Something is puzzling me. I have a Dual 1.25 GHz PowerPC G4, with 1 MB L3 Cache per processor, 768 MB DDR SDRAM of memory, running under Mac OS X 10.4.1. I dowloaded yesterday the 6.12.35 version of Boinc Manager (Darwin 8.11.0) and started to run SETI@home.
I must say I was disappointed in terms of speed. Indeed, each processor, running at 100%, is taking some 45 hours (!) to process a block of data... However, when I run a CPU I get the following results:
- 797 floating point MIPS (Whetstone) per CPU
- 2437 integer MIPS (Dhrystone) per CPU
It is not an awesome speed, I agree, but I've seen worse...
I also have a more modern machine, a portable PC with a GenuineIntel Intel(R) Core(TM) i7 CPU Q 720 @ 1.60GHz [Family 6 Model 30 Stepping 5](8 processors) running SETI@home. For this machine the benchmark results are:
- 1492 floating point MIPS (Whetstone) per CPU
- 7285 integer MIPS (Dhrystone) per CPU
Only the double per CPU as for the Mac, but in this case, each processor finishes a block of SETI data in approximately 2:30h (the NVIDIA GPU takes around 40 minutes).
So, anybody can tell me what's wrong with my Mac?
Thanks in advance,
Thank you for your attention to my problem. Answering to your question, it is an estimate, since none of the two calculations has yet finished. I started SETI@home on my Mac yesterday, stopped overnight and restarted it again this afternoon. One of the processes is running for 10:55 hours and completed only 28.25% of the job. The other in running for 10:03 hours and solely completed 19.18% of the task. Very odd, indeed...