S41.06 appears faster than D41.12
S41.06 appears slower than U41.01 but no validation errors yet.
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
A64 3700+ OCd
First result crunched fully with S41.06 came in at 2203s
looks like I'm going to have to start sacrificing small
animals in Akos' name!
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'm runnig S@H (Crunch3r V2.10 SSE3 Intel Pentium4) and S@H (akosf S41.06) on a Prescott with HT turned on.
When I started a new run on S41.06 (two WU in HT), it was some 10% slow on the first half than the pevious S40.04. Then Boinc manager switched to run 1 S@H and 1 E@H app. The second half of the E@H was some 10% faster than the previous S40.04 benchmark (it took 1:10h to the 50% mark and 45min to finish), and the S@H was also some 15% faster!!
Not being an IT expert, my hypotesys is runnig different appplications HT mode allows some sort of debottleneck on the HT CPU, and thus, an improvment in crunching speed.
Can someone confirm this? If so, is there a way to assing permenently a this parallel mode?
You should be able to do this by using Trux client, which you should use because of the credit calibration anyway.
But Trux has more features than this. One is to set "cpu affinity" by project. It should do the trick if you write 1 into the seti scope and 2 into the einstein scope of the truxsoft_prefs.xml. To generate a project scope add:
...
You should be able to do this by using Trux client, which you should use because of the credit calibration anyway.
MrS
Urm... I thought the project-specific CPU affinity feature required Trux Calibrating Client releases higher than the tx36 which is, I think, the highest in current release.
I'd love to be wrong--my Gallatin dearly loves running SETI on one side and Einstein on the other. Currently I limit it to about 15 or 20% SETI resource share because above that the probability of two SETI's running together gets rapidly higher. At 10% and 30 minute switch time, it pretty much never happens.
But Trux has more features than this. One is to set "cpu affinity" by project. It should do the trick if you write 1 into the seti scope and 2 into the einstein scope of the truxsoft_prefs.xml.
So I see S41.06 as much faster than the unfortunate S40.12 on this machine, but still not so fast as the S40.04 it currently runs. I'll try S41.06 on my other machines, and if it looks more promising on them, retry on this machine with another work unit and more carefully controlled conditions.
As to my other machines, the results are far more favorable to S41.06.
700 MHz Pentium III
has been running short units from Z1_0232.5 using S40.12 in about 3732 seconds. First full S41.06 from same set took 3074 seconds--CPU ratio 0.82.
Early indications from my 933 MHZ Pentium III strongly indicate substantial improvement, possibly better than the other two hosts reported here.
1.4 GHz Pentium M (Banias)
has been running long units from Z1_1333.0 using S40.12 in about 5520 seconds. First full S41.06 from same set in 4500 seconds--CPU ratio 0.82
looks like [Sleeper] is using
)
looks like [Sleeper] is using an XP 1800+
S41.06 appears faster than D41.12
S41.06 appears slower than U41.01 but no validation errors yet.
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: looks like [Sleeper] is
)
That sounds correctly. :D
A64 3700+ OCd First result
)
A64 3700+ OCd
First result crunched fully with S41.06 came in at 2203s
looks like I'm going to have to start sacrificing small
animals in Akos' name!
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'm runnig S@H (Crunch3r
)
I'm runnig S@H (Crunch3r V2.10 SSE3 Intel Pentium4) and S@H (akosf S41.06) on a Prescott with HT turned on.
When I started a new run on S41.06 (two WU in HT), it was some 10% slow on the first half than the pevious S40.04. Then Boinc manager switched to run 1 S@H and 1 E@H app. The second half of the E@H was some 10% faster than the previous S40.04 benchmark (it took 1:10h to the 50% mark and 45min to finish), and the S@H was also some 15% faster!!
Not being an IT expert, my hypotesys is runnig different appplications HT mode allows some sort of debottleneck on the HT CPU, and thus, an improvment in crunching speed.
Can someone confirm this? If so, is there a way to assing permenently a this parallel mode?
You should be able to do this
)
You should be able to do this by using Trux client, which you should use because of the credit calibration anyway.
But Trux has more features than this. One is to set "cpu affinity" by project. It should do the trick if you write 1 into the seti scope and 2 into the einstein scope of the truxsoft_prefs.xml. To generate a project scope add:
...
MrS
Scanning for our furry friends since Jan 2002
Using S41.06 on a A64 3000+ @
)
Using S41.06 on a A64 3000+ @ 2.3GHz
Short WUs: under 640 sec.
Long WU: under 2400 sec.
Incredible improvement over S40.12 times of 990 and 2900 seconds.
Akosf, you are a Miracle Worker!
kev
A64 X2 4400+ @2.85GHz, 2x XP 1800+ @2000+
RE: You should be able to
)
Urm... I thought the project-specific CPU affinity feature required Trux Calibrating Client releases higher than the tx36 which is, I think, the highest in current release.
I'd love to be wrong--my Gallatin dearly loves running SETI on one side and Einstein on the other. Currently I limit it to about 15 or 20% SETI resource share because above that the probability of two SETI's running together gets rapidly higher. At 10% and 30 minute switch time, it pretty much never happens.
RE: But Trux has more
)
Does that still utilize the hyperthreading?
Just started with the new app
)
Just started with the new app but I can tell the target graphic moves a lot quicker with S41 than it does with U41. (Prescott with H/T enabled.)
me-[at]-rescam.org
RE: Continuing my practice
)
As to my other machines, the results are far more favorable to S41.06.
700 MHz Pentium III
has been running short units from Z1_0232.5 using S40.12 in about 3732 seconds. First full S41.06 from same set took 3074 seconds--CPU ratio 0.82.
Early indications from my 933 MHZ Pentium III strongly indicate substantial improvement, possibly better than the other two hosts reported here.
1.4 GHz Pentium M (Banias)
has been running long units from Z1_1333.0 using S40.12 in about 5520 seconds. First full S41.06 from same set in 4500 seconds--CPU ratio 0.82