[email protected] vs. Power Consumption
At what point do the benefits of folding get passed by the benefits (of not folding) on the environment?
Some personal background. I used to fold with team 32 over at ocforums. I used it for stability testing and also to heat my room in the winter (I live in the basement). Now that energy efficiency is coming to the forefront and people are turning off lights and swapping incandescent for fluorescent, I've been turning my computer off when not in use. I've also purchased a second, very efficient, computer to download torrents that consumes a fraction of the energy mine does.
I'm wondering if pretty soon you'll be folding to find cures to problems caused by Global Warming, indirectly caused by folding.
Yours isn't the only similar view on these boards. I won't mention any names though..... :whistle:
I figure it's a trade-off and like to think that the fact I haven't owned a vehicle in 7 years makes up for the fact that I don't like to turn off my computers.
Pscout seems to have the best mix going when it comes to folding....
He's explained here a couple of times that he prefers to find the best power to output ratio for his folding rigs rather than my goal of max OC and power be damned. :)
I'm not sure however that I would go as far as setting up a folding "farm". The only computers folding in this household are ones that are actually used for other purposes. (I admit though that they probably don't need to be running 24/7).
I'm not dissing your viewpoint, but think that there are other power wasting pastimes that could be dropped before worrying about folding.
The benefit of [email protected] is for disease research mainly, cancer, Alzheimer's, Parkinsons, etc. It has come a long way and now there are multiple pharmaceutical companies which have expressed excitement about the published results and are using the FAH results in their own research.
An example of what is being done with the data is here: Dimerization of the p53 Oligomerization Domain: Identification of a Folding Nucleus by Molecular Dynamics Simulations
For me if I can help to get a drug developed to save or even make someone sick feel better it is more important than the small amount of power my folding boxes use.
Everything comes at the expense or opportunity cost of something else, and I dont think the world will end from folding. My advice is if you are worried about it, look for other ways to reduce your power consumption around the house, buy a smaller car, carpool, read instead of watch tv, etc. :)
As everyone has pointed out ... there are many trade offs.
I have limited the max number of rigs i run on my farm and focus more now on efficiency rather than max oc. If i had only 1 or 2 rigs as i used to, I would make the same tradeoff as sswillson for a higher oc. When we get 8 way's, I miight actually reduce the number of rigs a bit. And with intel finally joining the energy efficiency competition against AMD with c2d and now c2q, and business and consumers demanding it, i expect to keep moving to the lastest tech to take advantage of it ... not to mention that i like playing with the latest tech as long as it folds for humanity.
More environmentally friendly sources of electricity would help cuz the world is not gonna stop growing unless armagedon happens.
good feedback guys. i'm just curious to know what the collective thoughts are and get different opinions on the subject.
i'm on the fence leaning toward the low power consumption side. i do a lot to conserve power already around the house but i just know my system is probably best turned off when not in use.
i do believe greenhouse gases cause/accelerate some of these illnesses so it's a tough call.
Yes, great research has been done but in areas without hydro electricity, I find the environment will take a beating. Greenhouse gases produced by coal-fired power plants contributes to ailments like lung cancer and other respiratory diseases. Now let's be honest: nost of the US electricity comes from coal-fired plants. It makes me wonder if this whole folding craze is doing more harm than good.
Folding started as a way to use SPARE computer processes when they were not being used. Basically, if you were working on your computer the Folding client would work in the background. Now people are using PCs exclusively for folding and those power-hungry PCs run 24/7. It has become a competition that has exploded in popularity in the last two years and it does not look to be letting up.
I have no problem seeing people folding with energy efficient processors (stock clock X2 EE, C2D and Xeon) but I cringe every time I see someone folding 24/7 with a quad core, overclocked X2 or even worse...a GPU.
There are certain things I would NEVER do without if I was folding:
- An energy-efficient, stock-clock processor
- An 80Plus certified PSU
- ONLY RoHS components
OR if you would be building a rig only for folding I would tell you not to buy a computer at all....you should just buy a PS3 since its performance per watt in folding is second to none.
But that is just my opinion.
Sky ... I agree with most of what you said ... except the ps3 recommendation ... c2d or c2q is a much better performer in term of points per day per watt.
And while we can debate the use of points as a measure of value, it is the only measure that can be used usefully across clients and platforms. Tflops is another measure but since clients like the ps3 and gpu can only be used on a subset of folding simulations the points system is stanford's attempt at normalizing many factors.
If you have a C2D you probably don't have on-board video so you have to add the power consumption of a GPU at idle, plus the ram, hard drive, fans and quite a few other things. IMO, a C2D rig with an 8600GT, single HD, 2 fans 2GB of ram and a reasonably efficient PSU will be pulling in far more than 150W when folding.
Take for example my HTPC which was folding for the Guru3D team a while ago. It has a X2 3800+ EE (stock) processor, a 8600GT, single HD, 2GB of ram and an 80Plus Etasis PSU. When folding on both cores it pulled a constant 162W from the wall.
The same rig consumed ~ 96W at idle.
I expect i will also see a slight decrease when i switch it from running smp under windows to linux. It folds under windows while i finish tweaking it, but the windows smp client is currently a bit less efficient than the linux client.
My watt measurements are done on an APC ups and accurracy is ~ +/- 6 watts relative. I have not calibrated it with a load meter, but i think it is close enuf for my purposes.
IIRC, the ps3 clients are benchmarked to produce 900 ppd ... i never got one to fold with once i saw its power needs since i don't game and didn't need the blue ray, and it was kindof boring to fold with for an enthusiast like me. Similar reasons for why i no longer fold on gpu's.
By comparison, GPU folding had an incremental power consumption of ~ 45 W when i last measured it for a ppd of ~600-700. For a dedicated folding rig i dropped highend video cards since they cost ~35 W when idle.
There is an even more efficient setup running the dual core meroms in terms of ppd per watt, but it would take a lot more of them to scale like the quads do.
|All times are GMT -7. The time now is 04:42 AM.|