A few observation from my experience over the last few weeks.
Since the 6th Feb 19:03 I have only been doing two projects Seti at 75% and Einstein at 25% resource share .
Taking the figures from Boinc Stats and only using the figures from the 8th. I have been granted 65,500 for Seti, effectively on three cores, and 13,924 for Einstein, on one core. (the Seti figure is not an estimate it is 1605614 - 1540114)
But on Seti I use the optimised applications and I choose to do AP tasks and if no AP tasks are available, then will accept MB tasks.
The observation at Seti is that the optimised applications for Multibeam, Astropulse V5.00 and Astropulse 5.03 all give differing credit bonuses compared to the default applications.
So any test at Seti needs to say which applications are used. And preferably use different trial periods for each application used. So that default MB can be compared to optimised MB and similar tests for AP applications.
I would suggest looking at the maturity of the Seti apps that only the MB apps be used.
My personal view on AP is that there may be more changes as the new app, V5.03, slows down by about 20%, from ~12 hrs to >14 hrs, if there is large amounts of inserted radar blanking noise in the data.
It would certainly be interesting to run the 'parallel' test as well as the 'serial' test, and compare the credit total (sum of all four projects) for the two months. I suspect that BOINC will turn out to be more productive overall running different projects on the different cores (less competition for resources), but that's open for debate/experiment too.
I am pretty sure that I can stare at it, chew my nails but not touch it for a month. Can’t guarantee two months, tho. :oP
There's no need for that :-) I think Richard meant to compare the results of one month to those of the other. So there's only a one-month period to chew nails (though two times, but you could take a rest in between ;-).
Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)
OK, Just one caveat here (as a reminder). You can't just use the pending credit total for EAH in your calculation if you have any at the end of the week, since the granted won't be the same as the claimed for them.
I guess you'll have to wait until they post to present the bottom line. ;-)
OK, Just one caveat here (as a reminder). You can't just use the pending credit total for EAH in your calculation if you have any at the end of the week, since the granted won't be the same as the claimed for them.
I guess you'll have to wait until they post to present the bottom line. ;-)
Alinator
Or you could just multiply the pending total by 3.35
I've just checked the ratio of claim to award on 200 recent tasks (well, 184 after I'd taken out the pendings)
Average --- 3.350001807
Maximum --- 3.35045184
Minimum --- 3.349603417
New Einstein total including pending (guesstimated at 3.5X multiplier) over 168 hours: 90 618
Credit gain: 19 658
117.01 credit/hr
Now crunching Prime Grid (PPS-LLR)
Starting PG credit: 32 203
@ Alinator: the main reason that I started off with Seti and Einstein is that there is a couple weeks of crunching (PG and MW) and time for those pending credits to settle out. Weekly numbers will be adjusted as required at the end.
See ya next Monday. :o)
Your prototypical Generation 'Y' slacker, and damn proud of it.
The pending credit multiplier is 3.35, not 3.5. You've over-guesstimated your pending if you used 3.5.
Quote:
Weekly update.
New Einstein total including pending (guesstimated at 3.5X multiplier) over 168 hours: 90 618
Credit gain: 19 658
117.01 credit/hr
Now crunching Prime Grid (PPS-LLR)
Starting PG credit: 32 203
@ Alinator: the main reason that I started off with Seti and Einstein is that there is a couple weeks of crunching (PG and MW) and time for those pending credits to settle out. Weekly numbers will be adjusted as required at the end.
77.15/hr over 164 hours (due to 4 hour power outage)
Now crunching Milky Way, project supplied apps.
Starting 50 818
@ RandyC: Re pending. By next week both Seti and Einstein pending should have pretty much been sorted out and I will recalculate with more accurate numbers then.
See ya in a week. :o)
Your prototypical Generation 'Y' slacker, and damn proud of it.
So, yeah, stuff happens. And in my case, a little family emergency had me flying to the other side of the country for a week and a half, so I messed up the last MW test.
I had my neighbour re-start some other projects, so when those crunch out in 3-5 days, I will re-start the MW test.
Sorry about that. So, I should have MW results in a week to ten days, depending on when I get the new MW test started.
Stay tooned. . .
Your prototypical Generation 'Y' slacker, and damn proud of it.
A few observation from my
)
A few observation from my experience over the last few weeks.
Since the 6th Feb 19:03 I have only been doing two projects Seti at 75% and Einstein at 25% resource share .
Taking the figures from Boinc Stats and only using the figures from the 8th. I have been granted 65,500 for Seti, effectively on three cores, and 13,924 for Einstein, on one core. (the Seti figure is not an estimate it is 1605614 - 1540114)
But on Seti I use the optimised applications and I choose to do AP tasks and if no AP tasks are available, then will accept MB tasks.
The observation at Seti is that the optimised applications for Multibeam, Astropulse V5.00 and Astropulse 5.03 all give differing credit bonuses compared to the default applications.
So any test at Seti needs to say which applications are used. And preferably use different trial periods for each application used. So that default MB can be compared to optimised MB and similar tests for AP applications.
I would suggest looking at the maturity of the Seti apps that only the MB apps be used.
My personal view on AP is that there may be more changes as the new app, V5.03, slows down by about 20%, from ~12 hrs to >14 hrs, if there is large amounts of inserted radar blanking noise in the data.
RE: RE: It would
)
There's no need for that :-) I think Richard meant to compare the results of one month to those of the other. So there's only a one-month period to chew nails (though two times, but you could take a rest in between ;-).
Gruß,
Gundolf
Computer sind nicht alles im Leben. (Kleiner Scherz)
Just a weekly update: New
)
Just a weekly update:
New SAH total including new pending over 168 hours: 338 278
Credit gain: 12 046
71.7 credit/hr
Now crunching Einstein.
Starting Einstein credit: 70 960
See you in a week.
Your prototypical Generation 'Y' slacker, and damn proud of it.
Help feed the world's hungry. Free Rice.
OK, Just one caveat here (as
)
OK, Just one caveat here (as a reminder). You can't just use the pending credit total for EAH in your calculation if you have any at the end of the week, since the granted won't be the same as the claimed for them.
I guess you'll have to wait until they post to present the bottom line. ;-)
Alinator
RE: OK, Just one caveat
)
Or you could just multiply the pending total by 3.35
I've just checked the ratio of claim to award on 200 recent tasks (well, 184 after I'd taken out the pendings)
Weekly update. New
)
Weekly update.
New Einstein total including pending (guesstimated at 3.5X multiplier) over 168 hours: 90 618
Credit gain: 19 658
117.01 credit/hr
Now crunching Prime Grid (PPS-LLR)
Starting PG credit: 32 203
@ Alinator: the main reason that I started off with Seti and Einstein is that there is a couple weeks of crunching (PG and MW) and time for those pending credits to settle out. Weekly numbers will be adjusted as required at the end.
See ya next Monday. :o)
Your prototypical Generation 'Y' slacker, and damn proud of it.
Help feed the world's hungry. Free Rice.
The pending credit multiplier
)
The pending credit multiplier is 3.35, not 3.5. You've over-guesstimated your pending if you used 3.5.
Seti Classic Final Total: 11446 WU.
Weekly update: New PG
)
Weekly update:
New PG total including pending: 94 856
Old total (typo last week, oops) 82 203
Credit gain: 12 653
77.15/hr over 164 hours (due to 4 hour power outage)
Now crunching Milky Way, project supplied apps.
Starting 50 818
@ RandyC: Re pending. By next week both Seti and Einstein pending should have pretty much been sorted out and I will recalculate with more accurate numbers then.
See ya in a week. :o)
Your prototypical Generation 'Y' slacker, and damn proud of it.
Help feed the world's hungry. Free Rice.
So, yeah, stuff happens. And
)
So, yeah, stuff happens. And in my case, a little family emergency had me flying to the other side of the country for a week and a half, so I messed up the last MW test.
I had my neighbour re-start some other projects, so when those crunch out in 3-5 days, I will re-start the MW test.
Sorry about that. So, I should have MW results in a week to ten days, depending on when I get the new MW test started.
Stay tooned. . .
Your prototypical Generation 'Y' slacker, and damn proud of it.
Help feed the world's hungry. Free Rice.
All right, starting MW again.
)
All right, starting MW again. 6:30 pm PST, April 5.
Starting credit: 67 176.
Your prototypical Generation 'Y' slacker, and damn proud of it.
Help feed the world's hungry. Free Rice.