mersenneforum.org 1000039*2^n+1
 Register FAQ Search Today's Posts Mark Forums Read

 2021-03-15, 22:02 #1 Alex     "Alex_soldier (GIMPS)" Aug 2020 www.Mersenne.ru 2×3 Posts 1000039*2^n+1 Hello. Im working on 1000039*2^n+1 with n=[1..2.000.000] (k=1.000.039 is wellsieved). Now it was passed n=[1..1.000.000] (8150 tests). Code: 1000039*2^382+1 is prime! (121 decimal digits) Time : 119.587 ms. 1000039*2^466+1 is prime! (147 decimal digits) Time : 45.074 ms. 1000039*2^670+1 is prime! (208 decimal digits) Time : 48.150 ms. 1000039*2^1414+1 is prime! (432 decimal digits) Time : 41.688 ms. 1000039*2^8326+1 is prime! (2513 decimal digits) Time : 239.942 ms. 1000039*2^10810+1 is prime! (3261 decimal digits) Time : 390.164 ms. - First found in 1997 by Steffen Polster (SP code) 1000039*2^13102+1 is prime! (3951 decimal digits) Time : 642.368 ms. - First found in 1997 by Steffen Polster (SP code) It was no new primes Does anybody want to join me to finish the rest ~7650 tests? The range was sieved up to p=21e12 (~40 min per factor). PFGW / LLR is ~20 min per test.
2021-03-15, 23:35   #2
rogue

"Mark"
Apr 2003
Between here and the

11000100001002 Posts

Quote:
 Originally Posted by Alex Hello. Im working on 1000039*2^n+1 with n=[1..2.000.000] (k=1.000.039 is wellsieved). Now it was passed n=[1..1.000.000] (8150 tests). Code: 1000039*2^382+1 is prime! (121 decimal digits) Time : 119.587 ms. 1000039*2^466+1 is prime! (147 decimal digits) Time : 45.074 ms. 1000039*2^670+1 is prime! (208 decimal digits) Time : 48.150 ms. 1000039*2^1414+1 is prime! (432 decimal digits) Time : 41.688 ms. 1000039*2^8326+1 is prime! (2513 decimal digits) Time : 239.942 ms. 1000039*2^10810+1 is prime! (3261 decimal digits) Time : 390.164 ms. - First found in 1997 by Steffen Polster (SP code) 1000039*2^13102+1 is prime! (3951 decimal digits) Time : 642.368 ms. - First found in 1997 by Steffen Polster (SP code) It was no new primes Does anybody want to join me to finish the rest ~7650 tests? The range was sieved up to p=21e12 (~40 min per factor). PFGW / LLR is ~20 min per test.
If you have a good GPU, then you should be using srsieve2cl for sieving this sequence. If anything you might want to compare its speed to sr1sieve.

 2021-03-23, 08:37 #3 bur   Aug 2020 2·47 Posts The number of candidates for n < 1,000,000 seems so small for p < 21e12. I sieved k=1281979 up to p = 312e12 and was left with nearly 15,000 candidates for n < 1,000,000. And as rogue said, definitely use sr1sieve. It is by far the fastest for this type of search. For the Proth test, do you use LLR2? It has Gerbicz error correction which is very useful when working alone (i.e. no double checking). And best of luck to you! I tested all n < 2,500,000 on "my" k so far and the last prime occured around n = 415,000. I hope I'll find at least one more prime until the end of the sieve at n = 4,100,000.
2021-03-23, 09:00   #4
Happy5214

"Alexander"
Nov 2008
The Alamo City

5×113 Posts

Quote:
 Originally Posted by bur For the Proth test, do you use LLR2? It has Gerbicz error correction which is very useful when working alone (i.e. no double checking).
Doesn't "standard" LLR have Gerbicz error correction for Proth tests as of 3.8.24? I know it defaults to Fermat PRP tests with GBC for Riesel candidates now, so I would assume Jean's implemented it for Proth tests too.

 2021-03-23, 12:54 #5 Alex     "Alex_soldier (GIMPS)" Aug 2020 www.Mersenne.ru 2×3 Posts Thank you, rogue. Thank you, bur. Ive chosen k=1000039 because it is wellsieved. For the comparison sieving n=[1..1000000] up to p<1e6: k=1000039 has 16891 candidates k=1281979 has 39706 candidates k=1000039 has only 7 known small primes, it is not Sierpinski number. Now I have found several less candidates k, but k=1000039 was the first. I use Sr1sieve & PFGW. Last fiddled with by Alex on 2021-03-23 at 13:13
2021-03-23, 15:33   #6
rogue

"Mark"
Apr 2003
Between here and the

22·3·523 Posts

Quote:
 Originally Posted by Alex Thank you, rogue. Thank you, bur. Ive chosen k=1000039 because it is wellsieved. For the comparison sieving n=[1..1000000] up to p<1e6: k=1000039 has 16891 candidates k=1281979 has 39706 candidates k=1000039 has only 7 known small primes, it is not Sierpinski number. Now I have found several less candidates k, but k=1000039 was the first. I use Sr1sieve & PFGW.
llr is probably faster than pfgw for these numbers.