On my dual P3 800 the processing times for the new albert app have been in the range of 4-4.25 hours. With the einstein app processing times were just over 19 hours.
Nothing travels faster than the speed of light with the possible exception of bad news, which obeys its own special laws.
Douglas Adams (1952 - 2001)
hrm..had my first data glitch on EAH and it's on the albert app
1/4/2006 3:30:32 PM|Einstein@Home|Result r1_0770.0__432_S4R2a_1 exited with zero status but no 'finished' file
1/4/2006 3:30:32 PM|Einstein@Home|If this happens repeatedly you may need to reset the project.
1/4/2006 3:30:32 PM||request_reschedule_cpus: process exited
1/4/2006 3:30:32 PM|Einstein@Home|Restarting result r1_0770.0__432_S4R2a_1 using albert version 437
we'll see if it's a random, or a continuing problem.
Crunching times seem stable at 5 hrs 12 - 15 min on a P4 Prescott, 512L2, 2.8ghz 400 ddr.
another glitch
Granted, I have not done a massive amount of EAH WU, but the first two I've ever had came after Albert and on the heals of each other.
1/4/2006 8:29:18 PM|Einstein@Home|Result r1_0770.0__431_S4R2a_0 exited with zero status but no 'finished' file
1/4/2006 8:29:18 PM|Einstein@Home|If this happens repeatedly you may need to reset the project.
1/4/2006 8:29:18 PM||request_reschedule_cpus: process exited
1/4/2006 8:29:18 PM|Einstein@Home|Restarting result r1_0770.0__431_S4R2a_0 using albert version 437
NIMRUTH: You do not have enought ram to support the project or your CPU/OS. I consider this a help thread. I have at least 1 Sempron 3100 running at 2.4ghz. Check the forums. There is a huge amount of info out there on optimizing your system. 256 meg of ram at your rated speed is crippling your rig. If your mobo does not allow you to adjust HTT and change ram dividers, at least get a stick of 512.
My AMD Sempron 2500+ crunches Alberts at a little over 5.5 hours each.
My AMD Athlon XP 2000+ crunches Alberts at a little under 6.5 hours.
My AMD Athlon Tbird running at 1.4 GHz crunches Alberts at 7.0 hours.
My AMD Athlon Tbird running at 1.1 GHz crunches Alberts at 8.5 hours.
Interestingly, my Intel Celeron 1.8 GHz varies widely on the new Alberts.
Sample size of over 15 with the times ranging from 9.25 - 11.0 hours each...
First, my P4 3.2 gig seems to be running the latest alberts in ~7 1/4 hrs -- with two running simultaneously; my laptop pent. M at 1.6 gig runs one at ~6 1/2 hrs; my PIII 866 runs alberts at around ~16 hours or so.
Second, I think the coolest thing is my 866, believe it or not. If I had not found out about E@H earlier last year before I upgraded, that old rig would have been given away or trashed because it was 'too old for anyone to want.' Now, of course, it sits under my bed with a cable running to my router and it crunches einsteins 24/7 until it runs itself into the ground; probably several years or more. It's a cool thing to run older machines for nothing but boinc science!! Hey, every little bit helps.
Even better, I'm hitting up friends who might upgrade in the next year or so, who might otherwise give their machines to the computer store to scrap, to give their machines with an ethernet card to me so I can crunch more einsteins and rosettas, but mostly einsteins. THAT, fellow einsteins, is coolest!
But there is something odd that has been happening that I do need to mention: my RACs have been actually DECREASING since I changed my preferences to run 2 CPUs on my 3.2 gig rig (up from one). How can this be? I should be getting higher RACs running simultaneously, not lower ones.(?)
RAC is heavily biased towards the RATE at which work is returned. So, you are doing more, yet the speed at which the results are returning has slowed (you take longer with each result). This lowers the RAC.
Which is why many of us tell you to not get that wound up with RAC. It is like the gas guage in the car, an indicator of something, just not a very accurate one. If we were doing a pure average (or moving average) you would not have noticed the difference.
a little under 1 hr 50 mins. after reading the hyperthread bit i have a question tho. am benching measured integer over 6400 but einstien only sees one processor but should see 2 right and get an increase in benchmarks? more ram faster for boinc since its cpu intensive? its an oddball so probably just a windoze driver. running xpsp1 220*7 / 1555 mhz, 512 ram. its a clawhammer sse, sse2, etc, 64k l1 1mb l2 socket 754. running boinc 5.2.13 albert 4.37. also, under device mgr / system; numeric data processor (no drivers installed tho driver signing says msoft xp...)? not sure if i shouldnt just leave a good thing running :)
On my dual P3 800 the
)
On my dual P3 800 the processing times for the new albert app have been in the range of 4-4.25 hours. With the einstein app processing times were just over 19 hours.
Nothing travels faster than the speed of light with the possible exception of bad news, which obeys its own special laws.
Douglas Adams (1952 - 2001)
hrm..had my first data glitch
)
hrm..had my first data glitch on EAH and it's on the albert app
1/4/2006 3:30:32 PM|Einstein@Home|Result r1_0770.0__432_S4R2a_1 exited with zero status but no 'finished' file
1/4/2006 3:30:32 PM|Einstein@Home|If this happens repeatedly you may need to reset the project.
1/4/2006 3:30:32 PM||request_reschedule_cpus: process exited
1/4/2006 3:30:32 PM|Einstein@Home|Restarting result r1_0770.0__432_S4R2a_1 using albert version 437
we'll see if it's a random, or a continuing problem.
Crunching times seem stable at 5 hrs 12 - 15 min on a P4 Prescott, 512L2, 2.8ghz 400 ddr.
another glitch Granted, I
)
another glitch
Granted, I have not done a massive amount of EAH WU, but the first two I've ever had came after Albert and on the heals of each other.
1/4/2006 8:29:18 PM|Einstein@Home|Result r1_0770.0__431_S4R2a_0 exited with zero status but no 'finished' file
1/4/2006 8:29:18 PM|Einstein@Home|If this happens repeatedly you may need to reset the project.
1/4/2006 8:29:18 PM||request_reschedule_cpus: process exited
1/4/2006 8:29:18 PM|Einstein@Home|Restarting result r1_0770.0__431_S4R2a_0 using albert version 437
NIMRUTH: You do not have
)
NIMRUTH: You do not have enought ram to support the project or your CPU/OS. I consider this a help thread. I have at least 1 Sempron 3100 running at 2.4ghz. Check the forums. There is a huge amount of info out there on optimizing your system. 256 meg of ram at your rated speed is crippling your rig. If your mobo does not allow you to adjust HTT and change ram dividers, at least get a stick of 512.
Regards-tweakster
With an AMD 939 3500 a gig of
)
With an AMD 939 3500 a gig of ram not in sync (3200) running xp pro svc pack 2, I am getting times of 4 hours on albert! not to shabby ehh?
My AMD Athlon 64 3500+
)
My AMD Athlon 64 3500+ completes the new Alberts right at 4.0 hours.
Also - My AMD Sempron
)
Also -
My AMD Sempron 2500+ crunches Alberts at a little over 5.5 hours each.
My AMD Athlon XP 2000+ crunches Alberts at a little under 6.5 hours.
My AMD Athlon Tbird running at 1.4 GHz crunches Alberts at 7.0 hours.
My AMD Athlon Tbird running at 1.1 GHz crunches Alberts at 8.5 hours.
Interestingly, my Intel Celeron 1.8 GHz varies widely on the new Alberts.
Sample size of over 15 with the times ranging from 9.25 - 11.0 hours each...
A couple of things to
)
A couple of things to mentions for my rigs.
First, my P4 3.2 gig seems to be running the latest alberts in ~7 1/4 hrs -- with two running simultaneously; my laptop pent. M at 1.6 gig runs one at ~6 1/2 hrs; my PIII 866 runs alberts at around ~16 hours or so.
Second, I think the coolest thing is my 866, believe it or not. If I had not found out about E@H earlier last year before I upgraded, that old rig would have been given away or trashed because it was 'too old for anyone to want.' Now, of course, it sits under my bed with a cable running to my router and it crunches einsteins 24/7 until it runs itself into the ground; probably several years or more. It's a cool thing to run older machines for nothing but boinc science!! Hey, every little bit helps.
Even better, I'm hitting up friends who might upgrade in the next year or so, who might otherwise give their machines to the computer store to scrap, to give their machines with an ethernet card to me so I can crunch more einsteins and rosettas, but mostly einsteins. THAT, fellow einsteins, is coolest!
But there is something odd that has been happening that I do need to mention: my RACs have been actually DECREASING since I changed my preferences to run 2 CPUs on my 3.2 gig rig (up from one). How can this be? I should be getting higher RACs running simultaneously, not lower ones.(?)
(Click for detailed stats)
RAC is heavily biased towards
)
RAC is heavily biased towards the RATE at which work is returned. So, you are doing more, yet the speed at which the results are returning has slowed (you take longer with each result). This lowers the RAC.
Which is why many of us tell you to not get that wound up with RAC. It is like the gas guage in the car, an indicator of something, just not a very accurate one. If we were doing a pure average (or moving average) you would not have noticed the difference.
a little under 1 hr 50 mins.
)
a little under 1 hr 50 mins. after reading the hyperthread bit i have a question tho. am benching measured integer over 6400 but einstien only sees one processor but should see 2 right and get an increase in benchmarks? more ram faster for boinc since its cpu intensive? its an oddball so probably just a windoze driver. running xpsp1 220*7 / 1555 mhz, 512 ram. its a clawhammer sse, sse2, etc, 64k l1 1mb l2 socket 754. running boinc 5.2.13 albert 4.37. also, under device mgr / system; numeric data processor (no drivers installed tho driver signing says msoft xp...)? not sure if i shouldnt just leave a good thing running :)