been for 20 years.
It seems to me, and forgive me if im mistaken, that the science community that depends on the donations have come to accept that GPU users are just making use of wasted processing time, but the fact is that the WU's are gpu intensive and the cause higher power demands.
I feel the bonuses are linked with performance profiles for each hardware, and so when a system performance is restricted the bonus level of points falls, and is based soly on turbo completion of WU's.
If a user has a micro hyro power generator on there land, and they can use some of that power to power a computer its a win win senario, but most users can not achieve that, wind power and solar power at domestic setups can not power the levels often needed for 1080 TI system.
Power is expensive, and computer hardware is using more of it. People can debate that newer cards can generate more PPD than smaller GPU cards, but the cards are still using more power than the smaller cards.
I propose that the science community is going to need to plan for less, and forgive me again if this already the case, because originally WU's were completed by CPU's, and over the years CPU's have improved in efficiency while delivery more processing power, but in comparision of GPU's the amounts of ppd is small.
I think its a similar challenge to money donations when people pay by card. People are given the choice to donate 50p or not. I reason that most people find 50p to much so they donate occasionally, but if the amount was 5p most people donate more often. Say a 1000 transactions at 5p x 1000 = £50, or 5,000p. Whats more 50p or £50.
Im going to look at building a micro folding@home system and if i can, say a 50W an hour power consumption. Imagine if every house hold in the world could donate less than 25W to 50W toward folding@home we would record levels of distributed computing power.
Perhaps we need even smaller wu's ?
Thanks to AI and its high computer hardware & power demands, it could change the status quo of what we sore as normal for computer hardware standards for the considerable future.
Any how, good luck eveyone.
Dave Record
Im been participating with folding@home since 2003, and i donated 813 WU's or 125,732,757 Points just over the last couple of months, but i don't think i can donate any more, its because of implications of the cost of electricity for domestic users.
I came back in to this hobby with some enthusiasm and joking maybe i could run more GPU's, and its true i could run 4 maybe, but as electronics engineer i should have remembered the power consumption implications that hide away even for just 1 GPU.
The folding@home community had a boost in recruitment over covid, but has dropped to dedicated donators who are made up supporters who use it to either show off there processing power while donating, take advantage of systems idling away, or users that use cpu's to donate micro amounts and have been for 20 years.
It seems to me, and forgive me if im mistaken, that the science community that depends on the donations have come to accept that GPU users are just making use of wasted processing time, but the fact is that the WU's are gpu intensive and the cause higher power demands.
I feel the bonuses are linked with performance profiles for each hardware, and so when a system performance is restricted the bonus level of points falls, and is based soly on turbo completion of WU's.
If a user has a micro hyro power generator on there land, and they can use some of that power to power a computer its a win win senario, but most users can not achieve that, wind power and solar power at domestic setups can not power the levels often needed for 1080 TI system.
Power is expensive, and computer hardware is using more of it. People can debate that newer cards can generate more PPD than smaller GPU cards, but the cards are still using more power than the smaller cards.
I propose that the science community is going to need to plan for less, and forgive me again if this already the case, because originally WU's were completed by CPU's, and over the years CPU's have improved in efficiency while delivery more processing power, but in comparision of GPU's the amounts of ppd is small.
I think its a similar challenge to money donations when people pay by card. People are given the choice to donate 50p or not. I reason that most people find 50p to much so they donate occasionally, but if the amount was 5p most people donate more often. Say a 1000 transactions at 5p x 1000 = £50, or 5,000p. Whats more 50p or £50.
Im going to look at building a micro folding@home system and if i can, say a 50W an hour power consumption. Imagine if every house hold in the world could donate less than 25W to 50W toward folding@home we would record levels of distributed computing power.
Perhaps we need even smaller wu's ?
Thanks to AI and its high computer hardware & power demands, it could change the status quo of what we sore as normal for computer hardware standards for the considerable future.
Any how, good luck eveyone.
Dave Record