mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Conjectures 'R Us (https://www.mersenneforum.org/forumdisplay.php?f=81)
-   -   Recommended bases and efforts (https://www.mersenneforum.org/showthread.php?t=13196)

gd_barnes 2010-03-24 10:56

Recommended bases and efforts
 
1 Attachment(s)
Here are some recommended bases for testing. There are links to the applicable sieve files on the reservations web pages.

* - sieve file available
s - sieving in progress
(R=Riesel)
(S=Sierp)

Bases with 6, 7, or 8 k's remaining currently at n<200K; (approximate test time is 2 months on a modern quad):
[code]
# of suggested
base k's search reservation status date
S994* 7 n=100K-200K BOINC 2020/11/23

[/code]Bases with 2 k's remaining currently at n<300K; (approximate test time is 2 months on a modern quad):
[code]
suggested
base search reservation status date
(none)
[/code] Bases with relative difficulty currently < 25000**; (test time varies based on base and difficulty from 1-3 months on a modern quad):
[code]
# of suggested
base k's search reservation status date
S368* 3 n=200K-300K BOINC 2020/11/25

[/code]** - As shown on the unproven conjectures page here:
[URL="http://www.noprimeleftbehind.net/crus/vstats_new/crus-unproven.htm"]www.noprimeleftbehind.net/crus/vstats_new/crus-unproven.htm[/URL]

You can post reservations and statuses in this thread but after they are done, they will be moved to their applicable threads.

gd_barnes 2010-03-25 02:56

1 Attachment(s)
[quote=Batalov;209434]Reserving R96 as new to n=25K.

There are only 298 k's remaining after the script run at n<=1K, the results are attached. Not too shabby. Going on to n=25K.[/quote]

I've just about finished taking all of those new bases to n=2500 to give people a starting point and to give a good idea of how big of an effort each would be. Yes, that was one of the better ones considering its conjecture size.

I don't know if it will help you now but attached is a file of the 176 k's remaining at n=2500 for R96.

I am now showing the # of k's remaining for each of the new bases in the 1st post at n=2500 with the exception of R108, R192, and S192 that I'm not quite done with yet. I'll attach them to the 1st post as I finish them.


Gary

mdettweiler 2010-03-25 03:03

[quote=gd_barnes;209436]I've just about finished taking all of those new bases to n=2500 to give people a starting point and to give a good idea of how big of an effort each would be. Yes, that was one of the better ones considering its conjecture size.

I don't know if it will help you now but attached is a file of the 176 k's remaining at n=2500 for R96.

I am now showing the # of k's remaining for each of the new bases in the 1st post at n=2500 with the exception of R108, R192, and S192 that I'm not quite done with yet. I'll attach them to the 1st post as I finish them.


Gary[/quote]
BTW, a tip for attaching files to posts such as the first post here: to attach multiple files to a single post (without having to put them all in one zip file and therefore keep re-uploading every time one little file is changed), attach the first file to the post itself; then make additional post for each respective attachment, and use the "merge posts" feature to merge them into the first post.

gd_barnes 2010-03-25 03:38

[quote=mdettweiler;209438]BTW, a tip for attaching files to posts such as the first post here: to attach multiple files to a single post (without having to put them all in one zip file and therefore keep re-uploading every time one little file is changed), attach the first file to the post itself; then make additional post for each respective attachment, and use the "merge posts" feature to merge them into the first post.[/quote]

Why? That would be more effort. It's easier to:

1. Add an additional file to the zip on my machine.
2. Edit the post to delete the previous attachment and add the new one.

mdettweiler 2010-03-25 03:52

[quote=gd_barnes;209450]Why? That would be more effort. It's easier to:

1. Add an additional file to the zip on my machine.
2. Edit the post to delete the previous attachment and add the new one.[/quote]
Well, I guess what I was thinking is if the zip file was on the largish side it would be a pain to have to upload it again each time...though I do suppose the forum's attachment size limit keeps that in check somewhat.

gd_barnes 2010-03-25 03:59

[quote=mdettweiler;209454]Well, I guess what I was thinking is if the zip file was on the largish side it would be a pain to have to upload it again each time...though I do suppose the forum's attachment size limit keeps that in check somewhat.[/quote]

In thinking it over a little more, I think it would be about equally fast either way. I was thinking you'd have to create a number of new posts each time that equalled the total # of files that you had. But that isn't true. For each new file, you only have to create and merge one more new post.

I'll keep that in mind for future reference since it would be slightly easier for the person who needs the info. to have them in separate files and it shouldn't be any more hassle to do it that way. Thanks for the tip.

mdettweiler 2010-05-05 20:45

[quote=gd_barnes;214087]OK, I'll start sieving R181 and R182 for n=50K-100K later today. Go ahead and post a regular reservation in the bases 101-250 thread. I'll show them as reserved by you in this thread.[/quote]
Great, thanks! :smile:

[quote]These bases won't keep your quad busy for very long. The natural progression of things on your effort would make S208, R214, and R221 for n=50K-100K the next ones to tackle. S208 is heavier weight so would take longer unless a prime is found quickly. Let me know if you might be interested in those.

Another thing that I'd like to see tackled is the bases <= 250 that have 2 or 3 k's remaining that are only at n=25K. There are quite a few of those. It would be nice to see if we could add them to this thread (or prove them) by knocking out a k or 2 on them. That's something I'll probably start tackling myself over the next few months if others haven't done most of them already.[/quote]To itemize those:
S72 - 2 k's remaining at 40K
S133 - 3 k's remaining at 40K
S140 - 2 k's remaining at 25K
S157 - 3 k's remaining at 25K
S187 - 2 k's remaining at 25K
S217 - 3 k's remaining at 25K
R103 - 3 k's remaining at 40K
R158 - 3 k's remaining at 25K
R191 - 2 k's remaining at 50K
R213 - 2 k's remaining at 25K
R218 - 2 k's remaining at 25K
R236 - 2 k's remaining at 25K

Heck, if my quad can burn through those 1-k-remaining bases so fast (for some reason I was initially thinking that they'd take longer--not sure why), maybe I should just go for some of those instead. Tell you what--I'll reserve R181, R182, S140, and S187 to start with, if you don't mind sieving all that. (Note that they wouldn't have to be done all at once; I imagine even one core of your i7 sieving would be enough to keep up with my quad doing PRP if you do them one at a time and send them to me as they're ready.) That should clean up all remaining unreserved bases <200 with one k remaining at 50K, and then all unreserved Sierp. bases < 250 with 2 k's remaining at 25K.

Note that in all cases, I'd be stopping at n=100K: that means 50K-100K for R181 and R182, and 25K-100K for S140 and S187.

gd_barnes 2010-05-06 02:25

[quote=mdettweiler;214101]Great, thanks! :smile:


To itemize those:
S72 - 2 k's remaining at 40K
S133 - 3 k's remaining at 40K
S140 - 2 k's remaining at 25K
S157 - 3 k's remaining at 25K
S187 - 2 k's remaining at 25K
S217 - 3 k's remaining at 25K
R103 - 3 k's remaining at 40K
R158 - 3 k's remaining at 25K
R191 - 2 k's remaining at 50K
R213 - 2 k's remaining at 25K
R218 - 2 k's remaining at 25K
R236 - 2 k's remaining at 25K

Heck, if my quad can burn through those 1-k-remaining bases so fast (for some reason I was initially thinking that they'd take longer--not sure why), maybe I should just go for some of those instead. Tell you what--I'll reserve R181, R182, S140, and S187 to start with, if you don't mind sieving all that. (Note that they wouldn't have to be done all at once; I imagine even one core of your i7 sieving would be enough to keep up with my quad doing PRP if you do them one at a time and send them to me as they're ready.) That should clean up all remaining unreserved bases <200 with one k remaining at 50K, and then all unreserved Sierp. bases < 250 with 2 k's remaining at 25K.

Note that in all cases, I'd be stopping at n=100K: that means 50K-100K for R181 and R182, and 25K-100K for S140 and S187.[/quote]

Sure thing. I'll just stick one core of the I7 on it in the next couple of hours. I'll see how long it takes to sieve one of them and let you know. If necessary, I might initially put a 2nd core on it to help you fill up your quad more quickly. But since sieving is only 5-10% of any total effort, 1 sieving core should well outpace 4 testing cores.

Thanks for the list. That might be a good list for others to work from also. Concentrating on bases <= 250 is good because all bases <= 250 with CK <= 10K have been or are currently being searched to n=25K. So it's virtually set in stone that there will be no more bases <= 250 with 1 or 2 k's remaining barring some miracle unknown super-heavy-weight base with CK > 10K that manages to prime all of its k's except 1 or 2. That seems virtually impossible.

BTW, I'll make a SWAG estimate at about 1-2 CPU weeks avg. on these for each k for n=50K-100K. So a quad could likely blow through them in 2-4 days each (for n=50K-100K on the 1k bases; add about 20% for each k for n=25K-100K). I think the 2 k's that I did for S218 for n=50K-100K took about 8 and 9 CPU days respectively. I ran them on one core each.


Gary

gd_barnes 2010-05-06 19:24

Max,

For consistency, I added all 2k & 3k bases <= 250 that are at n<=100K. If you are going to search some 2k & 3k bases that are at 25K to 100K, then we may as well have a list of all of them that are at n<=100K. All of the others were either at 40K or 50K.

I also corrected your S133 from 25K to 40K, which prompted the additions.

The one that is "most behind" where it should be vs. its neighbors with a similar # of k's remaining is the 2 k's on S72, which are at n=40K.

BTW, all of the 1k & 2k bases that you've reserved are already recommended in this thread. These types of bases are the ones that it would be nice to have pushed higher.


Gary

gd_barnes 2010-05-16 02:50

I received a sieve file from Mathew for S49 sieved to P=500G. It is now posted on the web pages.

gd_barnes 2010-05-20 23:42

[quote=Mathew Steine;215548]I would like to reserve R192 to n=25K[/quote]


Great! Completion of this and a few other reservations will complete all bases on both sides <=200 with CK<20K to n=25K. I think that will be a significant milestone. Many bases with CK=10K-20K have been a large amount of work to get to n=25K.

To all,

I've added a link to our 1st post in this thread to the posting that contains all bases <= 250 that have either 2 or 3 k's remaining at n<=50K. See it directly below all of the recommended bases. Max originated the post and we are working together to keep it updated. With all 1k bases <= 200 now nearly complete to n=100K, that will be a good place to look for work. Perhaps we'll end up with a few more 1k (or proven) bases. :-) Another place to look would be in the 1k thread to extend all of those bases 200-250 to n=100K.


Gary


All times are UTC. The time now is 08:37.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.