re: PS3 The most recent major system upgrade includes Folding@home. It is in a bundle with/inside of Life with PlayStation. I choose to run F@h in Priority Mode only, not sharing the channel with real time clouds and news. Updates since the total system upgrade are downloaded as you enter Folding@home. It has done that at least once.

I dunno, I updated my system using an external download and it still says I need to download Life with PlayStation from the unit.

p.s. new to PS3, old to folding.

SMP goodness with new rig. Running the -bigadv version of fah6 and 16 FahCore_a2.exe. Stock Mac Pro 8-core 2.26 GHz 6GB. Temps 137 F and using ~300 watts. Hoping for bonus points.

Major bonus points with -bigadv Units. 25,403 for WU: Bonus 32083 for a total of 57,486.
33min 20sec frame times ~55.5 hours. A little over 1K pph.

WU #2683, 25403.00 points, 6 days. Big data file to upload: 100159508 bytes of core data

Wow, 25K WUs??

That's crazy. Is the 1k PPH estimate based on all 8 cores at 100% ?

WUs 2681, 2682, 2683 are 25,403 points. If you do one quickly enough there is a sliding bonus scale.

# 2681 completed in 55 hours with the bonus is 57,486 points = 1045.2 pph

Without the bonus qdtools reports 443 pph on the current # 2683 and the 2681 was 459 pph (in 64-bit).

The record appears to be 84,030 PPD. Top -bigadv systems http://foldingforum.org/viewtopic.php?f ... 14#p110479

How are you running it though?

SMP over 8 cores? Virtual machines running SMP? Inquiring minds want to know!

Bigadv Units under Linux and OS X run native with as many cores as you have >=8. The biggest ones I've seen are 32 core machines, 4 Quads with HT. The bonus calculation can be found on the following links. My 1st WU was 25k with a 32k bonus. One WU == 57,486 points. see: http://folding.extremeoverclocking.com/ ... =&u=204311

New bonus plan TRIAL (bigadv) http://foldingforum.org/viewforum.php?f=55 Linux and OS X SMP with 8 or more cores. Fast i7 cpus with HT can make the bonus deadline. (There's a laptop making bonuses ((a $4,950 i7 laptop)))

new release: extra-large work units http://foldingforum.org/viewtopic.php?t=10697

The special bigadv versions of fah6 are located at http://www.stanford.edu/~kasson/folding/ (uses stock mpiexec with the added "-np 16" flag" and stock FahCore_a2.exe. )

Mac Pro with 2 x Xeon E5520s runs 16 cores (I doubt that 8 would be any faster). It runs OS X 10.6.2 and I'm in straight console start "~$ fah6 &". qdtools is used to monitor the Folding.

This is the kind of rig that I was looking at to Fold SMP with the -bigadv switch. I have a full tower box that this would fit in. . .
It should be capable of 16,000 to 24,000 PPD running Linux. The 2.6 kernel has better times.

The 8 physical core requirement of the -bigadv WUs can be met by an i7 chip and a fast one can make the bonus times. There is a special version of fah6 for this. All info can be found here:

new release: extra-large work units http://foldingforum.org/viewtopic.php?t=10697

I decided to stay with Macs and my base 8-core Mac Pro is over 24,000 PPD. Click the EOC Stats link at the top of the page. Definitely hoping the the new 6-core Xeons are 1366 socket and play well with the 5520 chipset. At least then the price for the 4-core 2.93 GHz chips will drop from the current $1,495. {Interestingly Apple's Xeon chips are special and do not have the heat spreader on top of the cpu. Apple's heat sinks get better contact and better heat. dissipation. Anandtech swapped cpus in a Mac Pro for stock cpus with the heat spreaders in place and barely got the heat sinks mounted due to the added thickness, but it ran cool.}

Just for the hell of it I threw in a VelociRaptor 300 GB and made it the boot drive. The Mac Pro is noticeably peppier but it makes a barely measurable difference in frame times. Folding@home is mostly bound up by interprocess communication.. These are the biggest load averages I've ever seen. Network data between the 16 FahCore_a2.exe programs over 127.0.0.1 average just over 85 MB/sec
[Check: uptime 10:07. Network packets out 3,093 G out = 3,093,000,000/10.116=84.9 x 10^3]