manual results reporting bug for P-1 factoring
 

I have seen this bug several times; the manual reporting parser appears to not be able to pick up the B1 number.

(Results in the text file):
[Sun Dec 06 08:47:46 2009]
P-1 found a factor in stage #1, B1=600000.
UID: dgould, M51415913 has a factor: 3956827154570882591527327, AID:
19BBEEB4276C4E877A321773EDD2ADB5

(log received back at submission):
Factors found: 1
Processing result: M51415913 has a factor: 3956827154570882591527327, AID: 19BBEEB4276C4E877A321773EDD2ADB5
Insufficient information for accurate CPU credit.
For stats purposes, assuming factor was found using P-1 with B1 = 800000.
CPU credit is 2.4586 GHz-days.

[QUOTE=dgould;198256]I have seen this bug several times; the manual reporting parser appears to not be able to pick up the B1 number.[/QUOTE]
It doesn't actually pick up that it's P-1, or whatever type of factoring it was. Only that a factor was found.

This has come up on the forums before, but the posts elude me.

At least when a factor is found in stage 1 it usually gives a bit more credit than is due. If a factor is found in stage 2 it still gives credit as though it was in stage 1... c'est la vie.

If you want your real credit do not use the manual pages. Move the prime.spl file from the box with no access to the GIMPS server to a computer that has. The latter computer will report the results for the other. Just make sure you do not overwrite an existing prime.spl file.

Jacob

In my case I'm making several P-1 runs with numbers in the range 253000-254000 using B1=50M and B2=1000M. The first success was at exponent 253367 where the factor 428118424378877527039271 was found on step 1.

The system says that the computer found it by ECM using B1=50000.

This seems to be the same (or closely related) problem discussed in

[URL]http://mersenneforum.org/showthread.php?t=12818[/URL]

and posts #130-133 in

[URL]http://mersenneforum.org/showthread.php?t=11101[/URL]