| Author |
Message |
|---|
|
|
|
Time to moan about credits again :-)
There appears to be new longer pps workunits such as this one: http://www.primegrid.com/workunit.php?wuid=235306142
However, compared to the shorter pps tasks these ones are only giving about half the credits (129 for 3.6 hours work compared to around 1 credit/minute for the short ones). |
|
|
 rroonnaalldd Volunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
                 
|
|
1000sec for 10 credits or 13000sec for 130 credits.
____________
Best wishes. Knowledge is power. by jjwhalen
 |
|
|
|
|
|
Just checked some of my boxes. I'm getting about the same credit per time with many long and short units on it BUT some long units get about half credit compared to other long units, for about the same computation time. Some short units get above average credit too. |
|
|
DoESVolunteer tester
Send message
Joined: 11 Oct 08
Posts: 784
ID: 30382
Credit: 74,878,735
RAC: 0
            
|
Just checked some of my boxes. I'm getting about the same credit per time with many long and short units on it BUT some long units get about half credit compared to other long units, for about the same computation time. Some short units get above average credit too.
For 16000secs I am seeing 280 to 300 credit
thats on a gulftown i7
Duo cores are claiming about a third of that
____________
Member of AtP
Shown here is an Australian native rat (Ratus Kickarsus) |
|
|
|
|
|
Seems to be entirely random what is awarded.
on the same pc:
11000 sec, 130 credits
11000 sec, 228 credits |
|
|
|
|
Seems to be entirely random what is awarded.
on the same pc:
11000 sec, 130 credits
11000 sec, 228 credits
Not entirely random - the lower credit in your example was awarded to a double check where the initial result has a short 600 secs in the database, which skews the average of the 2 results. The higher credit was awarded where only you worked on it and therefore had a longer runtime. The credit seems to be proportional to the average time taken where more than 1 result is submitted.
____________
 |
|
|
|
|
|
The problem is that all results submitted by another project which only get double-checked have a standard CPU time of 600seconds...
In order for the credit to fairly and accurately reflect the work done on a workunit I believe the policy of the standard 600seconds should somehow be changed... |
|
|
|
|
|
I don't know if one of the admins changed it or if I missed it earlier,
but it seems that the run-time of a double-checker Wu was changed to 15,000 seconds. Thx for the quick adjustment!
An example:
http://www.primegrid.com/workunit.php?wuid=235359500 |
|
|
|
|
|
Yes , you are right: from 600 to 15.000 seconds.
Thx admins!
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
|
|
I don't know if one of the admins changed it or if I missed it earlier,
but it seems that the run-time of a double-checker Wu was changed to 15,000 seconds. Thx for the quick adjustment!
An example:
http://www.primegrid.com/workunit.php?wuid=235359500
Guess we have to complain to get this fixed so mine is :
WU 235299244
Oh well, at least there are Sophie's to do. |
|
|
|
|
Yes , you are right: from 600 to 15.000 seconds.
Thx admins!
thirded. thanks guys. |
|
|
|
|
|
But not always :(
http://www.primegrid.com/workunit.php?wuid=235314698
Still 600 seconds :(
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 903
                         
|
But not always :(
http://www.primegrid.com/workunit.php?wuid=235314698
Still 600 seconds :(
That's probably because this WU was created before they made the change. In fact, it was created before the first post in this thread.
The first person to whom this WU was assigned waited 41 hours before aborting it.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Could you increase the standard CPU runtime of a double-check WU from 15,000s to maybe 20,000s...?
Newest workunit took me 21,683s (on an i7 @ 2,3GHz)... |
|
|
|
|
Could you increase the standard CPU runtime of a double-check WU from 15,000s to maybe 20,000s...?
Newest workunit took me 21,683s (on an i7 @ 2,3GHz)...
20 000 is little to high but we will soon be there so from now new generated will get 20k.
Lennart
|
|
|
|
|
|
PPS LLR wu' properties still show "estimated task size" 1605GFLOPs", which is causing buffer to behave strange (I'm getting work for a whole day, having set 0,2 in the additional buffer size in prefs). I have crunched a few dozen long wu's by now (the runtime is pretty accurate) so that info should come from the server side. If so, please correct that as soon as possible, please. |
|
|
|
|
|
My up-most appolgy to PG and the wing man..For some reason my PPS.llr than ran for 15 minutes turned into 3:30 to 4:30 hours. Even tho my cach is a day on all..well I had to abort quit a bit of wu's because of time dead lines. Sorry..
____________
My first Prime
20208428036625*2^666666-1 (SGS)
PROUD MEMBER OF TEAM CARL SAGAN |
|
|
|
|
|
Wow, same problem here.
I didn't have a big cache, but my boinc still thinks a WU takes 9 minutes, instead of the 3.45 hours. So I have a incredible amount of WU's to do before December 25. I will have a try, but I can't promiss I won't have to send some back unfinished.
____________
 |
|
|
|
|
|
If you see right now that you cannot finish WU before deadline, abort it. Your wingman will wait in any case, but will wait less if you abort now and project sent those WU to another. If you wait until deadline, and project abort WU then your wingman will wait few days more....
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
|
These a few days can be weeks or more while calculating SoB-units. Aborting is then the best solution ever.
____________
Best wishes. Knowledge is power. by jjwhalen
|
|
|
|
|
|
I have aborted a fair amount this morning, since I will not be able to complete over 200 units in 3 days.
Luckily I was able to check my other rigs at the office today so my wingmen doesn't get screwed for a few days ;)
____________
|
|
|
