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 2016-04-10 07:21

1k bases S638 and S785 have now been sieved to P=20T.

rebirther 2016-04-10 07:25

[QUOTE=gd_barnes;431200]1k bases S638 and S785 have now been sieved to P=20T.[/QUOTE]

Can you make the thread sticky? I will grab these 2 bases as soon as possible ;)

gd_barnes 2016-05-03 18:52

Added R694 (2 k's remaining for n=100K-200K) and S866 (1 k remaining for n=200K-400K) to the recommended list. Both have sieve files.

As the lowest difficulty base remaining, S428 with 1 k remaining for n=200K-400K is still the top recommendation. It also has a sieve file.

LaurV 2016-05-04 09:00

[QUOTE=gd_barnes;433034]As the lowest difficulty base remaining, S428 with 1 k remaining for n=200K-400K is still the top recommendation. It also has a very deep sieve file to which LaurV worked a lot.[/QUOTE]
Fixed it for you :razz:

gd_barnes 2016-08-15 22:37

I have added as a priority for BOINC R625 and S625 for n=25K-100K. I have fully sieved them to P=15T and the files are posted on the reservations pages. These bases have k's with several different search depths due to the base 5 effort. That and the fact that they have fewer than 100 k's remaining at a search depth of n=25K is why I am recommending them. Note that I only sieved the k's and n-ranges that need to be tested to bring them fully up to n>=100K. That is why there are a different # of k's remaining shown on the pages than are in the sieve files.

gd_barnes 2017-03-29 06:11

Suggestion for BOINC: Search all Riesel bases with < 120 k's remaining for n=25K-100K. Most or all of them have sieve files.

I noticed that the Sierp bases are, in general, well ahead of the Riesel bases now.

rebirther 2017-03-30 17:33

[QUOTE=gd_barnes;455698]Suggestion for BOINC: Search all Riesel bases with < 120 k's remaining for n=25K-100K. Most or all of them have sieve files.

I noticed that the Sierp bases are, in general, well ahead of the Riesel bases now.[/QUOTE]

ok, noticed ;)

gd_barnes 2017-04-21 06:06

Suggestion for BOINC: R693 for n=50K-100K with 58 k's remaining. This base was just completed to n=50K.

gd_barnes 2017-05-05 03:28

Added S646 for n=10K-25K to the recommended list. It only has 298 k's remaining.

gd_barnes 2017-05-05 18:37

Added a bunch of Riesel bases for n=25K-100K for BOINC. All have sieve files.

Added some Sierp bases for n=10K-25K. Sieving has not been done on these.

MisterBitcoin 2017-05-05 19:11

You can remove R877, srbsieve just passed n=7,6K.

rebirther 2017-05-05 19:25

[QUOTE=gd_barnes;458376]Added a bunch of Riesel bases for n=25K-100K for BOINC. All have sieve files.

Added some Sierp bases for n=10K-25K. Sieving has not been done on these.[/QUOTE]

Can you update the reservation table? The R576 is still reserved since 2016, the remaining k doesnt match the recommended table.

gd_barnes 2017-05-05 19:40

[QUOTE=rebirther;458381]Can you update the reservation table? The R576 is still reserved since 2016, the remaining k doesnt match the recommended table.[/QUOTE]

Everything is fully up to date. The R576 k's that are reserved are R24 k's that Ian (MyDogBuster) is working on. The sieve file contains the k's that are not being worked on that have only been searched to n=25K. The number of k's remaining in the recommended post here reflects just those k's. BOINC could reserve the base and use that sieve file to search the non-reserved k's.

This will happen on all bases that are a perfect square where there are overlapping k's with the k's from the base that is a square root of the base.

gd_barnes 2017-05-05 19:42

[QUOTE=MisterBitcoin;458379]You can remove R877, srbsieve just passed n=7,6K.[/QUOTE]

I will remove it when it is done. Also it is recommended to n=25K. You are only taking it to n=10K. For these smaller conjectured bases that only take a few days to test to n=10K, may I suggest testing them to n=25K instead?

MisterBitcoin 2017-05-05 19:49

[QUOTE=gd_barnes;458384]I will remove it when it is done. Also it is recommended to n=25K. You are only taking it to n=10K. For these smaller conjectured bases that only take a few days to test to n=10K, may I suggest testing them to n=25K instead?[/QUOTE]

I´ll test up to n=10K with srbsieve, higher values will be tested with LLR which is faster.

gd_barnes 2017-05-05 19:54

[QUOTE=MisterBitcoin;458385]I´ll test up to n=10K with srbsieve, higher values will be tested with LLR which is faster.[/QUOTE]

I'm a little confused. So are you reserving R877 to n=25K? Will you be reserving S646 to n=25K? If so, those can be done in one reservation where you report the status at n=10K and then again when finished.

MisterBitcoin 2017-05-05 20:01

[QUOTE=gd_barnes;458386]I'm a little confused. So are you reserving R877 to n=25K? Will you be reserving S646 to n=25K? If so, those can be done in one reservation where you report the status at n=10K and then again when finished.[/QUOTE]

My plan is to finish R877 to n=10K and release it, than I´ll take S646.
In the same time doing S646 I will do the remaining sieve files for the recommended 10-25K bases.

gd_barnes 2017-05-05 20:03

Great thanks. I'll go ahead and reserve S646 to n=25K for you.

gd_barnes 2017-08-24 22:11

I have added a few more recommended for misc. testing and for BOINC.

gd_barnes 2017-10-19 08:37

I have added 11 new bases to the list that have 1, 2, 3, or 4 k's remaining. Only one of the bases have been sieved so they are also a sieving recommendation. All have a "difficulty" value of < 10000 as shown here:
[URL]http://www.noprimeleftbehind.net/crus/vstats_new/crus-unproven.htm[/URL]

This means that the bases should, relatively speaking, be faster to test than most bases and many should be able to be proven over the next few months and years. That is...they are somewhat "behind" where they should be tested. :smile:

The bases with low difficulty values are a good way to come up work that gives us the most bang for our buck.

gd_barnes 2018-01-07 06:59

1k bases R398 and R662 for n=200K-500K recommended for BOINC. R662 now has a fully sieved file. I'm working on sieving R398. These are the two lowest difficulty bases remaining on the Riesel side.

I also added S212 for k=6 only that has a fully sieved file. The other 3 k's on this base have been searched to n=500K. k=6 has only been searched to n=200K and so needs to be caught up.

gd_barnes 2018-01-15 01:43

Added R615 and R643 for BOINC. 2k and 3k bases to n=250K. R615 has a sieve file. I'll start working on a file for R643 later this week.

gd_barnes 2018-02-11 05:12

Added R411, R495, and R675 n=25K-100K for BOINC. R675 sieving just completed. R411 and R495 sieving is in progress. MisterBitcoin is working on R411 and I am working on R495.

Also added a new designation in the first post. "s" means that sieving is in progress.

gd_barnes 2018-02-16 17:45

I added a few more n=25K-100K bases. We are gradually getting some of them sieved.

R495 is ready to go. :smile:

gd_barnes 2018-03-02 20:31

R411 and S810 sieve files are ready to go. :smile:

gd_barnes 2018-03-11 02:18

I have restructured the first post in this thread to more closely align with our 2018 project goals. A lot of bases were added. Many of the 2 and 3 k bases for n=100K-200K are modest efforts and none of them are currently reserved. We should be able to add some 1kers out of those and maybe a proof or two.

pepi37 2018-11-28 12:10

Soon I will start with preparing new set of sieve files for YOYO. Queue will be empty in 10-15 days: so if you have some requests for sending any bases, please send me PM. Also REB can make request if he need sieve files for BOINC.

masser 2018-11-28 14:26

[QUOTE=pepi37;501134]Soon I will start with preparing new set of sieve files for YOYO. Queue will be empty in 10-15 days: so if you have some requests for sending any bases, please send me PM. Also REB can make request if he need sieve files for BOINC.[/QUOTE]

There are 12 unreserved sequences on the CRUS website list, Top 20 Conjectures with 1k Remaining by Highest Weight, with minimum decimal length less than 600000. Would those be good candidates for YOYO or BOINC?

gd_barnes 2018-11-29 05:24

[QUOTE=pepi37;501134]Soon I will start with preparing new set of sieve files for YOYO. Queue will be empty in 10-15 days: so if you have some requests for sending any bases, please send me PM. Also REB can make request if he need sieve files for BOINC.[/QUOTE]


All bases with 4 or 5 k's remaining at n=100K. You could sieve those for n=100K-300K. They will be a goal for next year. Also I think there are a few 1k'ers that are at n=300K. Those could be sieved for n=300K-1M.

pepi37 2018-12-14 18:46

Ten S bases with 4 -8 K from 100-300K is sent to YOYO.
I will prepare this weekend at lest two more batches and sent it also :)

Felix 2019-01-03 12:50

do you plan to update the listing of the bases with the new 2019 project goals?

gd_barnes 2019-01-03 13:05

I'm in the process of doing that now. The 1k conjectures are in there.

gd_barnes 2019-01-04 00:28

I have added all of the new recommended bases for our 2019 goals to post 1 here. :smile:

gd_barnes 2019-05-14 05:11

I have added S140 to the 1k recommended list currently reserved by matzetoni. It recently dropped to a 1ker. :-)

gd_barnes 2019-09-06 00:43

Ian released recommended base S1002. So it is now open for BOINC or anyone else to take.

gd_barnes 2019-09-08 22:19

S686 now has a sieve file.

gd_barnes 2019-09-28 00:59

S892 now has a sieve file. :smile:

gd_barnes 2019-12-23 06:24

Recommendation for BOINC: S595 for n=25K-100K.

I just finished sieving this one to P=1T.

Reb, this is one that you recently tested to n=2500 and then Ian and I searched it to n=25K. It turned out to have only 165 k's remaining at n=25K for a conjectured k > 300K much to my surprise. :smile:

rebirther 2019-12-23 06:28

[QUOTE=gd_barnes;533416]Recommendation for BOINC: S595 for n=25K-100K.

I just finished sieving this one to P=1T.

Reb, this is one that you recently tested to n=2500 and then Ian and I searched it to n=25K. It turned out to have only 165 k's remaining at n=25K for a conjectured k > 300K much to my surprise. :smile:[/QUOTE]


This can take a while. S612 still reserved and not started yet due some retests and 2 times server crash :(

gd_barnes 2019-12-23 19:55

[QUOTE=rebirther;533417]This can take a while. S612 still reserved and not started yet due some retests and 2 times server crash :([/QUOTE]

Sorry to hear that.

Yes it is a lot of work. No rush. S595 fits in the category of the 2018 goal of testing all bases with < 400 k's remaining to n=100K. I missed it due its larger conjecture. BOINC did a lot of testing with 300-400 k's remaining for n=25K-100K that year.

matzetoni 2020-01-12 21:15

any recommended bases soon? no rush, just asking :)

gd_barnes 2020-01-12 21:21

[QUOTE=matzetoni;534984]any recommended bases soon? no rush, just asking :)[/QUOTE]

Later today I'll post a list of 2020 project goals.

gd_barnes 2020-01-13 05:34

2020 project goals bases have been added to the 1st post in this thread. :smile:

pepi37 2020-01-13 09:46

Gary what is you recommendation for those sequence that doesnot have sieve files:from 100K to 300K
Should I make sieve in start from 100K to 1M and make deeper sieve or I should make sieve from 100K to 500K and make lower sieve depth.

Or to do mix: make sieve from 100K to 1M but with lower sieve depth and then when sieve pass 500K make additional sieve to higher sieve depth?

gd_barnes 2020-01-13 10:38

[QUOTE=pepi37;535017]Gary what is you recommendation for those sequence that doesnot have sieve files:from 100K to 300K
Should I make sieve in start from 100K to 1M and make deeper sieve or I should make sieve from 100K to 500K and make lower sieve depth.

Or to do mix: make sieve from 100K to 1M but with lower sieve depth and then when sieve pass 500K make additional sieve to higher sieve depth?[/QUOTE]


For bases currently at n=100K sieve n=100K-500K to whatever typical lower depth you've been using. With many more k's, n=500K-1M won't be needed for a long time.

For bases currently at n=200K or 250K sieve n=200K-1M or 250K-1M to the the usual 1e15.

dannyridel 2020-01-19 13:32

I really want to contribute, but I only have a small computing force. Do you have any recommended bases that have a relatively small amount of work needed to be done? For example some base with 1k and a range of about 10k?

gd_barnes 2020-03-13 03:13

8 sieve files for the aforementioned recommended bases have been added to the reservations page. :smile:

pepi37 2020-06-22 14:13

S342 7 n=100K-200K
S360 8 n=100K-200K
S446 7 n=100K-200K
S452 8 n=100K-200K
S488 7 n=100K-200K
R748 8 n=100K-200K
R752 8 n=100K-200K
R766 6 n=100K-200K
R767 7 n=100K-200K
R770 6 n=100K-200K
R830 6 n=100K-200K
R894 7 n=100K-200K
R926 7 n=100K-200K
R994 6 n=100K-200K
R1009 7 n=100K-200K


All listed bases sent to sieving in range 100K to 500k
Sieve depth 5e14

gd_barnes 2020-07-02 06:55

Pepi,


Are there any sieve files that are complete from this recommended list? My personal server will run out of work in < 2 days.


Gary

pepi37 2020-07-02 07:35

[QUOTE=gd_barnes;549596]Pepi,


Are there any sieve files that are complete from this recommended list? My personal server will run out of work in < 2 days.


Gary[/QUOTE]
S342 will be over in three days
S360 is just started..

dannyridel 2020-07-08 11:33

[QUOTE=pepi37;549597]S342 will be over in three days
S360 is just started..[/QUOTE]

I'm starting to see S452 sieve tasks now. Been sieving a lot at yoyo recently. (well, a lot for me!)

pepi37 2020-07-08 11:41

[QUOTE=dannyridel;550010]I'm starting to see S452 sieve tasks now. Been sieving a lot at yoyo recently. (well, a lot for me!)[/QUOTE]
In next two weeks all will be over :)

MyDogBuster 2020-07-09 14:40

No hurry now. Gary is on vacation out of town for at least another 10 days.

gd_barnes 2020-07-10 03:41

I can still see you! :smile:

gd_barnes 2020-07-21 00:06

Some additional sieve files have been posted for these recommended bases.

pepi37 2020-07-22 21:47

[QUOTE=gd_barnes;551133]Some additional sieve files have been posted for these recommended bases.[/QUOTE]
When this batch is over what is next recommendation bases?

gd_barnes 2020-07-22 23:21

1. Finish all bases with 6, 7, and 8 k's remaining for n=100K-500K.

2. Bases with difficulty < 25000 (some of them will overlap with #1 or #3) for n=??-500K.

3. Bases with 2 k's remaining for n=200K-500K or 200K-1M.

gd_barnes 2020-07-29 02:46

More sieve files posted. :-)

pepi37 2020-08-01 23:23

Sent for sieving
R320
R337
R394
R409
R534
R588
S961
S988

Still sieving ( will be over in few days)
R830
R894
R926
R994

Sieving over ( sent to Gary)
R 770

dannyridel 2020-08-03 03:05

Hi pepi, I'm somehow getting base 212 k=6 tasks, is that just a leftover from before?

pepi37 2020-08-05 07:46

[QUOTE=dannyridel;552375]Hi pepi, I'm somehow getting base 212 k=6 tasks, is that just a leftover from before?[/QUOTE]
From where you get it?
That is my reservation

dannyridel 2020-08-10 04:33

[QUOTE=pepi37;552597]From where you get it?
That is my reservation[/QUOTE]

I just got it from yoyo. Sieve tasks.

pepi37 2020-08-10 07:26

[QUOTE=dannyridel;553096]I just got it from yoyo. Sieve tasks.
Hi pepi, I'm somehow getting base 212 k=6 tasks, is that just a leftover from before? [/QUOTE]


Please give me clear answer
You write -I'm somehow getting - if you are getting-someone is sent to you those files-who?

pepi37 2020-08-19 22:04

New bases sent for sieving ( 2 k remain)
Sieve depth 5e14


S365
S376
S394
S444
S447
S684
S753
S766
S774

gd_barnes 2020-10-18 02:22

Pepi,

I see that most of the files for bases in the 1st post have been sent to Yoyo for sieving.

Can you go ahead and also send sieve files for R887, S368, and S994 to Yoyo?

The target is to have them all sieved by mid to late November so that we have time to get them all tested by the end of the year. :-)

Thanks!

pepi37 2020-10-23 07:34

[QUOTE=gd_barnes;560179]Pepi,

I see that most of the files for bases in the 1st post have been sent to Yoyo for sieving.

Can you go ahead and also send sieve files for R887, S368, and S994 to Yoyo?

The target is to have them all sieved by mid to late November so that we have time to get them all tested by the end of the year. :-)

Thanks![/QUOTE]




No problem Gary
Will be sent this weekend.
P.S I noticed Yoyo sieving speed is "slower" then usual, but end of the Year, yes, all will be done by that date!

gd_barnes 2021-01-01 21:54

I have now posted the recommended bases that correspond to the 2021 goals.

pepi37 2021-01-01 21:57

[QUOTE=gd_barnes;567980]I have now posted the recommended bases that correspond to the 2021 goals.[/QUOTE]

Ok, lets start sieving to make those goals. 😊

gd_barnes 2021-01-01 22:04

[QUOTE=pepi37;567981]Ok, lets start sieving to make those goals. 😊[/QUOTE]


Sieve all unsieved recommended bases if the resources are there! :-)

rogue 2021-04-06 18:55

I'm hoping that someone starts sieving the conjectures at n=100K which have only 9 or 10 remaining k.

gd_barnes 2021-04-06 19:27

[QUOTE=rogue;575347]I'm hoping that someone starts sieving the conjectures at n=100K which have only 9 or 10 remaining k.[/QUOTE]
That was suggested to Pepi in the sieving thread. Our big sieve for goal #4 was holding it up. It shouldn't be much longer before it starts.

pepi37 2021-04-06 23:28

When started, will be done quickly so you should always have enough data for processing

gd_barnes 2021-05-06 19:41

Some sieve files have been posted for the 9-10k conjectures. :-)

pepi37 2021-05-13 20:53

R536sieve.txt
R558sieve.txt
R567sieve.txt
S908sieve.txt


sent to Gary just now :)

pepi37 2021-05-13 21:30

All recommended bases 100-200 K are sent to sieve ( or already in sieve process)

gd_barnes 2021-05-22 19:38

More sieve files added.

pepi37 2021-05-22 22:19

R298 n=200K-300K
R315 n=200K-300K
R319 n=200K-300K
R424 n=200K-300K
R439 n=200K-300K
R471 n=200K-300K


Next step is sending those sequence for sieving, and after this is done, I expect from Gary new recommended bases. Also , as always is anyone here has some personal wish, can contact me via PM.

pepi37 2021-05-30 17:58

Gary, all recommended sieve files are prepared and sent to Yoyo.

Awaiting further instructions! :)

pepi37 2021-06-05 23:26

In next few days I will sent another set of data for sieving.
If somebody has any wishes let me know, else, I will take random bases without any sieve files and make them.

gd_barnes 2021-06-06 06:40

[QUOTE=pepi37;580089]In next few days I will sent another set of data for sieving.
If somebody has any wishes let me know, else, I will take random bases without any sieve files and make them.[/QUOTE]

If you see any 1k bases that are not sieved to n=1M, that would be the first priority.

Another area to sieve would be bases with 11 or 12 k's remaining at n=100K. You could sieve those for n=100K-500K. If there are too few of those you could go up to <= 15 k's remaining at n=100K. Eventually those will make their way onto the project goals list in future years.

Thanks!

pepi37 2021-06-06 11:21

[QUOTE=gd_barnes;580115]If you see any 1k bases that are not sieved to n=1M, that would be the first priority.

Another area to sieve would be bases with 11 or 12 k's remaining at n=100K. You could sieve those for n=100K-500K. If there are too few of those you could go up to <= 15 k's remaining at n=100K. Eventually those will make their way onto the project goals list in future years.

Thanks![/QUOTE]


t16_b117_k1474.npg
t16_b257_k40.npg
t16_b317_k44.npg
t16_b318_k89.npg
t16_b430_k15321.npg
t16_b497_k8.npg
t16_b789_k96.npg
t16_b883_k66.npg


Sent for sieving!

gd_barnes 2021-06-13 06:20

I have added R578 to n=200K to the recommended list. This base is low-weight and is the lowest "relative difficulty" base that we have that is not in the 2021 goals. Therefore it can easily be searched higher.

I am not adding it to the 2021 project goals.

Pepi, please have this base sieved by Yoyo to n=500K when you get a chance. Thanks.

pepi37 2021-06-15 11:17

[QUOTE=gd_barnes;580851]I have added R578 to n=200K to the recommended list. This base is low-weight and is the lowest "relative difficulty" base that we have that is not in the 2021 goals. Therefore it can easily be searched higher.

I am not adding it to the 2021 project goals.

Pepi, please have this base sieved by Yoyo to n=500K when you get a chance. Thanks.[/QUOTE]


Noticed!
I will sent that in next few days!


All times are UTC. The time now is 08:50.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.