Lost assignment?

[mdettweiler]

Originally Posted by lfm

As for wasting his old 25%, he did that when he quit for 6 months or more and allowed it to be reassigned. We can't just wait forever for these guys to get back around to us. When you quit you should accept the consequences. If the project as a whole has to accept a little wasted capacity, that's nothing special either.

Actually I expect the point will be mute, if he dropped out once before he probably will again without completing any result.

I don't believe that was the case: given that the exponent is in the 100M digit range, it would seem more likely that the exponent just ran into the 1-year limit rather than the OP having actually "quit".

[lfm]

Originally Posted by mdettweiler

I don't believe that was the case: given that the exponent is in the 100M digit range, it would seem more likely that the exponent just ran into the 1-year limit rather than the OP having actually "quit".

It would have to run for 6 months without a progress report even? Is it offline and still running? It can't be that old a machine if it really did get to 25%, its hard to immagine its not on the internet at all but I guess anything is possible.

[JuanTutors]

lfm, I appreciate your sentiment, but I don't plan to waste a year of work because my copy of prime95 didn't communicate with the server.

[lfm]

Originally Posted by dominicanpapi82

lfm, I appreciate your sentiment, but I don't plan to waste a year of work because my copy of prime95 didn't communicate with the server.

Why didn't it send in regular updates? was its connection to the internet bad or disabled? If you don't fix it, it will probably happen again. And why did it take so long for you to notice?

[petrw1]

http://www.mersenneforum.org/showpos...98&postcount=9

The v5 server reserves the right to reassign exponents that take longer than a year to complete. This won't happen for 100M exponents. The warning is for users that reserve small exponents that will take years to complete. In the past, we've been very, very tolerant of reservations that take several years to complete. The new policy tries to strike a better balance.

[mdettweiler]

Originally Posted by petrw1

http://www.mersenneforum.org/showpos...98&postcount=9

In this case, that seems to be a moot point since the 1-year deadline only kicks in when the client is reporting progress to the server often enough to keep extending the "normal" reassignment dates. In this case, server communication didn't occur, so it would have been reassigned after 60 days IIRC (long before the 1-year deadline).

[lfm]

Originally Posted by mdettweiler

In this case, that seems to be a moot point since the 1-year deadline only kicks in when the client is reporting progress to the server often enough to keep extending the "normal" reassignment dates. In this case, server communication didn't occur, so it would have been reassigned after 60 days IIRC (long before the 1-year deadline).

60 days -> 6 months actually

It seems the current assignment for the number was made more than a year after the initial assignment so there might have been yet another assignee and timeout in the interim.

[ATH]

dominicanpapi82 you should in Prime95 choose Options - Preferences and set "Days between sending new end dates" to 30 or lower, or in prime.txt add the line:
DaysBetweenCheckins=30
or lower than 30.

[JuanTutors]

Originally Posted by ATH

dominicanpapi82 you should in Prime95 choose Options - Preferences and set "Days between sending new end dates" to 30 or lower, or in prime.txt add the line:
DaysBetweenCheckins=30
or lower than 30.

I can't set that option anymore. It will delete my worktodo.txt file. I will have to submit my work manually when it finishes (in 2013!) The only other option I see is contacting the guy who got reassigned my exponent and letting him know, but from what I have seen that's not part of policy.

[mdettweiler]

Originally Posted by dominicanpapi82

I can't set that option anymore. It will delete my worktodo.txt file. I will have to submit my work manually when it finishes (in 2013!) The only other option I see is contacting the guy who got reassigned my exponent and letting him know, but from what I have seen that's not part of policy.

What if you took the exponent and reformatted its worktodo.txt line like this:
Doublecheck=exponent,TF bits,1
i.e., as a doublecheck without any assignment key. If I understand correctly, that should register the exponent as a doublecheck and then pick up exactly where the save file left off before.

I've never tried this myself; can someone confirm whether it's possible to register a doublecheck for an exponent while the first-pass test is still in progress?

[ATH]

Originally Posted by mdettweiler

What if you took the exponent and reformatted its worktodo.txt line like this:
Doublecheck=exponent,TF bits,1
i.e., as a doublecheck without any assignment key. If I understand correctly, that should register the exponent as a doublecheck and then pick up exactly where the save file left off before.

I've never tried this myself; can someone confirm whether it's possible to register a doublecheck for an exponent while the first-pass test is still in progress?

As was discussed in this thread: http://www.mersenneforum.org/showthread.php?t=13472 all LL tests have random shifted S0 start value, so the test will count as a doublecheck even if its left at Test= in worktodo.txt