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

[Butta]

Could the RSX be included in the folding to help out CELL?

 

[Dave Baumann]

Could the RSX be included in the folding to help out CELL?

RSX is derived from the GeForce 7 architecture which, due to its threading implementation, doesn't lend itself to the types of processing F@H requires. The Pande group concentrated on ATI's R5xx parts and is also currently looking at G80. Look for posts on this forum by mhouston if you want more details.

 

[Arwin]

I think I pay € 14,4 per month for this ... Don't think I'll keep it on like this all year, though. But right now, it's racking up WUs rather nicely and will continue to do so all week.

By the way, I think they updated the FAQ about PS3. It reads that the PS3 does GPU like stuff relatively faster than CPU type stuff, but it is also capable of CPU type stuff reasonably fast and they're working on optimising some work-types for that now.

Some more screenshots:

This one is folding and browsing combined. There's one downside to this I noticed, which is that the new posts button doesn't work properly in this mode. If you go back to the regular XMB browser, it runs fine.

 

[Arwin]

Some updates on the fluctuations in Flops count

3/26/2007 Update on flops count
We have been looking into the flops count and its large variations and have found one more issue. The initial stats were based off the average we had seen during testing (yielding approximately 25 GFLOPs for a single PS3). However, the pre-launch testing period used big proteins which will result in higher GFLOP utilization. When we went live, we started our initial post-launch phase with small proteins to test the scientific validity; these smaller proteins have more overhead (since they spend less time calculating the force -- which is highly optimized) and thus the GFLOPS are lower now. As we switch back to the larger proteins, we expect to see an increase in the FLOPS per machine, and thus the overall FLOP count will change dramatically. We stress that there is a wide variation in FLOPS we can get (easily a factor of 3x) and so we expect the number to vary widely until we reach some steady state average.

 

[Shifty Geezer]

For the sake of Sony's PR, they want to flush the system with large proteins and get the 3x increase, to hit the Petaflop landmark!

 

[speng]

Anyone know how to set the folding@home program as the screensaver?

Also, I tried to get my bookmarks up in the browser on the folding@home page but couldn't. So I guess you just typed in the URL itself?

Speng.

 

[mckmas8808]

Anyone know how to set the folding@home program as the screensaver?

Also, I tried to get my bookmarks up in the browser on the folding@home page but couldn't. So I guess you just typed in the URL itself?

Speng.

1. At the XMB crossbar highlight Folding@Home.
2. Hit the triangle button.
3. And it should give you the option to run every 10,20, or 30 minutes after idle.


And also using the internet while folding you DO have to go to google. And you can do that by going to the stats page at Folding at Home. You can't type anything in the URL.

 

[Arwin]

For the sake of Sony's PR, they want to flush the system with large proteins and get the 3x increase, to hit the Petaflop landmark!

Well, no ... they were at 990 when they changed it for testing, so that's definitely not it! Instead, I think they cut it short on purpose because they wanted to verify their science before they announce the first Petaflop to the world, lest they get a lot of backlash over blown up results (they say as much on their site).

 

[speng]

1. At the XMB crossbar highlight Folding@Home.
2. Hit the triangle button.
3. And it should give you the option to run every 10,20, or 30 minutes after idle.


And also using the internet while folding you DO have to go to google. And you can do that by going to the stats page at Folding at Home. You can't type anything in the URL.

Thanks much.

 

[Gradthrawn]

+1 Team Rank (133 -> 132)? Or is it too early to say for certain yet?

 

[Deepak]

Wow, Team B3D's output has seen a huge jump. I am still there at no 56.

 

[Arwin]

I managed to have my PS3 contribute 12 since Friday, and I think patsu even squeezed out 16 by now. With a few more PS3s contributing like that, we recent FW 1.60 owners are doing our little thing, I guess!

My girlfriend does like this very much so who knows we will keep doing this longer than just this week ...

 

[patsu]

Wow, Team B3D's output has seen a huge jump. I am still there at no 56.

I have to fold more than 23 work units (2-3 more days) to break into 100.
I'm at 18 now... taking my own sweet time.

 

[Sis]

I have to fold more than 23 work units (2-3 more days) to break into 100.
I'm at 18 now... taking my own sweet time.

lol, you act like you're the one doing the calculations!

I, on the other hand, am doing the calculations by hand. On paper. And mailing them in. Hoping to complete my first work unit in 2017.

 

[patsu]

lol, you act like you're the one doing the calculations!

I, on the other hand, am doing the calculations by hand. On paper. And mailing them in. Hoping to complete my first work unit in 2017.

That's because my master is not at home
(Now where is that PS3 be-headed Bishop picture when I need it )











But seriously... I feel bad halting Folding for Resistance sometimes. We'll see how long this keeps up.

 

[Gradthrawn]

But seriously... I feel bad halting Folding for Resistance sometimes. We'll see how long this keeps up.

User: *Hits circle*

F@H: "Are you sure you want to let another loved one die of cancer?"

User: "Damnit.... no.... I guess" *cancels exit*

 

[inefficient]

lol, you act like you're the one doing the calculations!

I, on the other hand, am doing the calculations by hand. On paper. And mailing them in. Hoping to complete my first work unit in 2017.

You grossly underestimate how much slower you are than the Cell processor!

You would not be able to complete 1 billion floating point calculations on pen and paper within your life time even if you were some sort of savant. The Cell can do dozens of these every second.

 

[Sis]

You grossly underestimate how much slower you are than the Cell processor!

You would not be able to complete 1 billion floating point calculations on pen and paper within your life time even if you were some sort of savant. The Cell can do dozens of these every second.

Huh. Maybe true. I had started out trying to calculate how long the single work unit was going to take me, but I got a divide by zero error so I just guessed at 10 years. I figured once I got into a rhythm it'd go pretty quick.

 

[inefficient]

Huh. Maybe true. I had started out trying to calculate how long the single work unit was going to take me, but I got a divide by zero error so I just guessed at 10 years. I figured once I got into a rhythm it'd go pretty quick.

Puny human!

 

[patsu]

User: *Hits circle*

F@H: "Are you sure you want to let another loved one die of cancer?"

User: "Damnit.... no.... I guess" *cancels exit*

Pretty close, pretty close.

That and also I spent too much time playing (It's a really bad idea to have PS3 sit besides my work computer).