I'm also adding code to 23.5 to check EVERY iteration for an impossible result
such as -2, -1, 0, 1, 2. This test will be very, very quick.
But no matter how quick this additional test will be, it will slow down all machines globally, while a false positive report would only "slow down" two machines to double check for a limited time!!!
If we consider a double check will take 15 days on an average machine, and 15 days (in the same time) on another machine, this makes 30 days x 24 x 3,600 = 2,592,000 seconds.
While if we consider that this additional test takes only 3 minutes per exponent overall in average, there are over 40,000 machines working in the project, I don't know the exact number but if we consider half that number work on LL tests, 20,000 x 180 = 3,600,000 seconds. (There's a lot more machines assigned LL tests!)
So this makes a difference of 1,008,000 seconds and this of course just in the case a false positive was reported!!!
So I think the additional test is not needed, it would only slow down the project...
_________________________________________________________________
Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail
_________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
