Unexpected termination of PM-1

[Miszka]

I'm using Prime95v. 30. 7b4
During implementation:
Pminus1=N/A,1,2,3779,-1,40000000000,2000000000000,68,"7559,2086009,6084191,274476660553,5086072485847,6306850735777,3476227576267836521,2078022893280176212649"
The program aborted after finishing the first stage and calculating B2=999*B1=39960000000000 without any message, writing to the log, or sending the result to the GIMPS server (See Photo1).

I restarted the program forcing the value B2=2000000000000
Pminus1=N/A,1,2,3779,-1,40000000000,2000000000000,"7559,2086009,6084191,274476660553,5086072485847,6306850735777,3476227576267836521,2078022893280176212649"
The program has again terminated unexpectedly (See Photo2).

Perhaps the B2 value used (calculated) is too large and exceeds the allowable limit?
I don't know, because there is no announcement.

[Viliam Furik]

Originally Posted by Miszka

I'm using Prime95v. 30. 7b4
During implementation:
Pminus1=N/A,1,2,3779,-1,40000000000,2000000000000,68,"7559,2086009,6084191,274476660553,5086072485847,6306850735777,3476227576267836521,2078022893280176212649"
The program aborted after finishing the first stage and calculating B2=999*B1=39960000000000 without any message, writing to the log, or sending the result to the GIMPS server (See Photo1).

I restarted the program forcing the value B2=2000000000000
Pminus1=N/A,1,2,3779,-1,40000000000,2000000000000,"7559,2086009,6084191,274476660553,5086072485847,6306850735777,3476227576267836521,2078022893280176212649"
The program has again terminated unexpectedly (See Photo2).

Perhaps the B2 value used (calculated) is too large and exceeds the allowable limit?
I don't know, because there is no announcement.

I had similar issues before, with small exponents and huge B1/B2. I don't remember the details exactly, but I think I had problem with B2 being too large. That's probably the reason for you too.

[Miszka]

And what's the biggest value B2 can be?

[Viliam Furik]

Originally Posted by Miszka

And what's the biggest value B2 can be?

Not sure... But it's easy to find out. Try which ones work! It's like a guessing game.

[kriesel]

Originally Posted by Viliam Furik

Not sure...

Or read the source code, perhaps while a guess runs.

[Viliam Furik]

Originally Posted by kriesel

Or read the source code, perhaps while a guess runs.

Frankly, that didn't occur to me...

[Prime95]

Originally Posted by Miszka

I'm using Prime95v. 30. 7b4

Pminus1=N/A,1,2,3779,-1,40000000000,2000000000000,68,

Prime95 is the wrong tool for the job. Use GMP-ECM.

[tuckerkao]

Originally Posted by Miszka

And what's the biggest value B2 can be?

It depends on the available memories of the user's PC. Many larger factors require the bounds to be up to 2^14 or greater which can be out of reach by many models. ECM has the higher chance to find such the useful factors, but require many curves to be run before able to find 1.

[ATH]

Originally Posted by Miszka

And what's the biggest value B2 can be?

With GMP-ECM you can run stage2 in as many stages as you want, going as high as you want. First you run stage1 and save the result:

ecm.exe -v -pm1 -save M3779stage1.txt 4e10 4e10 < M3779.txt

The "M3779.txt" file is just this line:

(2^3779-1)/7559/2086009/6084191/274476660553/5086072485847/6306850735777/3476227576267836521/2078022893280176212649

Then you can just run stage2 like this:
ecm.exe -v -pm1 -resume M3779stage1.txt 4e10 4e10-1e13 < M3779.txt
ecm.exe -v -pm1 -resume M3779stage1.txt 4e10 1e13-2e13 < M3779.txt
ecm.exe -v -pm1 -resume M3779stage1.txt 4e10 2e13-3e13 < M3779.txt
etc.

You can adjust the interval you want to run each time depending on how much memory you want to use. 1e13 takes 2784 MB. Generally I think running as high an interval as possible each time is faster than running many small intervals, but don't remember if that is always the case. I seem to remember testing than sometimes it was not.

[Miszka]

Originally Posted by Prime95

Prime95 is the wrong tool for the job. Use GMP-ECM.

I made several attempts to determine the value of B2 to complete my task.
Only B2=B1=40000000000 was successful.
In this experimental way I found out that for small Mersenne numbers the P-1 method is not a good method.

[Prime95]

Originally Posted by Miszka

I made several attempts to determine the value of B2 to complete my task.
Only B2=B1=40000000000 was successful.
In this experimental way I found out that for small Mersenne numbers the P-1 method is not a good method.

Keep the save file. Version 30.8 may be able to complete stage 2.