Hi,
At 12:54 AM 10/16/98 +0000, C0c0 wrote:
>This is the second time it happens with my P100, but the first one it
>didn't restarted. Maybe i should just do factoring. Anyway, i dont
>understand what kinda hardware problem can it be.
>
>Iteration: 4208297/5832553, ERROR: SUM(INPUTS) != SUM(OUTPUTS),
>1043988048484624 != 1049324106667991
>
>Sorry for my bad english, and regards from spain.
Your English is quite good. You can recover your lost work if you
have a backup of the p5832553 file, otherwise I'm afraid you've
suffered a great loss.
If you are not running version 16.x or later, I'd suggest upgrading.
Version 16's handling of intermediate files is a little more robust.
The errors you are getting could be due to an overheating CPU - try
running prime95 for a while, then open the box and see if the CPU
is too hot to touch.
The errors could also be due to flaky memory chips (either cache or main
memory). If this is the case, it will be difficult to track down.
As to switching to factoring, that is an option. Upgrading to version 17
gives you another option - double-checking. You could still find a Mersenne
prime (not likely) and the LL test will be 16 times faster than the
large exponents you are currently testing.
Sorry for the troubles,
George
