First half d41.12/u41.01 result came back invalid.
could have been either app. Will post again for all
U41.01 WU.
The reason some people get results back faster than others is because they
have a large enough cache that the other hosts in the quorum
finnish before they do.
Dave
There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman
How does everyone else get valid/invalid results seemingly within minutes? It always takes me hours or days to get results.
Those folks have a 4+ day backlog of WUs, so when they finally return WUs, they are the 3rd result, completing the quorum and getting validation almost immediately. Not to mention, Akos' clients only take minutes to crunch . . . :)
EDIT: Oops. Beaten to the punch. What he said, he he.
How does everyone else get valid/invalid results seemingly within minutes? It always takes me hours or days to get results.
I'd say just plain luck. And / or being in a group with other optimized cruchers. Because one thing is for sure: no halfway modern 24/7 machine can keep up a cache of einstein WUs with these clients and the 32WU/day/CPU quota (except if they're using BOINC Studio).
When I look into my results it's not uncommon to see 2 co-crunchers with optimized apps as well. Then we all finish our work within one day -> fast feedback. Though it's more common to have one regular client and one optimized co-cruncher. Then it depends on the cache setting of the *slow* machine.
First half d41.12/u41.01 result came back invalid.
could have been either app. Will post again for all
U41.01 WU.
two more invalid with u41.01
there was a definite speed increase, but
the invalid results negate it.
gone back to d41.12
Dave
There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman
RE: How does everyone else
)
Bad news travell fast? :-)
I found U41 not suitable for Intels - short WU took ~32 minutes, S4106 ~16.
A64 3700+ First half
)
A64 3700+
First half d41.12/u41.01 result came back invalid.
could have been either app. Will post again for all
U41.01 WU.
The reason some people get results back faster than others is because they
have a large enough cache that the other hosts in the quorum
finnish before they do.
Dave
There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman
Newer Semprons also have
)
Newer Semprons also have SSE3.
Results with my Sempron 3000+
U41.01 2,660.72
S41.06 2,960.84
S40.12 3,779.44
I guess we are one step closer to that black hole forming in Hungary. :-)
I am impressed as always, Akosf.
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.
All short WUs are invalid! I
)
All short WUs are invalid!
I got no "long" WUs to test on.
(Opteron 144)
RE: How does everyone else
)
Those folks have a 4+ day backlog of WUs, so when they finally return WUs, they are the 3rd result, completing the quorum and getting validation almost immediately. Not to mention, Akos' clients only take minutes to crunch . . . :)
EDIT: Oops. Beaten to the punch. What he said, he he.
Congratulations again to
)
Congratulations again to AkosF for his wonderful work on these einstein clients!
Anyway, one question looking the result of my 2 last wu:
http://einsteinathome.org/task/27309537 (invalid ! D41.12->U41.01)
http://einsteinathome.org/task/27312708 (complete U41.01 - stderr reported below) 5.2.2
2006-05-01 22:55:07.4531 [normal]: Optimised by akosf U41.01 --> 'projects/einstein.phys.uwm.edu/albert_4.37_windows_intelx86.exe'.
r2006-05-01 22:55:07.4687 [normal]: Started search at lalDebugLevel = 0
2006-05-01 22:55:08.7187 [normal]: Checkpoint-file 'Fstat.out.ckp' not found.
2006-05-01 22:55:08.7187 [normal]: No usable checkpoint found, starting from beginning.
2006-05-01 22:56:04.3750 [normal]: Fstat file reached MaxFileSizeKB ==> compactifying ... done.
2006-05-01 23:33:35.2656 [normal]: Fstat file reached MaxFileSizeKB ==> compactifying ... done.
2006-05-01 23:51:05.6718 [normal]: Search finished successfully.
[/i]
What is this "r" beginning the 2nd line ??
Could it cause wrong or invalid results ?
Anyway! ... Again great congratulations from the whole team "Est" from France to you AkosF !
RE: How does everyone else
)
I'd say just plain luck. And / or being in a group with other optimized cruchers. Because one thing is for sure: no halfway modern 24/7 machine can keep up a cache of einstein WUs with these clients and the 32WU/day/CPU quota (except if they're using BOINC Studio).
When I look into my results it's not uncommon to see 2 co-crunchers with optimized apps as well. Then we all finish our work within one day -> fast feedback. Though it's more common to have one regular client and one optimized co-cruncher. Then it depends on the cache setting of the *slow* machine.
MrS
Scanning for our furry friends since Jan 2002
RE: A64 3700+ First half
)
two more invalid with u41.01
there was a definite speed increase, but
the invalid results negate it.
gone back to d41.12
Dave
There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman
First validated result was
)
First validated result was invalid. Started with S41.06 finished with U41.01
http://einsteinathome.org/task/27058746
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.
RE: Running U41 on Intel D
)
S41.06 is the best I've found for the 820. Running 3602 seconds per.