PFGW 4.0.4 (with gwnum v30.10) Released

[earshot]

powmod via script gives me incorrect results when it starts using Generic modular reduction. I was learning using script en scriptify and started by making a Mersenne Prime script. It first checks the candidates by powmod base 3 and when it is successful it runs a complete test. It finds all the Mersenne primes up to 607 but for p>=647 modular reduction is used and incorrect results are generated. Is this a known bug or am I doing something wrong?

[paulunderwood]

The rep-unit function "r()" prints a lot of extranious spaces after the number.

For example:

./pfgw64 -f0 -od -q"r(5794777)" | wc
PFGW Version 4.0.1.64BIT.20191203.x86_Dev [GWNUM 29.8]

No factoring at all, not even trivial division
                     
      1       2 5794826

[ATH]

This works, it gives a file with 5794777 1's:

./pfgw64 -f0 -od -q"r(5794777)" | tail -c +13 | head -c 5794777 > rep5794777.txt

The number 13 in "tail -c +13" has to be 6 more than the number of digits in 5794777.

[Puzzle-Peter]

Dear all,


Norman Luhn (Cybertronic) and I are trying to prove the following number using PFGW:


(520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1


there is a file helper.txt which contains
520461*2^55931+1
2584831267



The command we are using is

pfgw -tc -l -hhelper.txt <inputfile>


I tried the following PFGW versions:
Linux Version 3.8.3 64bit
Linux Version 3.7.8 64bit
Windows 4.0.0/1 64bit (Help - About says 4.0.0, onscreen it says 4.0.1)
Windows 3.6.5 32bit and 64bit


All tests run on Intel CPUs



They all gave me "is prime" with an output that looks like this:

Primality testing (520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 [N-1/N+1, Brillhart-Lehmer-Selfridge]
Running N-1 test using base 5
Running N-1 test using base 13
Running N+1 test using discriminant 29, base 3+sqrt(29)
Calling N-1 BLS with factored part 33.33% and helper 0.02% (100.01% proof)
(520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 is prime! (236.0713s+0.0011s)

Norman can't reproduce the primality on his AMD machines, he only gets Fermat and Lucas - PRP.


We tried to determine what's going on but we can't pinpoint the issue. I have no AMD, he has no Intel machines. Could some people please try and post PFG Version, CPU, OS and result? this test only takes a few minutes to run.


Thanks a lot
Peter

[Cybertronic]

Additions:


I get also on a Laptop SONY Vaio Intel processor "Fermat-Lucas-PRP"


PFGW Versions: Very old up to V4.0 32/64bit

OS Windows 10 / Windows 7


command line
pfgw2 -tc -hxyz cs




Output: OS Windows 10 , Ryzen 7 1700 3GHz

Z:\PFGW>pfgw2 -tc -hxyz cs
PFGW Version 1.2.0 for Windows [FFT v23.8]


***WARNING! file cs may have already been fully processed.

Primality testing (520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 [N-1/N+1, Brillhart-Lehmer-Selfridg
Reading factors from helper file xyz
Running N-1 test using base 5
Running N-1 test using base 13
Running N+1 test using discriminant 29, base 3+sqrt(29)
Calling N+1 BLS with factored part 0.02% and helper 0.00% (0.07% proof)
(520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 is Fermat and Lucas PRP! (903.1017s+0.0100s)




xyz:

numbers
520461*2^55931+1
2584831267

[kruoli]

Puzzle-Peter:
Calling N-1 BLS with factored part 33.33% and helper 0.02% (100.01% proof)
Cybertronic:
Calling N+1 BLS with factored part 0.02% and helper 0.00% (0.07% proof)

The program is doing different tests; the last one is definitely the wrong "tool" for this number, so it is not able to detect the number being prime. But the question remains, why it executes a different tests here…

[paulunderwood]

Originally Posted by Cybertronic

xyz:

numbers
520461*2^55931+1
2584831267

Does xyz contain the word "numbers"?

[Cybertronic]

Only this two lines


520461*2^55931+1
2584831267

[paulunderwood]

One thing to try: rename "xyz" to "helper" and run with -hhelper

[Cybertronic]

Okay, I will try -hhelper


Other files in use was -hhelper.txt , without success

[axn]

~/Desktop/pfgw_linux_4.0.1/distribution$ ./pfgw64 -tc -l -hhelper.txt input.txt 
PFGW Version 4.0.1.64BIT.20191203.x86_Dev [GWNUM 29.8]

Output logging to file pfgw.out
Primality testing (520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 [N-1/N+1, Brillhart-Lehmer-Selfridge]                                    
Reading factors from helper file helper.txt
Running N-1 test using base 5                                                  
Running N-1 test using base 13                                                 
Running N+1 test using discriminant 29, base 3+sqrt(29)                                    
(520461*2^55931+1)*(43439253939*(520461*2^55931-1)^2-3)+1 is prime! (294.1606s+0.0005s)                                    

~/Desktop/pfgw_linux_4.0.1/distribution$ uname -a
Linux R5 5.4.0-88-generic #99-Ubuntu SMP Thu Sep 23 17:29:00 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

~/Desktop/pfgw_linux_4.0.1/distribution$ lsb_release -a
No LSB modules are available.
Distributor ID:	Ubuntu
Description:	Ubuntu 20.04.3 LTS
Release:	20.04
Codename:	focal

~/Desktop/pfgw_linux_4.0.1/distribution$ lscpu
Model name:                      AMD Ryzen 5 3600 6-Core Processor