SCE Joins Stanford's Folding@home Program (B3D ID=32377)

The GPU numbers are skewed though by only doing certain types of calculations right?

They really need to get this thing running on G80 so I can join in! :cry:

With the impact the PS3 is having it seems a bit of a waste to have my CPU working on this!
 
They really need to get this thing running on G80 so I can join in! :cry:

With the impact the PS3 is having it seems a bit of a waste to have my CPU working on this!

Your regular computer CPU could handle things wich the PS3-clients dosn't do..
Sure, it won't do as much workload as the PS3's are doing, but it might do a different workload, and you never know if your simulated molecule is the one wich solves the Alzheimer riddle. :)
It sure beats having your computer running a screensaver tough. :)
 
Do you still remember the link ? I know xbd/Carl mentioned it in this thread but it would be great if we can find more references.

Meanwhile... we achieved 403 TFLOP with 16464 PS3s.

EDIT: Hmm.... 435 TFLOP and 17770 PS3s now
Heh, nope Carl was my source to. :)

But 403T/16464= 24.5G 435T/17770= 24.5G should point in the direction of him being right.
After all there's still quit a lot of headroom up till the theoretical peak performance of the 6SPEs.

Your regular computer CPU could handle things wich the PS3-clients dosn't do..
Sure, it won't do as much workload as the PS3's are doing, but it might do a different workload,
Would be nice with some reasoning backing up your postulate, so it doesn't look like "I hope because I do not want the PS3 to do well in any scenario". :)
 
AFAIK the PS3 client does all types of calculations.

Anyway, I contributed my first PS3 WU last night, while I was sleeping. It's 10% into its second WU now, but I might pause the Wu-machine to do some more gaming. ;)
 
My PS3 is almost 58% into its 3rd work unit.

Dsamn but does it ever get hot around the top vents! I could live wiht the fan spinning a bit faster and having a cooler system. Still it works just fine so I guess it's made to run that way..

It makes a bit more noise than when playing DVDs (and it gets loads hotter) but it's a realyl quiet games consone anyway. Only wii and gamecube are quieter of all the fan-cooled consoles I've ecxperienced and the difference to gamecube is fairly slight.

My old PS2 ws a howling monster in comparison to this.
Peace.
 
My PS3 is almost 58% into its 3rd work unit.

Dsamn but does it ever get hot around the top vents! I could live wiht the fan spinning a bit faster and having a cooler system. Still it works just fine so I guess it's made to run that way..

It makes a bit more noise than when playing DVDs (and it gets loads hotter) but it's a realyl quiet games consone anyway. Only wii and gamecube are quieter of all the fan-cooled consoles I've ecxperienced and the difference to gamecube is fairly slight.

Actually the German magazine measured the Wii, PS3 and 360, and the Wii came out louder than the PS3 when running a game disc. Albeit by only 0.1 Sone, 0.9 vs 1.0 ... ;) Not surprisingly, the 360 was a 'bit' higher, at something like 3.8 sone.

But I'm sure that when they're just doing calculations, the PS3 will be louder than the Wii.

Actually, the PS3 is a bit louder than I expected. But certainly still tonnes quieter than my old PS2, and I can't tell when it is spinning a disc or the HDD is active. The only thing you hear is the cooling fan, or the air it pushes out. So I expect that it will get quieter in later revisions still.

Which is good, because I will eventually want two. Already I am thinking I need to move it from the living room to my game room (for testing with the wheel) and back ... :D (yes, I have two ps2s as well)
 
Crossbar said:
Thinking about, I find it pretty amazing!

Sony must be pretty confident about the build quality of their system. Imagine all units running 24-7 at close to maximum stress with the fan working continuously.

Yes... I'm impressed by the build quality of PS3 so far. It's been up for close to 4 months (since I bought it), mostly in the Resistance clan "page". Didn't pay much attention to it.

Should you do the same, remember to clean up the dust "hair" around the ventilation every week :D. I did so only after 3 months, so it was pretty thick (like a man's beard).

Maybe those are the sort of thing's they'll tweek over time? It seems, from the Stanford forum, that they are working quite closely on the clients front end/usability.

I hope so ! In fact, they *should* continue to polish the entire PS3 user experiences (e.g., standardize/clarify the use of the "Playstation" button)

What do you have to through to quit the client?

Press triangle, down arrow to "Quit Folding" and then wait for 0.5 - 1 second for XMB to appear. It's easy but it can be annoying for someone impatient (especially if he has to do it everytime before playing a game).

Well, yeah, no one knows <looks for 'broken-crystal-ball-smilie>, but don't you think it'd be -really great- to have something that could contribute on such a large scale? Although if PS3 gets anywhere near PS2 numbers I think even ~1% would provide 'wow' TFLOPS, but more's always better, right?

Hell yeah ! It would satisfy my geekhood. If Sony has sold through 3 million at this point, then we are looking at around 30,000. Right now, we are at 564 TFLOP with 23027 PS3s.
 
Press triangle, down arrow to "Quit Folding" and then wait for 0.5 - 1 second for XMB to appear. It's easy but it can be annoying for someone impatient (especially if he has to do it everytime before playing a game).

Just hit "O," assuming no menus are up. ;-)
 
Argh ! yes... forgot about that... or the "Playstation" button. Both will allow you to quit F@H in 2 button presses.
 
Argh ! yes... forgot about that... or the "Playstation" button. Both will allow you to quit F@H in 2 button presses.
With a slight difference. When you "Quit" by hitting circle, the default selection is "Yes." When you quit by hitting the PS button, the default selection is "No." The PS button "Quit" being universally "No" by default. I just thought that was a little odd. :p
 
With a slight difference. When you "Quit" by hitting circle, the default selection is "Yes." When you quit by hitting the PS button, the default selection is "No." The PS button "Quit" being universally "No" by default. I just thought that was a little odd. :p

:LOL: UI guidelines would say PS button is correct, but I like the default "Yes" better for F@H. This is one of the examples for Sony to clean up their overall user experience.
 
Actually the German magazine measured the Wii, PS3 and 360, and the Wii came out louder than the PS3 when running a game disc.
Sorry to have to say this but if tyhey really came up with such results they're completely incompetent. Hard to imagine for being german and all..a people renown for being efficient and all :cool:

I have a wii also and it's SO QUIET. Regardless of if it spins a disc or not. The PS3 cooler is very quiet also but it's not quieter than a wii. There's a slight rumble emanating from inside that black tower that makes it a bit "noisier". It's nothing worth complaining about though and it's very inobtrusive since it's not high in pitch. PS3 is mazaingly silent for the power it has.

Not surprisingly, the 360 was a 'bit' higher, at something like 3.8 sone.
Approximate descripitions of sound during gameplay from the 3 major consoles this round:
Wii: hhhhhhh
PS3: rrrrrrrrrrr
360: UUUUUUUUWWWWWWW
:cool:
Still I wub my 360. Wouldn't give it up for anything. Really comfy controller also. One of the best ever apart from the D-pad.
Peace.
 
:LOL: UI guidelines would say PS button is correct, but I like the default "Yes" better for F@H. This is one of the examples for Sony to clean up their overall user experience.

I'm hoping they'll keep "Yes" as default within F@H (but "No" as default for the PS button). While it's not consistent, it does make it quicker, and more convenient, to exit. "O" -> "X" Done! I'm thinking this was purposefully done that way.
 
You can see the current 12 PS3 projects here:

http://fah-web.stanford.edu/psummary.html

It's surprising to see that there's a maximum of about 2 days elapsed time within which to submit a finished work unit. With work units taking about 8 hours to run, a PS3 must be folding for an average of at least around 4-5 hours per day to produce any useful results.

With so many PS3s folding, let's hope there's enough work for them to all do!

Oh and this comment about FLOPs is very interesting:

http://forum.folding-community.org/fpost173708.html#173708

Just counting FLOPS can be deceptive. The GPU does a lot of FLOPS, but is less efficient with them. For example, the GPU doesn't allow a scatter (random access write) so we are forced to do some calculations twice: the force on i & j is related to the force on j & i, but saving that data requires a scatter, so we just recalc.

That's the faster way to go on the GPU, so it's the right thing to do there. It also brings the FLOPS *way* up, but it means it *requires* 2x the FLOPS just to do the same calculation.

unfortunately we don't know what the situation is with PS3 FLOPs, but it's very likely that every FLOP counts on PS3. It's not possible to discern from that comment what proportion of GPU calculations are done "twice".

Jawed
 
Heh, nope Carl was my source to. :)

LOL, well I'm not working off of any insider information here or anything... it's just 1 SPE is always grabbed by the OS itself, and since we also know RSX not to be a part of the folding ops themselves, that just leaves us with a max of 6 SPEs. :)

By the way the contribution PS3 is makign thus far is awesome; after the Europeans settle in post-launch a couple of weeks from now, we should try to coordinate a worldwide "24 hours of folding" to see what maximum throughput we can attain.

PS - PS3 is presently at 577 TFLops!! :oops:
 
I need to stop playing GOW 2 and VF5 just long enough to help out my fellow comrades.
Indifferent2.gif


Little of topic:
Has anyone tried a Linux version of folding on the PS3? Just to gage how efficient or fast the output of the PPU maybe by itself. Taking into consideration that someone has actually gotten “foldingâ€￾ working for YDL on the PS3. It would make for interesting project…
 
Has anyone tried a Linux version of folding on the PS3? Just to gage how efficient or fast the output of the PPU maybe by itself. Taking into consideration that someone has actually gotten “folding” working for YDL on the PS3. It would make for interesting project…

Is that a question? Or are you saying someone has gotten Folding@Home working on PS3 Linux.

There is no PPC Linux client so it should not be possible. The only current Linux binaries are for x86. And the only PPC binaries are for OSX.
 
Is that a question? Or are you saying someone has gotten Folding@Home working on PS3 Linux.

There is no PPC Linux client so it should not be possible. The only current Linux binaries are for x86. And the only PPC binaries are for OSX.

It was a question taking into account that a PPC version of folding did/could exists. Honestly, I could have sworn a PowerPC folding version did exists.

Edit: Ah, MacOS
 
Back
Top