I was testing 9,028,373 from a worktodo.ini file that said
Test=9028373,63
That is all the worktodo.ini file said. I use manual prime retrieval and
asked for, and got from George, three new numbers as 9,028,373 was my only
number.
The worktodo.ini file was changed to
Test=9028373,63
Test=new number 1,60
Test=new number 2,60
Test=new number 3,60
There are line feeds at the end of the ,63 and ,60 so there are no extra
blanks anywhere in the worktodo.ini file. The changes were made while
9,028,373 was running with notepad or one of the other simple MicroSoft
utilities. The changes were tehn saved and the worktodo.ini file has the
four numbers.
Suddenly, 9028373, which was at 14% done, disappeared!! In its place,.the
P-III (450 MHz) was factoring new number 1 (pass 1 of 16). There is
nothing, nada, zilch, zippo, about the loss of 9028373 in the results file.
Why? It is true that I briefly shut down as I often do to clear Java files,
but I invariably go to the LL test screen, click for STOP and then click for
EXIT. At startup, the LL testing starts up again automatically (thanks to
some one who gave me fine explicit instructions as to how to do this). I
believe the STOP and then EXIT saves the latest p and q files so I lose only
a few minutes, not up to 1/2 hour or whenever the last p and q files were
saved.
I have often used Norton Utilities to (a) system check for errors and there
is a persistant error that crops up from playing Java based chess (after 15
or so games; it takes that many to win just once) and to (b) speed disk,
(another Norton Utilities feature), i.e., reorganize the C drive. But I
have been doing this for MONTHS AND MONTHS while running LL tests (more than
250 done over the years) and have never had an ill effect.
I will have to restart 9028373 but then, why did it stop? Is this a
microsoft feature (i.e. a bug that in an undocumented yet powerful add on)?
Also, should I just stop the factoring passes and restart so that I am back
again on prime 9028373?
The MERSENNE directory where I have all the files and programs has a
p9028373 file, a q9028373 file and a p file for the new number. I'd like to
resume on the p and q files for 9028373, obviously.
Anybody can respond. Ask for more info if you need it.
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
