Newer milestone thread

[kracker]

Well, I just checked the top 5 or 6 in the list and they were assigned to either me, chalsall or monst.

[c10ck3r]

Would taking all 461 exponents to 70/71/72 bits be considered poaching? That could eliminate a couple 20 exponents, I'm sure. (And by sure, I mean that I'm guessing without checking the average bit-depth for the actual assignments)

[petrw1]

Originally Posted by kracker

Well, I just checked the top 5 or 6 in the list and they were assigned to either me, chalsall or monst.

27 here. Not all scheduled in December but I could do some rearranging.

[richs]

TheMawn, I've been a participant for 15 years and I request that you don't poach my DC assignments under 30402457. They will all be completed in the next two months.

[Uncwilly]

Originally Posted by Uncwilly

462 is a bit too much work to expect a "Christmas miracle" this year (or even a tidying up by the end of the year.)

Originally Posted by philmoore

This quote is from October 3rd - I just want to point out that progress has been pretty steady on this first double-check milestone since then, and current GIMPS strategy has been to only hand out the lowest available double checks to "most trusted" participants.

Yes, we have been making good steady progress on the bottom end. I see no need to change anything. If for the past 6 months there had been 25 until the milestone, I might expect one of GIMPS "Christmas Miracles". But, with the steady march (I have been watching for us to be at 1/2 of the Oct 3rd value), I think that your right to look for the completion of multiple milestones in 2014. I am guessing that the 462 will be down to 0 by Pi Day '14. Remember that 100.1 kph > 100 kph.

[Brian-E]

Originally Posted by c10ck3r

Would taking all 461 exponents to 70/71/72 bits be considered poaching? That could eliminate a couple 20 exponents, I'm sure. (And by sure, I mean that I'm guessing without checking the average bit-depth for the actual assignments)

This might be a good use of resources if you assume that anyone alienated by having their long-running DC rendered worthless when a factor is found nonetheless continues to contribute to the project with the same enthusiasm as before. But I don't think the assumption holds.

Originally Posted by richs

TheMawn, I've been a participant for 15 years and I request that you don't poach my DC assignments under 30402457. They will all be completed in the next two months.

+1 !
While it's true that I've participated for far less time, I do however also relate to this sentiment strongly.

[bloodIce]

Any hope for those ~3,2K low-hanging fruits from 2M to 2.5M, which need a push from 62 to 63 bits http://www.mersenne.info/trial_facto...ata/3/2000000/? I am sure some factors will pop up. With your resources this will be a half day work. I occasionally do several expos in this range, but it will be nice if the heavy guns shoot once.

[petrw1]

Originally Posted by bloodIce

Any hope for those ~3,2K low-hanging fruits from 2M to 2.5M, which need a push from 62 to 63 bits http://www.mersenne.info/trial_facto...ata/3/2000000/? I am sure some factors will pop up. With your resources this will be a half day work. I occasionally do several expos in this range, but it will be nice if the heavy guns shoot once.

I have a pea-shooter available. :)

[LaurV]

Those are in the "normal mfaktc" range, so it would be no big effort. I made a list with all of them which are not reserved, about 3K. One non-OC, gtx580 can do two in 5 minutes (yes, you need two copies to max the card, the exponents are still small for big GPU sieving jobs, and the total GHZd/d maximum is about 340, or 360 with CPU sieving, but you waste also cpu core) so there would be 3000*5/4=3750, about 62 hours on two cards. You can't get the 440 GHzd/d like for the LL front. I may give it a run these days.

[ET_]

Originally Posted by LaurV

Those are in the "normal mfaktc" range, so it would be no big effort. I made a list with all of them which are not reserved, about 3K. One non-OC, gtx580 can do two in 5 minutes (yes, you need two copies to max the card, the exponents are still small for big GPU sieving jobs, and the total GHZd/d maximum is about 340, or 360 with CPU sieving, but you waste also cpu core) so there would be 3000*5/4=3750, about 62 hours on two cards. You can't get the 440 GHzd/d like for the LL front. I may give it a run these days.

I'd like to do some of them too, but it seems that GPU72 can't handle them. I guess I should reserve them on Primenet...

Edit: No way I should find another way to get them.

Luigi

BTW, a shared list of valid exponents would be nice to work with...

[chalsall]

Originally Posted by ET_

I'd like to do some of them too, but it seems that GPU72 can't handle them.

GPU72 can handle just about anything...

If there is demand, a "Great Internet Mersenne Factor Search" option could be easily added. Just like we did for the 332M range. This would actually take me very little time.

I must be understood that the GIMFS effort would not help GIMPS in any way, since it would only involve searching for factors for MP candidates which have already been proven to be composite (read: two matching LLs).

But as as part of the ethos of the GIMPS community, those participating are free to do whatever they want with their own time, kit and money so long as it doesn't negatively impact other participants.

A show of hands who wants this? Please include desired ranges.