Bases 6-32 reservations/statuses/primes

[gd_barnes]

Originally Posted by rogue

BTW, all base 28 k (Sierpinski) are tested to n = 100,000. I made a mistake and did not test the base 28 Riesels. You can unreserve them

I will reserve the Sierpinski and Riesel base 27 k.

Base 30 is nearing completion to n = 100,000, but is on a slower PC, so it will take a little while longer.

I am assuming that you will be unreserving the Sierp base 28 k's also. Is that correct?


Gary

[gd_barnes]

Originally Posted by KEP

Is taking riesel base 27 with k=706 with n>100K<n<1M! I'll do it this way: First sieve to 10000e9 or less on a not very dedicated machine. Then once work is only availeable to one of my dual cores I'll test those candidates remaining using either LLR or WinPFGW. This time I'll stick to my reservation, because it might keep my computer busy and it really want interfere with my initial goal of getting the base3 conjecture up to k=1e9 or k=2e9 by the end of the year :)

KEP!

Ps. Just sent on relief from hospital, as the update on the base 3 riesel conjecture attack says, so if things doesn't improve magnificantly this weekend or early next week I might not be able to bring you an update for the next week. But I guess that progress is about 10-25% for the k>2M<100M range. Thanks for your understanding!

Originally Posted by rogue

BTW, all base 28 k (Sierpinski) are tested to n = 100,000. I made a mistake and did not test the base 28 Riesels. You can unreserve them

I will reserve the Sierpinski and Riesel base 27 k.

Base 30 is nearing completion to n = 100,000, but is on a slower PC, so it will take a little while longer.

Rogue,

See the above quote. KEP has already taken Sierp base 27. Would you still like to do the remaining k on Riesel base 27 and perhaps go ahead with Riesel base 28?


Gary

[rogue]

Originally Posted by gd_barnes

Rogue,

See the above quote. KEP has already taken Sierp base 27. Would you still like to do the remaining k on Riesel base 27 and perhaps go ahead with Riesel base 28?

Gary

I missed his reservation. I'll still take the Riesel base 27. I'll also take Riesel base 26.

[KEP]

Originally Posted by rogue

I missed his reservation. I'll still take the Riesel base 27. I'll also take Riesel base 26.

OK it seems that Rogue is taking my Base 27 for Riesel. This is OK, but I'm unreserving it then. It means that I'll loose 72 hours of work, but also it means that I'll be able to finish the Base 19 sierpinski and the Base 12 for sierpinski faster. Actually I'll now split the base 12 reservation in portions and run it on 3 cores and continue running base 19 on 1 core. On a sidenote, for Sierpinski base 19 I've already found 12 primes. So only 1,600 primes to go

KEP!

[rogue]

Originally Posted by KEP

OK it seems that Rogue is taking my Base 27 for Riesel. This is OK, but I'm unreserving it then. It means that I'll loose 72 hours of work, but also it means that I'll be able to finish the Base 19 sierpinski and the Base 12 for sierpinski faster. Actually I'll now split the base 12 reservation in portions and run it on 3 cores and continue running base 19 on 1 core. On a sidenote, for Sierpinski base 19 I've already found 12 primes. So only 1,600 primes to go

KEP!

I'm confused. Were you doing Sierpinski base 27 or Riesel Base 27? I thought you were only doing Sierpinski? I certainly don't want to poach.

[gd_barnes]

Originally Posted by KEP

OK it seems that Rogue is taking my Base 27 for Riesel. This is OK, but I'm unreserving it then. It means that I'll loose 72 hours of work, but also it means that I'll be able to finish the Base 19 sierpinski and the Base 12 for sierpinski faster. Actually I'll now split the base 12 reservation in portions and run it on 3 cores and continue running base 19 on 1 core. On a sidenote, for Sierpinski base 19 I've already found 12 primes. So only 1,600 primes to go

KEP!

Originally Posted by rogue

I'm confused. Were you doing Sierpinski base 27 or Riesel Base 27? I thought you were only doing Sierpinski? I certainly don't want to poach.

I'm sorry guys. I had double-checked my web pages yesterday because there have been so many updates the last several days. They were correct but I made a complete mis-statement and reversed what I should have said to Rogue about KEP's reservation.

Even though the web pages had it reflected correctly, here is what I said:

KEP has already taken Sierp base 27. Would you still like to do the remaining k on Riesel base 27 and perhaps go ahead with Riesel base 28?

Here is what I SHOULD have said:

KEP has already taken Riesel base 27. Would you still like to do the remaining k on Sierp base 27 and perhaps go ahead with Riesel base 28?

Argh! It's heck trying to keep all of the bases straight. lol

So, KEP please keep your reservation and don't lose so much work. Rogue, if you wouldn't mind taking SIERPINSKI base 27 and leaving Riesel for KEP, that's actually how I updated the web pages yesterday. And KEP, I had still shown you as having reserved RIESEL base 27 on the pages yesterday. It was my misstatement in the thread here that caused the problem.

Rogue, based on taking Sierp base 27, do you still want Riesel base 26? Both Sierp and Riesel are available for base 26.

Sorry again...
Gary

[KEP]

@ Gary:

You're forgiven, we can all make mistakes. I'll keep my Riesel Base 27 reservation, and in seconds from now I'll start sieving again. Optimal sieve depth has been reached for Sierpinski Base 12, so I'll do LLR on 1 core. So to sum up, following is work that I do:

1. Test Sierpinski Base 12 up to n=250K or prime found
2. Test Riesel Base 27 up to n=1M or prime found
3. Test Sierpinski Base 19 up to n=100K or all k's primed

Thanks. By the way Gary, I read your statements about using NewPGen, I really has a hard time seing how sieving a fixed n using the increment can save a lot of work and help remove candidates over billions or maybe even trillions. But if you test it out using NewPGen, then please let me know. Of course its always faster if we test a trillion k to n=2,500, which would leave about 40,000,000 candidates remaining. However due to sieve limits in srsieve, the only option availeable will be to sieve and then run WinPFGW or LLR through the remaining candidates of the approximately 40M candidates remaining. This want be much of an issue, but we sure need some script or program that can remove the candidates from a main file containing all the k/n pairs remaining at n=2,500. In this main file we need to change the remaining candidates to the new n value by search and replace and then replace the sieve depth with 0, before starting a new sieve and eventually running LLR or WinPFGW, and then loop the process. Maybe I should produce a more elaborate report on my ideas, but at least I hope this made sence. I know it will require a lot of manual work though ... but it may still be faster than bringing down only a million k's to n=25,000 each day for either of the Base 3 conjectures. But I really hasn't tested that way out yet. Also we has to consider if there will be any k limits when using NewPGen like there e.g. is when using multiple k's in srsieve.

Regards

KEP

Ps. Maybe someone knows developers that can combine srsieve, WinPFGW, NewPGen, sr2sieve and other wishfull sievers, which can accept the fact that they remove primes and keeps composites and use the most nescessary codings to provide certain proves of primes aswell as fast sieving for "removing" or skipping composite k's faster and save them for testing by next n since they were not prime yet .

[rogue]

OK, then I'll take Sierpinski base 27.

[gd_barnes]

Originally Posted by rogue

OK, then I'll take Sierpinski base 27.

OK, I'll put you down for Sierp base 27 and Riesel base 26.

Thanks guys.

[KEP]

Here is 22 primes for sierpinski base 19:

389404*19^10013+1
275586*19^10026+1
364744*19^10027+1
19036*19^10034+1
339634*19^10075+1
543406*19^10106+1
554706*19^10114+1
642324*19^10133+1
741084*19^10135+1
230536*19^10138+1
480564*19^10139+1
238534*19^10143+1
178594*19^10169+1
690138*19^10169+1
583876*19^10172+1
134386*19^10184+1
557826*19^10230+1
416944*19^10237+1
651604*19^10237+1
124734*19^10271+1
592614*19^10271+1
545494*19^10313+1

I've now begun a sieving of the range up to n=25,000. I expect to see further primes faster in a week or so. But hey only 1,590 primes to go . Working hard. The Base 12 Sierpinski is at 404*12^109474+1 no prime yet. Also Base 27 Riesel is still sieving on 2 cores. Have fun and take care everyone.

KEP!

[michaf]

A small update on Riesel base 24 here:

The following k/n pairs are prime:

8076 17333
26374 17500
14199 17590
27086 17605
27656 18311
26771 18531
17254 18532
10684 18570
4659 18684
976 19189
2314 19284
10016 19775
20274 19794
31491 19985
30979 20160
21594 20298
1571 20425
8076 20445
25426 20461
15056 20863
26771 20871
10674 20912
27046 20933
21739 21168
28621 21249
10601 21603
18226 22199
16609 22354
26374 23076
12261 23247
31491 23303
10684 24180
21971 24533
11653 24904

Which mean that now a total of 170 k’s are remaining (including 2 mob’s)
One mob was eliminated: 976*24^19189-1 also got 23424*24^19188-1

I’ve checked upto 25k now