broadbandit
Premium
join:2000-07-21
Morrisville, PA
clubs:
| [E@H] Confused about machine crunch times
Today, I'm looking at the log for an AMD 1800+ using Trux's calibrating client set to calibrate ALL. Both my s@h and e@h seem to be crunching in the low 2 hours (s@h 2:40)(e@h 1:52) the last that I've seen, Thursday night, my P3866Mhz was crunching using the same client and turning results in for seti at about 4hours, e@h it was taking 8hours! I've checking my fast rigs at work at they also seem to be working on e@h at 8hours each! I figured that the work units handed out for e@h were just larger. So, with that being said, then why is my AMD 1800 (1.5ghz) crunching through these in under 2hours? Confused!
--
It ain't easy being greazy..
Load "*",8,1 |
|
m00kie
Yes, Nurse Ratched
Premium
join:2000-09-11
| I've had 2 hour wus and 8 hr wus with Einstein. Seems to be typical. Their work units seem to vary much more than Seti does.
--
Expand your universe... Join BBR Team Starfire |
|
Xaak
You'll find me at T S W B.org
Premium
join:2002-06-19
| reply to broadbandit
Yeah, we're used to relatively consistent wus from Seti. Einstien has large differences, and so does Rosetta, which I'm playing around with now. I get some Rosetta wus that take 6-7 hours, and others at closer to 2 hours, with claimed credits related to crunch times.
Also, keep in mind that while Intel is king on Seti, especially with the optimized clients, other projects may take better to other processors due to the way they're written and optimized.
--
Xaak
Xaak Consulting, LLC |
|
parkut
Crunch Addict
Premium
join:2001-12-15
Harrison Township, MI
clubs:
 ·AT&T Midwest
 ·WOW Internet and C..
| reply to broadbandit
One of my Debian Linux XP18 systems averages a result every 7606 seconds - 2hrs 7 minutes (average of 108 results)
Trux's calibrating boinc manager has nothing to do with the client projects actual crunch time.
The Optimized Seti@Home client from Crunch3r only works on SAH, and not on any other project.
Trux's calibrating will fiddle with the reported crunch time and credits, but not with the ACTUAL crunch time.
--
Hello, my name is Bill and I'm a crunchaholic...
Proud to be the current host of Crunchenstein #1, #3, #4 and #5 |
|
broadbandit
Premium
join:2000-07-21
Morrisville, PA
clubs:
| reply to broadbandit
Thanks everyone! makes sense now. I thought that my P4 2.8Ghz's weren't getting enough gas to chomp through E@H wu like the AMD 1.5Ghz was. 
--
It ain't easy being greazy..
Load "*",8,1 |
|
72245156
TSWB.org
Premium,ExMod 2000-04
join:2000-07-11
Winnipeg, MB
| Einstein works better with AMD. Barton 2500 stock speed does Einstein long WU's around 5 hours, short ones around 1.5 hours.
--
Join BroadbandReports.com's SETI@Home Team
Don't let your computer's idle time go to waste! |
|
broadbandit
Premium
join:2000-07-21
Morrisville, PA
clubs: | reply to broadbandit
And I'da though Albert would have been an Intel guy.. Thanks everyone! |
|
