This client identify my cpu as "03.05.2007 13:55:44||Processor: 2 GenuineIntel Intel(R) Pentium(R) 4 CPU 3.20GHz [x86 Family 15 Model 4 Stepping 1] [fpu tsc sse sse2 mmx] " wich is not correct, cause i have sse3. I have an IBM "branded" thinkcenter with a intel pentium 4 540 cpu. The core of cpu is known as Prescott.
So it identify my cpu incorrectly and downloads incorrect modules, speed is really slow.

P.s. cpu-z, and other report that i have sse3 ;-)
pp.s. sorry for a bad English.

If I remember right...

Windows doesn't know about sse3. BOINC gets its information on processor features from Windows. Thus, BOINC can only report what Windows tells it.

Off the top of my head, I don't know of any projects that send apps based on CPU features. Docking makes use of some of the information for their use of homogeneous redundancy (making sure that workunits go to similar computers because of differences in floating point math).

What projects are you running? How slow is slow (links to workunits from your computer and workunits from similar computers would help)? Is there anything else taking up CPU cycles? Thermal throttling?

Just throwing out ideas...

---

Kathryn :o)

If I remember right...

Off the top of my head, I don't know of any projects that send apps based on CPU features. Docking makes use of some of the information for their use of homogeneous redundancy (making sure that workunits go to similar computers because of differences in floating point math).

May be i`am not correct, but some time ago "Einstein" and "Seti" users had precompiled versions of executables that were optimized to specific processor http://einstein.phys.uwm.edu/forum_thread.php?id=3898

What projects are you running? How slow is slow (links to workunits from your computer and workunits from similar computers would help)? Is there anything else taking up CPU cycles? Thermal throttling? Just throwing out ideas...

Seti and Einstein. When i connect an exe file is downloaded and it is called einstein_S5R2_4.17_windows_intelx86.exe.
About slow, my notebook (ibm r51 Pentium-M 1.4ghz) at home calculated Einstein to 7% for a time of 2 hours. And here i have 1.7% for a time of 1 hour 40 minutes.

Seti's is a generic application. If you want the optimized version, check out the sticky at the top of the Number Crunching forum.

Einstein is a different beast. You're right, their science app used to have some optimization in it. They're starting up a new phase of data analysis and the app has no optimization in it. It's actually turning out to be a bit buggy as well. There's multiple threads discussing the increase in work unit run times on their boards.

---

Kathryn :o)