Bases 4-32 reservations/statuses/primes

[gd_barnes]

Originally Posted by Siemelink

Ah yes, I forgot to check your sieve files, I did download them. Can you put up a link to the sieve file from the reservation page? Then we would have a practical place to keep the sieve files.
I like sieving so I'll switch to another. I'll take 258*27^n-1

Laters, Willem.

OK, I can do that. There's only 1-2 left after you took most of them but I'll still do it because we'll have many in the future. Good idea.

If you like sieving, would you be willing to help me do a large sieve on Sierp base 16 for all remaining 55 k's for a team drive like is done at Riesel Prime Search (RPS)? We could split the duties sieving to n=100K. I could then divide up the file into smaller pieces like they do at RPS so that everyone can have a manageable portion of a relatively large effort.

If so, I'll post a new thread called "Conjectures 'R Us 1st team drive".

Edit: I will unreserve 288*13^n-1.


Gary

[tcadigan]

Originally Posted by gd_barnes

If you'd like to pick up a smaller search effort that also LLR's as fast as base 2, we've got 55 k's for Sierp base 16, all tested to just n=25K that are ready for testing. At some future time, we also want to start on Riesel base 16 from the beginning.


Thanks,
Gary

actually I think I'll take you up on that. I need a break from the base 4 stuff for a bit anyway. I'll do 7773 until n=100K base 16 attached is the latest dat file for my base 4

[gd_barnes]

Originally Posted by tcadigan

actually I think I'll take you up on that. I need a break from the base 4 stuff for a bit anyway. I'll do 7773 until n=100K base 16 attached is the latest dat file for my base 4

Great, we'll reserve it for you. And thanks for sharing your sieved file. I know of a person who might want to work on it for a little while. Would you be OK with that or is your break from it temporary? Since it's officially the 'Sierpinski Base 4' project, I would inform him to report it as such if he found a prime.

In the near future, I think I'm going to coordinate a team 'drive' so to speak. We will sieve all of the k's on Sierp base 16 together that haven't been already reserved. Then we can divide up the sieved file into n=~5K 'chunks' that contain all k's.

What's great about team efforts is that after we're done sieving, people can just pick up very small or very large range of the already sieved files depending on their tastes and resources and still be searching multiple k's at once. I like them because they 'spread the risk around' so to speak. Like what happened to you on base 4 Sierp, it's easy to get stuck on one k that just refuses to yield a prime.

I'll hold off just a little while on the team drive and see if others might pick up a k or two more on base 16 for themselves. I think it's important to allow people the individuality of searching a k by themselves in addition to the team efforts that allow efficient sieving and searches across many k's.

Edit: You may want to post your break from Sierp Base 4 in that thread also. Alternatively if you and Jean Penne prefer to abandon that thread and report statuses here, that works for us and saves me from having to check it occassionally. If you both agree, I'll just have Citrix lock the thread after a period of time.


Gary

[gd_barnes]

Carlos (em99010pepe) asked for some sieved files that we have for the effort so that he could LLR them. I sent him Sierp base 12 and Riesel base 13 (one k on each remaining) and so am reserving them for him up to n=100K.


Gary

[em99010pepe]

When I free up some machines (7 cores, 1 already working) I'll put them on here to give you guys a little boost....

[gd_barnes]

Originally Posted by Citrix

Gary, Could we have a separate webpage for powers of 2. I think these will be the most popular.
By the way the webpage looks really nice.

Originally Posted by Siemelink

Hi Gary,

All your coordination effort is really nice. I checked out the webpages, they are solid stuff. I just had one minor thought, is it possible to keep the headers of the table in view when scrolling down?
Laters, Willem

A follow-up on these requests. After thinking about the powers of 2 request, I think it as well as the headers request are good ideas. I'm a bit of a newb at HTML and it is very strange attempting to keep the headers on when scrolling down. After getting a response from Karsten (Kar_bon) on how to do that, I should be able to do it. I'll work on both of these later this weekend.

FYI, I've analyzed powers of 2 up to 256 now. The conjectures on base 64 are low at k=14 and k=51 and were easily proven. On base 128, they're low with both at k=44 but I wasn't able to quite prove them. Riesel had one k remaining at n=5K. Sierp had 4 k's remaining at n=40K (searched to n=5K and converted higher ones from base 2) but 3 of them cannot be proven for the same reason as k=4 on Sierp base 32. So no good for those.

I think that base 256 may be the next power of 2 effort after bases 4 and 16. It has a Riesel conjecture of k=10364 and a Sierp conjecture of k=1221. Sierp's is a little low but Riesel may yield some k's to search very high. Ultimately I'll check it to base 1024 or possibly 4096 but I see no reason to go any higher. There may not be any that are good > base 256.

Based on almost split poll results, I think it may make sense to simultaneously do some powers of 2 while continuing to fill in bases <= 32.

Although there isn't a separate page for powers of 2 yet, I will add base 64 in just a little while and add base 128 later today (Friday) on the current pages.

If you haven't voted on the direction of the effort yet, get your voice heard! It is the poll in this forum. It's not public so I have no idea how people vote.


Gary

[gd_barnes]

Originally Posted by em99010pepe

When I free up some machines (7 cores, 1 already working) I'll put them on here to give you guys a little boost....

Hah! A LITTLE boost! Everyone, with a couple of new quads running 50% faster than anything I have, this guy eats sieved files for lunch and dinner too!

I'll see if I can get some sieving going on for ~45-50 of the base 16 k's this weekend as the beginning of a team effort. I want to save a few for individual efforts. I think he said it'd be ~2 weeks before the other cores free up.


Gary

[tcadigan]

Originally Posted by gd_barnes

Great, we'll reserve it for you. And thanks for sharing your sieved file. I know of a person who might want to work on it for a little while. Would you be OK with that or is your break from it temporary? Since it's officially the 'Sierpinski Base 4' project, I would inform him to report it as such if he found a prime.

...

Edit: You may want to post your break from Sierp Base 4 in that thread also. Alternatively if you and Jean Penne prefer to abandon that thread and report statuses here, that works for us and saves me from having to check it occassionally. If you both agree, I'll just have Citrix lock the thread after a period of time.


Gary

let'em have at it! I'd be more interested in seeing it fall to finding a prime than selfishly holding onto it.

as for the thread closure...I'm fine either way. I'd just want to make sure that the files don't get lost along the way. I think there's relatively recent newpgen's for all k's and even 2 different dats attached throughout the thread (1 for original search, 1 for expanded search)

[gd_barnes]

I have completed testing the one k remaining on Sierp base 11 to n=100K and the two k's remaining on Sierp base 23 to n=60K. They are now released for testing.

Carlos, Willem, or whomever wants to test Sierp base 23 further, I'm attaching the remainder of my sieved file up to n=100K to this post. It's well sieved to P=600G.


Gary

[Siemelink]

Originally Posted by gd_barnes

I have completed testing the one k remaining on Sierp base 11 to n=100K and the two k's remaining on Sierp base 23 to n=60K. They are now released for testing.

Carlos, Willem, or whomever wants to test Sierp base 23 further, I'm attaching the remainder of my sieved file up to n=100K to this post. It's well sieved to P=600G.


Gary

Yeah, that's for me. I started it yesterday already at 60k. I am now off for the holidays so I wonder what I'll find when I return.

Laters, Willem.

[gd_barnes]

Originally Posted by Siemelink

Yeah, that's for me. I started it yesterday already at 60k. I am now off for the holidays so I wonder what I'll find when I return.

Laters, Willem.

Very good. Got ya down. Maybe you'll find a top-5000 prime when you return! n>=73.7K would do it for base 23.

Note for everyone: As you probably well know, this is not an official prime-search project so any top-5000 prime correctly reported should not include a project ID.


Gary