PS3 Downtime To Fight Disease

Aerenel writes, "CNN reports that Sony has teamed up with Folding@home to use the PS3 to study how proteins are formed in the human body and how they sometimes form incorrectly. From the article: 'Donating [a gamer's] PS3's down time to researchers could help cure Alzheimer's, Parkinson's or mad cow disease.' PS3 users will be able to download a software package that tracks when the PS3 is not being used. While gamers are in school, at work, or asleep, their system's Cell processor can be used to perform simulations for research organizations. The PS3, due in November, has gotten serious negative press in the past few months, and this refreshing good news may win back the hearts of gamers still undecided about purchasing the system."

Comments

Well...

[urbanradar]

This definitely seems like a good thing. But I wonder, will gamers really let that influence their purchasing decisions? Honestly, I have my doubts.

Re:In other words...

[urbanradar]

RTFA. It's on a voluntary basis. If you don't like it, don't do it. But it certainly is nice to have the possibility, and, as it seems, officially approved by Sony.

Fight my disease

[OSS_ilation]

Do I have Alzheimer's, or did I read about this last month?

Huh?

[Phanatic1a]

The PS3, due in November, has gotten serious negative press in the past few months, and this refreshing good news may win back the hearts of gamers still undecided about purchasing the system."

If I'm already ambivalent about spending that much money on a game system, the question "What will the game system, which I bought to play games, do when I'm not playing games on it?" is not likely to be a significant influence on my decision.

Not HAHA

[eldavojohn]

Well, this article is kind of a dupe but the CNN article has a lot of new information.

One of tidbits is that the researchers have to dumb down the PC distributed version so that it runs on even the slowest computer. In fact, they have to play to the lowest common denominator. With the PS3, it's standardized so they can inch out every bit of performance from the chipset. On top of that, they know there will only be on GPU so they can write the renderer for that and you'll see the protein folding on your screen. It will look all science-y and you can navigate around it. People might like this as a screen saver or conversation piece. The researchers are also hoping that it attracts people to also install it on their computers to aid in this endeavor.

Re:Not HAHA

[moro_666]

While probably 10% offtopic, i still hope this gets out to all the science app builders :

  Please make your applications scalable in % of cpu power used :)

  I have a laptop here, cute 3200 bogomips under it's belly, but i refuse to burn my system's cpu in favor of curing cancer. I wouldn't mind to give you 10% or 20% of the cpu power, because that wouldn't heat it up, but your applications that burn at 110% of the power available, are just not usable for most partly "idle" machines. Moreover, you'd get a bunch of workforce from fileservers which are idling on the cpu 90% of the time .. but as long as your applications just slay their performance in cold blood and cause them to run at nearly nuclear explosion temperatures (renicing the process will help against the first issue sometimes, but not the latter), you are just losing possible helpers.

  One minimalistic "sleep" or "delay" into your mainloop, and whoop's , you're going to get more work done than you have ever before. Until then, nothing will fold on my machines over here.

Re:Not HAHA

[thePowerOfGrayskull]

Why would you insist on "scaling" applications when you could just set the priority(win)/nice(*nix) level lower? (i.e., higher nice value in linux, lower priority setting in windows). That way, it will always yield to every other system that needs to use the CPU. (And nevermind the belief that a process can use only 'part' of a single-core CPU)

Re:Not HAHA

[proxima]

Why would you insist on "scaling" applications when you could just set the priority(win)/nice(*nix) level lower? (i.e., higher nice value in linux, lower priority setting in windows). That way, it will always yield to every other system that needs to use the CPU. (And nevermind the belief that a process can use only 'part' of a single-core CPU)

Because even if my computer isn't running anything important, running it at 100% CPU usage will invoke the full cooling capacity of the system. For a desktop, this means a louder fan noise. For a laptop, however, it may not be well-suited to 24/7 100% CPU utilization. The fans are small and can get _very_ noisy. The bottom of the laptop can get so hot as to be unusable on your lap; quite possibly, you could shorten the lifespan of your components.

Re:Not HAHA

[Anonymous Freak]

BOINC allows such a setting, but it's buried deep within the settings.

The United Devices client has it at 50% by default, and is easier to configure.

Re:asfaasf

[ArcherB]

I'm going to pay $600 so I can run a second copy of something that's already spinning away on my PC right now?

No, you are going to pay $600 to play games and run Linux on a Cell processor. The F@H thing is just a perk that you can either use to double (or quadruple) what you are doing on your PC right now or not. If you pay $600 for a PS3 just to do F@H, you are an idiot.

The next supercomputer...

[RetlawST]

...will be built using a cluster of PS3s.
 
  This cluster will be able to help cure cancer AND allow Madden 2007 to play an entire season in three minutes.

Pass

[mac123]

I'm waiting for the project that will have me use all of the spare time for my incredibly overpowered home CPUs (and with it ample amounts of electricity) to 'fight global warming'!

Re:Potential power costs?

[mspohr]

I guess you are exaggerating to illustrate your point but the power consumption of the PS3 is unlikely to be 500 watts (50 to 100 watts is more likely) and your power cost of 20 cents is two or three times what most people pay (I pay 10 cents in California). Your estimate is probably 10x to 20x too high.

Yes, it would cost electricity but most likely only a few dollars a month, not $750 a year and most likely wouldn't burn out the electricy grid.

Re:How the..

[kpearson]

Have you read anything about Folding@home before you started criticizing it? It is run by a public university, not a privately-held organization. The university will not profit from the research. If you can't bother to learn about the project, at least read it's FAQ.

Is it tax deductible...

[Chabil Ha']

If so, buying a PS3 could pay for itself over time. Seriously, I would signup, have them e-mail me CPU usage statistics muliplied by an industry standard rate as a reciept. They're happy to have CPU cycles, I'm happy to have the tax deduction.

Now if that's the case, can something similar be done with other CPU cycle donations to other projects? I have a multi-core server that spends most of its time idle. Might as well put to some use, eh?

Re:In other words...

[grommit]

Riiiight.. because the heating/cooling cycles of electronic equipment being turned on and off repeatedly is *much* better for it.

At least try to come up with a valid reason to not use it.

Re:Potential power costs?

[radish]

I agree that 500W is a little much, but according to this page the 360 takes 145W. I'd imagine the PS3 to be similar.