Thanks, that is great. It will be some days until current exponent finish, so I can see if it works.

2 "expired" assignments completed anyway...
 

Now my LL tests have been changed to DC.

In both cases the prior assignee was close to done when they were expired.
Might there be a case for some "wiggle" room if obvious progress is being made and completion seems (i)eminent?

55780411
55914317

[QUOTE=petrw1;393540]Now my LL tests have been changed to DC.

In both cases the prior assignee was close to done when they were expired.
Might there be a case for some "wiggle" room if obvious progress is being made and completion seems (i)eminent?

55780411
55914317[/QUOTE]

I dropped the second since it was not started...the second is in fact now showing as a DC assignment.

So if a person has exceeded the allowable time and lost the assignment; (rules as defined; I agree)
then it gets assigned to another (ME); (rules as defined; I agree)
then the competes it after all and reports in; (and gets credit for the LL; I like that the work is allowed to finish but I think it should be a DC as some disincentive for exceeding the time limit allowed)
then mine get converted to a DC.

Any other opinions?

[QUOTE=petrw1;393640]I dropped the second since it was not started...the second is in fact now showing as a DC assignment.

So if a person has exceeded the allowable time and lost the assignment; (rules as defined; I agree)
then it gets assigned to another (ME); (rules as defined; I agree)
then the competes it after all and reports in; (and gets credit for the LL; I like that the work is allowed to finish but I think it should be a DC as some disincentive for exceeding the time limit allowed)
then mine get converted to a DC.

Any other opinions?[/QUOTE]First person to report is logged as normal LL result.
Next person to report is logged as DC result.
Subsequent reports are TC and higher.

Why mess with that and create confusion and complication?

I don't agree that being forced into a DC is in any way a disincentive. Why does it matter what it is called? The important thing is that the work is not lost and it is logged accordingly.

[size=1]I don't see the point of dropping the second assignment. The work still needs to be done anyway. It is only the order that it is done in that changes. The only thing that negatively affects progress is unnecessary duplicated extra work.[/size]

[QUOTE=petrw1;393540]In both cases the prior assignee was close to done when they were expired.
Might there be a case for some "wiggle" room if obvious progress is being made and completion seems (i)eminent?
[/QUOTE]

There is already wiggle room, so this should not happen very often. George posted this SQL code in the "Newer milestones thread":
[url]http://www.mersenneforum.org/showpost.php?p=387598&postcount=1473[/url]

So assignments gets 3.33 days extra for every 1% completed above 10% and there is an extra unspecified grace period if they are close to finished. I do not think we should add too much "wiggle room" or the rules will cease to function as intended.

Grace periods tacked onto grace periods is counterproductive.

Yes, it is sad if you were at 99% when you lost ownership of the assignment, but we TOLD you how many days you had and you AGREED to the terms, and THEN you were given extra time on top of it. They were at 99% when all their bonus time ran out, meaning they were probably at 80% when their actual time ran out.

They should NOT be doing assignments then can only complete to 80% in the allotted time. There's a reason why the preferred assignments option is OFF by default.

I am grateful for this conversation, in that it induced me to have a look at the rules. I am currently running my first-first time LL in a couple of years, after many LLDCs. I had gotten it in my head that only 60 days are allowed for preferred LL assignments instead of 90. I've been on track to finish this 56M in 53 days, but was a little concerned, lest something intervene to slow things down. It is a relief to see that an 8350 core is not [U]quite[/U] as slow as I had been thinking. :showoff:

[QUOTE=ATH;393652]There is already wiggle room, so this should not happen very often. George posted this SQL code in the "Newer milestones thread":
[URL="http://www.mersenneforum.org/showpost.php?p=387598&postcount=1473"]…[/URL]
So assignments gets 3.33 days extra for every 1% completed above …[/QUOTE]


Unfortunately the grace period is only applied to assignments that are grandfathered by the old rules. The grace period is not applied to assignments handed out in the last 11 months.

[QUOTE=flagrantflowers;393660]Unfortunately the grace period is only applied to assignments that are grandfathered by the old rules. The grace period is not applied to assignments handed out in the last 11 months.[/QUOTE]

Correct....the assignments that were expired and then assigned to me were both about 99% done reporting progress every day (that is, NOT abandoned.)

So NO, not grace on top of grace because these new (non-grandfathered) currently have NO grace.

[QUOTE=petrw1;393662]Correct....the assignments that were expired and then assigned to me were both about 99% done reporting progress every day (that is, NOT abandoned.)

So NO, not grace on top of grace because these new (non-grandfathered) currently have NO grace.[/QUOTE]

From my poking around in the data earlier, it seems like that kind of thing does happen from time to time anyway, even under the new assignment rules. Maybe not as often, but it can and does.

As far as what happens... well, the first one to check in a result is quite literally a first-time check, and the next one to check in a result is, by definition, a double-check. True, sometimes the person checking it in second is the one that was officially assigned that work, but short of redesigning the client to halt all progress if it's exponent was reassigned, or ignoring results without a valid assignment ID, I don't know what could be done.

In nearly all cases it won't matter, because first and double checks are equally needed. The rare case would be if the assignment was prime and it was reassigned. But still, I'd say there's a case to be made even then that if the original assignee happened to report in before the new assignee, the credit would go to that original person. It'd be weird and awkward, but there's no getting around the fact that so-and-so found it first, no matter the circumstances. It'd be interesting if it were a prize-winning find... I wouldn't want to get into the middle of that because money complicates things even more than just the credit of having your name listed with the other discoverers.

I suppose it might be the type of thing that's even covered under the terms of use of the software? If not, maybe it'd be a useful addition, just in case.

[QUOTE=Madpoo;393755]But still, I'd say there's a case to be made even then that if the original assignee happened to report in before the new assignee, the credit would go to that original person. It'd be weird and awkward, but there's no getting around the fact that so-and-so found it first, no matter the circumstances. It'd be interesting if it were a prize-winning find... I wouldn't want to get into the middle of that because money complicates things even more than just the credit of having your name listed with the other discoverers.

I suppose it might be the type of thing that's even covered under the terms of use of the software? If not, maybe it'd be a useful addition, just in case.[/QUOTE]

The terms say that the first to report the prime to the PrimeNet servers gets the prize.