The chances of getting any new development in the PRP side of the project is slim to none I would say (based on past experience).
But on the sieve side of things it's possible that the SR2Sieve client could be modified. Might be worth having a look around on some other projects forums and trying to track down the guy who's developing it.
Sieving sounds like the best option for the PS3.
The main PPC chip could be responsible for the candidate prime sieving and each SPH/BSGS test is handed out to an individual cell. The BSGS code would have to be adapted to run within the memory constraints (256KB per cell) but I can think of some possible ways for this to work.
Anyone with a PS3 running Linux (and spu-gcc etc installed) willing to give me an account and ssh access so I can fiddle around with some ideas?
Quad 2.5GHz G5 PowerMac. Mmmmm.
My Current Sieve Progress: http://www.greenbank.org/cgi-bin/proth.cgi
I'm currently working with Bryan ...
http://www.rieselsieve.com/forum/viewtopic.php?t=1136
I'd love to see if a SOB client could run and eventually optimized to use all 7 cores.
Send me an email IronBits at cox.net and I'll open up a hole in the router and let you play to your hearts content.
You'll need ssh of course ... When I get home this evening, I'll open the door and let you in.
Yellow Dog Linux and gcc is already installed etc.
Excellent, you'll have email in about 5 minutes!
Quad 2.5GHz G5 PowerMac. Mmmmm.
My Current Sieve Progress: http://www.greenbank.org/cgi-bin/proth.cgi
Posting Permissions
Reply With Quote
