Beta App including SSE2/3 optimisations

Udo
Udo
Joined: 19 May 05
Posts: 203
Credit: 8945570
RAC: 0

RE: I have a Pentium II (no

Message 44617 in response to message 44616

Quote:
I have a Pentium II (no SSE) running Linux and BOINC 5.4.9. My ACE is about 0.81 on both SETI and Einstein, while DCF is 0.82 on SETI app 5.12 and 0.59 on Einstein app 4.16. Does this mean that Einstein runs better than SETI on my CPU? I gave a 3:1 relative share in favour of Einstein.
Tullio

Hello,
the DCF is an factor which describes how much a science application differs from its own standard.
Example:
Lets assume the initial Einstein S4 application had an DCF of 1
Then the estimated end time will be exactly met.
Akos optimized the application, now the computation is done 5 times faster. Then the DCF will drop to 0.2 - after that the estimated end time will also be correct.
Reducing DCF is done in small steps (each result finishing earlier than estimated reduces DCF a little bit) but increasing DCF is done in big steps (results finishing later than estimated increase DCF to that value that matches to the current processing time.

After all - you cann't compare DCFs between different projects!
Udo

Udo

tullio
tullio
Joined: 22 Jan 05
Posts: 2118
Credit: 61407735
RAC: 0

RE: Hello, the DCF is an

Message 44618 in response to message 44617

Quote:


Hello,
the DCF is an factor which describes how much a science application differs from its own standard.
Example:
Lets assume the initial Einstein S4 application had an DCF of 1
Then the estimated end time will be exactly met.
Akos optimized the application, now the computation is done 5 times faster. Then the DCF will drop to 0.2 - after that the estimated end time will also be correct.
Reducing DCF is done in small steps (each result finishing earlier than estimated reduces DCF a little bit) but increasing DCF is done in big steps (results finishing later than estimated increase DCF to that value that matches to the current processing time.

After all - you cann't compare DCFs between different projects!
Udo


Udo, thanks for the explanation. I am running also QMC@home, where my DCF is 1.5. This means that a WU shall be completed later than estimated. I see that in that project the CPU time/clock tima ratio is lower than in the other two.Also CPU efficiency is lower. Thanks again.
Tullio

Ziran
Ziran
Joined: 26 Nov 04
Posts: 194
Credit: 356403
RAC: 1626

RE: RE: That's odd.

Message 44619 in response to message 44602

Quote:
Quote:
That's odd. Akos's s4 sse3 app was significantly (~20% IIRC) faster than the sse1/2 variants.

1. On the same CPU?

I looked at my old posts about speeds with Akos S4 applications.

On my Sempron 3000+ I had the following speeds with these different versions.

U41.01 2,660.72 (SSE3)
S41.06 2,960.84 (SSE)
S40.12 3,779.44 (SSE)

Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.