mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Open Projects (https://www.mersenneforum.org/forumdisplay.php?f=80)
-   -   Very Prime Riesel and Sierpinski k (https://www.mersenneforum.org/showthread.php?t=9755)

robert44444uk 2008-01-27 13:45

A good week
 
Found the following VPS for Riesel series:

R 2502262772385 60 100/9732 100/10000
R 2781948761147 60 100/9271 100/10000
R 2342836014713 60 100/6520 109/10000 122/19093 !!! Reserved by me
R 1746492605077 60 100/8014 102/10000
R 2256305169303 60 100/9821 100/10000
R 2319747344799 60 100/9595 102/10000
R 2476995416951 60 100/9951 100/10000
R 2421745267415 60 100/9769 100/10000

Near misses:
R 1907954478839 60 99/10000
R 2002078724075 60 99/10000
R 1965188880093 60 99/10000
R 2183225888567 60 99/10000

There are now 92 known Riesel VPS

robert44444uk 2008-01-28 12:28

Sorry for post 10 and 12, the correct multiplier M(x) is 67 and 61, and not 66 and 60 as posted.

kar_bon 2008-01-28 13:59

i would like to make a page with all Riesel-type payams with all known primes, the searched ranges (highest n), contributor and some other datas.
currently i'm working on R 29979474409 67 (data from here [url]http://www.primepuzzles.net/puzzles/puzz_006.htm[/url]).
what i need too is the factored k (here k=29979474409=191*6217*25247) and e(67)=3*5*11*13*19*29*37*53*59*61*67.
i tested upto n=92k for now:
103 primes upto n=10k (100th at n=5968) and 129 primes at all at n=74659.
i could include this page at [url]www.rieselprime.org[/url] then.
tell me your opinion.
Karsten

roger 2008-01-29 02:05

That makes sense Karsten,

AFAIK, there aren't too many people working on these series at the moment, and a formal webpage could bring in more members (and their computers :wink:)!

I'm working on the Proth side, but I'd be happy to send you my work if you include both plus and minus payam levels. (Now if I just had some VPS's :ermm:)

Good luck to you!

roger

EDIT: Also working on ep83 now. Have sieved to y=100B, and am currently testing the rest.

kar_bon 2008-01-29 03:25

to see how i could look alike see here:
[url]www.rieselprime.org/test/SummaryRieselPayam.htm[/url]

only a first try, some background infos to display and many many more data to collect.
so i need all data i can get especially [b]all[/b] prime n!

karsten

roger 2008-01-29 03:41

What are the highlighted and bold entries? I'm guessing the ones in a box are the 100th prime.

kar_bon 2008-01-29 10:35

oh sure, sorry, look here for the 'syntax' i used in the summary-pages at all:
[url]www.rieselprime.org[/url] and the left menu -> "Infos Data Section (26.01.08)".
all colors and special things are described there!
karsten

robert44444uk 2008-01-29 12:22

I think it is a good idea to start that sort of page. Up to now we have been playing around at each of the Payam levels without any real purpose other than to find VPS.

The problem is with the numbers of Riesels to put up. I have tested in excess of 10 billion y values (!!!), so putting up everything is not really feasible!!!

The VPS has the advantage of being easily understood - if there are at least 100 primes by n=100 then it is very special - that is not to say that 99/10000 is not special as well, it is only one behind. Can I suggest that the page puts all Riesel VPS up and others such as the M(163) example which are the best at their level. I have not kept records of primes found, only prime counts, so the almost 100 values would need to be run again.

A second page could capture all y*M(x) which are less than 2^40, with a slightly lower threshhold of primes by 10000, as these are of interest to the very active LLRers. It could form the basis for a future drive. I am thinking this threshhold should be maybe 85 or 90 primes by 10000.

We could also create a page for Proths - the problem here is the appetite for people to reserve values, there are 175 known VPS, tested ranges are unknown (I found many of them but did not keep a record of tests) and many have been taken to high values of n, where testing with pfgw is painfully slow. However LLR- capable Payam Proth values have not really been explored at all, to my knowledge. That would be very interesting area of exploration, using payambase.

Finally, I have kept extensive up to date records of "drag racing" Riesels, I have the best performer at each E level, and each prime found. Similar tables for Proths need to be built.

There are gaps in our knowledge. I have sketchy information on Riesels tested back in 2003 which are in my table, but for which prime information is lacking. The information on these values would need to be rebuilt.

As a side exercise it would be interesting to find the least prime Payam. I have an inkling that there are no covering sets that do not contain the small primes or primes with low order base 2. The worst performing Payam number has its first prime at about n=230, this would be an interesting search.

kar_bon 2008-01-29 13:11

that's good to hear.
in the first step i think i could include all Riesel-VPS in that page with the information you/we got: especially the 100th prime (near misses VPS are welcome too) and/or number of primes to a level say n=10k. to get these prime n could be done time after time. reservations should also be displayed like now. then it would be helpful to make a extra page with the best candidate of every E-level with only small infos.
others information could useful: like your mentioned depth of k you searched at an E-level so noone won't do double work in that ranges.
the sierpinski-side of payam numbers could be created later (by now there're not so much results at all).
so i think i could start with the informations in this thread to extend the page and see how it looks like.
the format of report "R 1043827764761 53 100/8888 103/10000" is good, keep it, so there are infos to put in that page immediately.
Robert, i got your tables you sent me time ago. these data i can use too.
so quite now many work to do first. but you can send me more data worthly to put on!
karsten

PS: got anyone admin-rights to this thread? i think the posts could be deleted then if the infos are in that page and keeps this thread clear. perhaps the post #1 could be extended with more infos and link to the site (when 'official').

kar_bon 2008-01-31 00:12

the first more data are online now. many will follow. only Riesel numbers from this thread so far.
look the link i gave in post 16.
karsten

robert44444uk 2008-02-02 06:39

Karsten, Roger et al

I am running a huge batch of 61's and will find approx another 10 VPS. Will post these later and then will post the complete list of VPS Riesels to this forum.

Regards

Robert


All times are UTC. The time now is 18:44.

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