Message boards :
Number crunching :
P4 vs G4
Message board moderation
Author | Message |
---|---|
Mntbighker Send message Joined: 1 Feb 00 Posts: 9 Credit: 62,517 RAC: 0 |
I have a Powerbook G4 1.33 and a P4 2.8 with HT running Linux. From the reported stats page on the Seti site I got these benchmark numbers... Float: G4 - 984 million / P4 - 331 million Integer: G4 - 3955 million / P4 - 680 million I am running the standard available clients. Obviously there are some glaring differences in optimization. I am interested in comments. Is the G4 BOINC menubar client code well optimized for altivec? I also had no luck building any of the recent cvs linux boinc clients on my P4. I get these errors mostly.. libtool: link: cannot find the library `' make[2]: *** [crypt_prog] Error 1 make[2]: Leaving directory `/home/moorcroft/My Downloads/boinc/lib' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/moorcroft/My Downloads/boinc' make: *** [all] Error 2 The sanity test works prior to the build. My P4 is my primary system with horspower and spare cycles (runs IMAP for me) so I would like to get it working properly. Mark Moorcroft (Mac geek) ELORET Corp. - NASA/Ames RC AIM - ssevenup YIM - mntbighker |
N/A Send message Joined: 18 May 01 Posts: 3718 Credit: 93,649 RAC: 0 |
Welcome to the Fora! BOINC Menubar hasn't been AltiVecced at all, but Google for Team MacNN and you'll find a well optimized copy. As for the benchmarks, well... weird stuff happens. I posted my specs are in the Crunching Apples thread, and pointed out that a VirtualPC environment benchmarked higher than the host! I haven't ever used CVS, so I can't help you there. This forum is chock-full-of-BOINCers who'll know |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.