Proposed LL assignment and recycle rules

[Prime95]

Assignment rules (top 3000 exponents):
	1)  No v4 users
	2)  No manual reservations
	3)  Reliability is high.
	4)  Speed is satisfactory.
	5)  "Days of work" setting is short.

Recycle rules (top 1500 exponents):
	1)  If expected completion date is not updated for 60 days.
	2)  If assignment made before 2014-02-01 and:
	  2a) assignment is 12 months old and < 50% complete it is recycled.
	  2b) assignment is 15 months old it is recycled.
	3)  If assignment made after 2014-02-01 and:
	  3a) assignment is 6 months old and not started it is recycled.
	  3b) assignment is 9 months old it is recycled.

Assignment rules (exponents > 100M):
	1)  No restrictions

Recycle rules (exponents > 100M):
	1)  If manual assignment, recycled if no result reported in 5 years.
	2)  If prime95, recycled if assignment is greater than 1 year old and
		LL has not begun.
	3)  If prime95, recycled if assignment is greater than 2 years old and
		expected completion date has not updated for 180 days.

Assignment rules (remaining exponents):
	1)  No v4 users
	2)  Speed is satisfactory.

Recycle rules (remaining exponents):
	1)  If expected completion date is not updated for 60 days.
	2)  If prime95 and assignment is one year old and not started it is recycled.
	3)  If assignment is two years old it is recycled.

[Aramis Wyler]

Could the TF and DC rules be the same, aka they would be the 'LL' rules whether it was regular LL range, DC range, or LMH range. I think that would simplify understanding, and the amount of time could simply be based on the size of the exponent + 90 days for the queue.

[TheMawn]

I think because DC's are 4-8 times shorter than LL's the timelines could be shortened, but that's just me. It looks like the rules are very similar so Aramis has a good point.

EDIT: I like the recycling after two years no matter what, based on the number of errors likely to accumulate after a two-year test.

EDIT: What are the poaching rules? Does the user being poached upon get the credit? I noticed a DC rule regarding to work being completed by another user. In the case of LL, if someone else does the LL, the user being poached on can continue his LL and "double-check" while being credited for a first-time LL?

[ET_]

Following the rules, it seems to me that v4 clients will only get LL > 100M.

But v4 clients are likely to be set up on older, slower computers.

Did I misunderstand the rules?

Luigi

[petrw1]

Originally Posted by ET_

Following the rules, it seems to me that v4 clients will only get LL > 100M.

But v4 clients are likely to be set up on older, slower computers.

Did I misunderstand the rules?

Luigi

I believe it was posted recently on a diff thread that V4 only gets DC and over 33M.

[kracker]

Originally Posted by TheMawn

I think because DC's are 4-8 times shorter than LL's the timelines could be shortened, but that's just me. It looks like the rules are very similar so Aramis has a good point.

Why? DC's are for slower computers that take almost the same time as a LL on a fast computer. I have 6 Core2 Duo's on DC's, although quite inefficient compared to a Haswell, they still can produce.

IMO, I really like what is going to be/implemented. I just hope we will keep it as loose/flexible as possible to avoid even accidentally stepping on people, even if is very rare.

[ET_]

Originally Posted by petrw1

I believe it was posted recently on a diff thread that V4 only gets DC and over 33M.

Thanks!

Luigi

[TheMawn]

Originally Posted by kracker

Why? DC's are for slower computers that take almost the same time as a LL on a fast computer. I have 6 Core2 Duo's on DC's, although quite inefficient compared to a Haswell, they still can produce.

Right. Slower computers. Thanks.

[kracker]

Originally Posted by TheMawn

Right. Slower computers. Thanks.

[TheMawn]

Originally Posted by kracker

That might have come out wrong. I meant to thank you for letting me know what I had missed.

[kracker]

Originally Posted by TheMawn

That might have come out wrong. I meant to thank you for letting me know what I had missed.

Sorry as well. I took your previous comment as sarcasm. It is and always will be my opinion that slower computers are not useless.

EDIT: Well, only to a certain extent of course.