Posts by Dagorath

1) Message boards : Number crunching : PS3 (Message 1187) Posted 28 Dec 2007 by Dagorath Post: That was possible for Folding@home so why not for BOINC. It is possible but for many projects (perhaps including Milkyway) there is no incentive to port their application to PS3. There is no incentive because their application will not run any faster on PS3 than it does on a PC. PS3 is designed to do the kinds of calculations required for graphics and rendering 3D worlds at amazing speed. Folding and YoYo science application also use the type(s) of calculations PS3 can do very fast. As for the kinds of calculations required by your average BOINC project, PS3 performs those calculations no faster than your average PC so those projects have no incentive to port to PS3.
2) Message boards : Number crunching : AMD64 (Message 1122) Posted 19 Dec 2007 by Dagorath Post: Even though there's an official AMD64 client for Linux, it refers to too many dynamic libraries and requires a fairly recent Linux setup to run on. So, one more time, I'm making available the AMD64 Linux client here. It refers to a minimal set of standard dynamic libraries whose version requirements should be satisfied by Linux systems up to 2 or 3 years old, however it was built with a fairly recent version of GCC, 4.1.2. The drill's still the same: Untar the file "boinc_5.10.30_x86_64-pc-linux-gnu.tgz" at an existing BOINC working directory. Read the file "README.x86_64-pc-linux-gnu". The official AMD64 Windows client can be found here. For more information, see the BoincStats Forum. HTH I'm pretty sure we have access to AMD64 machines (one of our clusters run this i believe). I'll try and get our labstaff to put boinc on it so i can compile a binary for this architecture. A 64-bit binary would be cool. As you probably know already, you'll see a speed increase if the algorithm uses lots of integer calculations and it may not cost you a lot of man-hours to get that benefit (perhaps just a few minor tweaks and recompile). What you might not know is that there are other ways to get speed increases with 64-bit though implementing those methods might take additional work. Only you and your team can know if the potential speed boost is worth the effort but you might want to look into using SSE/SSE2 functions and registers under 64-bit to speed up floating point ops. IIRC, the SIMAP project got a decent speed boost by taking advantage of SSE/SSE2 for one of their predominately floating point apps. I am no expert on optimizations for 64-bit but Augustine, Crunch3r and others are. They've helped other projects and might have time to give your team some pointers too. Meanwhile, your 32-bit app(s) are doing well on my 64-bit Linux machines. Keep up the good work.