Your'e right Honza. I was running S41.06 and not U41.01 as I thought. I did try U41.01 and according to my preliminary calculations a wu takes ~5300s. That is considerably slower than S41.06 on the same hardware.
Preliminary crunch times on Yonah with U-series: ~80% in one hors, which makes it ~75 mins or 4500 sec for z1_1379.
Will run S4106 for other z1_1379 for comparation.
Good news.
I remade the FISTTP code without optimize its environment and got a VALID result with better time. It's interesting. That was also interesting that U41 was slower on some machines than S41.
Good news.
I remade the FISTTP code without optimize its environment and got a VALID result with better time. It's interesting. That was also interesting that U41 was slower on some machines than S41.
I am glad to see that you have gotten the FISTTP code (I assume part of SSE3?) working well. It seems that the faster Intel computers work better with S41 while the faster AMD hosts prefer U41. Someone please correct me if I'm wrong.
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
I am glad to see that you have gotten the FISTTP code (I assume part of SSE3?) working well. It seems that the faster Intel computers work better with S41 while the faster AMD hosts prefer U41. Someone please correct me if I'm wrong.
>Our AMD X2's were faster using U41 (our 2 lonely Celerons should do much better as well!). Looking forward to Akosf's revisions.....Cheers, Rog.
good morning
akos, you don't have to say sorry if there's a problem with one of your releases.
you did a lot for the einstein@home community and we can't thank you enough for that.
if someone gives you a bad rating because he had a problem with one of your executables he's an idiot and should be sentenced to use a standart albert until the end of his life ;)
cu abschaum
good morning
akos, you don't have to say sorry if there's a problem with one of your releases.
you did a lot for the einstein@home community and we can't thank you enough for that.
if someone gives you a bad rating because he had a problem with one of your executables he's an idiot and should be sentenced to use a standart albert until the end of his life ;)
cu abschaum
I am glad to see that you have gotten the FISTTP code (I assume part of SSE3?) working well. It seems that the faster Intel computers work better with S41 while the faster AMD hosts prefer U41. Someone please correct me if I'm wrong.
While I do not have an Intel SSE3 candidate, certainly the U client went through the roof on my AMD X2. On that part I definitely agree. But I don't have any results returned to see if they are valid. Have two results pending for U client then switched back to S client since of users' feedback.
:
your thoughts - the ways :: the knowledge - your space
:
Sorry to post again, but I had no answer (maybe no one knows the right answer ?)
What is this "r" beginning the 2nd line of the stderr_txt?? (This letter seems to appear only on the last AkosF versions optimized clients!)
Could it cause (or result from) wrong or invalid results ?
Your'e right Honza. I was
)
Your'e right Honza. I was running S41.06 and not U41.01 as I thought. I did try U41.01 and according to my preliminary calculations a wu takes ~5300s. That is considerably slower than S41.06 on the same hardware.
Preliminary crunch times on
)
Preliminary crunch times on Yonah with U-series: ~80% in one hors, which makes it ~75 mins or 4500 sec for z1_1379.
Will run S4106 for other z1_1379 for comparation.
Akosf wrote here
)
Akosf wrote here :
link
Hello! Don't use U41.01 because it doesn't work well. Sorry...
edit: I will look after and correct it.
[
Good news. I remade the
)
Good news.
I remade the FISTTP code without optimize its environment and got a VALID result with better time. It's interesting. That was also interesting that U41 was slower on some machines than S41.
RE: Good news. I remade the
)
I am glad to see that you have gotten the FISTTP code (I assume part of SSE3?) working well. It seems that the faster Intel computers work better with S41 while the faster AMD hosts prefer U41. Someone please correct me if I'm wrong.
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: I am glad to see that
)
>Our AMD X2's were faster using U41 (our 2 lonely Celerons should do much better as well!). Looking forward to Akosf's revisions.....Cheers, Rog.
good morning akos, you don't
)
good morning
akos, you don't have to say sorry if there's a problem with one of your releases.
you did a lot for the einstein@home community and we can't thank you enough for that.
if someone gives you a bad rating because he had a problem with one of your executables he's an idiot and should be sentenced to use a standart albert until the end of his life ;)
cu abschaum
RE: good morning akos, you
)
Let me say this is a very good posting!
cu,
Michael
RE: I am glad to see that
)
While I do not have an Intel SSE3 candidate, certainly the U client went through the roof on my AMD X2. On that part I definitely agree. But I don't have any results returned to see if they are valid. Have two results pending for U client then switched back to S client since of users' feedback.
:
your thoughts - the ways :: the knowledge - your space
:
Sorry to post again, but I
)
Sorry to post again, but I had no answer (maybe no one knows the right answer ?)
What is this "r" beginning the 2nd line of the stderr_txt?? (This letter seems to appear only on the last AkosF versions optimized clients!)
Could it cause (or result from) wrong or invalid results ?
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.