Posted 9 Jan 2015 by
Conan
Thanks Rebirther for fixing up my computer totals, all adding up OK now.
Conan
Posted 9 Jan 2015 by
Conan
For the ""correction"" to be done properly, not only does the overall total need correcting (which mine has been), but the Computer totals has to be corrected as well.
My computers still add up to over 160,000 points, yet my overall is only about 56,000.
Could not all those work units be re-validated so their value changes from 900 to 5 and then the computers will correct themselves?
Thanks
Conan
I have tried to revalidate but BOINC does it wrong, the value of 5 is there for a few seconds and will add it again to credits total after that its switching to 900 again and count it too.
The RAC will normalise again if you are crunching some new WUs. If you need a change in the host tell me the number. I could not correct all the things I need it this will take too long.
Thanks for the reply Rebirther.
I did a bit of maths and as all my short task that I processed are still on my computer, I was able to work out my correct totals.
Current User Total (all 4 computers) = 56,681
Host 424 current total = 33,479 credits
Processed 26 affected short tasks, 26 x 900 = 23,400
Corrected value is 26 x 5 = 130
So Host 424 should have 33,479 - 23,400 + 130 = 10,209 credits
Host 425 current total = 41,416 credits
Processed 31 affected short tasks, 31 x 900 = 27,900
Corrected value is 31 x 5 = 155
So Host 425 should have 41,416 - 27,900 + 155 = 13,671 credits
Host 426 current total = 47,578 credits
Processed 32 affected short tasks, 32 x 900 = 28,800
Corrected value is 32 x 5 = 160
So Host 426 should have 47,578 - 28,800 + 160 = 18,938 credits
Host 427 current total = 37,708 credits
Processed 26 affected short tasks. 26 x 900 = 23,400
Corrected value = 26 x 5 = 230
So Host 427 should have 37,708 - 23,400 + 130 = 14,438 credits
My totals now are 10,209 + 13,671 + 18,938 + 14,438 = 57,256 credits
The difference between my new total of 57,256 and the one you show for me (56,681) is the 5 credit value of the work units that you corrected.
You appear to have removed all 900 credits per WU but did not give back the 5 credit value each WU should of been given.
If this does not make sense please let me know.
Thanks
Conan
Posted 9 Jan 2015 by
Conan
For the ""correction"" to be done properly, not only does the overall total need correcting (which mine has been), but the Computer totals has to be corrected as well.
My computers still add up to over 160,000 points, yet my overall is only about 56,000.
Could not all those work units be re-validated so their value changes from 900 to 5 and then the computers will correct themselves?
Thanks
Conan
Posted 3 Jan 2015 by
Conan
What badge level should I give?
current settings:
10k
50k
100k
500k
1M
5M
10M
50M
That's pretty standard. Go for it.
I have changed it to
10k
50k
250k
500k
1M
5M
10M
25M
In the FAQ, you show 9 levels of badges with a 2M level included. If this is the final badge levels should not the 2M be 2.5M to blend in with the 250k and 25M levels?
Just thinking out loud (and typing as I think)
Conan
Posted 28 Dec 2014 by
Conan
I noticed this output on a work unit
11:08:37 (21223): wrapper (7.2.26012): starting
11:08:37 (21223): wrapper: running llr64 ( -d -oPgenInputFile=input.prp -oPgenOutputFile=primes.txt -oDiskWriteTime=10 -oOutputIterations=50000 -oResultsFileIterations=99999999)
12:41:38 (21223): llr64 exited; CPU time 5523.472304
12:41:38 (21223): called boinc_finish
14:26:34 (21765): wrapper (7.2.26012): starting
14:26:34 (21765): wrapper: running llr64 ( -d -oPgenInputFile=input.prp -oPgenOutputFile=primes.txt -oDiskWriteTime=10 -oOutputIterations=50000 -oResultsFileIterations=99999999)
14:26:35 (21765): llr64 exited; CPU time 0.002999
14:26:35 (21765): called boinc_finish
Note how the CPU time went from 5,523.472304 seconds to 0.002999 seconds which then records a 0.00 second CPU time when uploaded.
This task
916010 899422 424 27 Dec 2014, 17:41:17 UTC 28 Dec 2014, 4:28:05 UTC Completed and validated 5,583.16 0.00 150.00 Sierpinski / Riesel Base - long v0.01
Conan
Posted 10 Dec 2014 by
Conan
If the problem is can't get any work units, it is probably because there is no work at the moment.
I think there are 32 (currently) work units waiting to be returned and then new work will be generated, that is going on from what I have read in the forums.
Conan
Posted 9 Dec 2014 by
Conan
I think I know whats wrong. I have built the wrapper files on win7 in vs2012 but not with the xp toolset, thats why the wrapper cant be run on xp machines. I will recompile it and see how its working. The llr app has the toolset.
Thanks Rebirther,
I just have to wait for some new work to test this out.
Conan
Posted 7 Dec 2014 by
Conan
Just joined and connected 4 computers, 2 Linux 64 bit and 2 Windows 32 bit.
My first results are all errors for 1 of the Windows 32 bit computers
<message>
couldn't start CreateProcess() failed - : -148
</message>
It runs Primegrid with no issues so I don't know what this error means.
I am waiting for the other computers to return results so I can see if it is a common problem for me.
Conan
Posted 7 Dec 2014 by
Conan
http://primes.utm.edu/primes/home.php
I need your vote if you want to add your find by your own name into the prime database. Another idea was to keep all the primes under an alternative name by mersenne.
Over at Primegrid, they have the option in your account to add your details, such as real name, to be submitted to the Primes listing on your behalf. The volunteer does not then have to do anything other than find the Prime.
I have no problem with this and my real name is listed in my account over there so Primegrid can submit primes found for me (not that I have found many).
Conan