At 02:46 AM 12/4/2006, Brian Beesley wrote: >In any case >a detected roundoff error of 0.5 cannot be reliably recovered from as it's >impossible to guess accurately whether the correct integer is the "floor" or >the "ceiling".
Actually, prime95 recovers from roundoff errors up to 0.6 (by redoing the iteration using a safer, but 3 times slower, method). I think Brian is trying to say that if you do no roundoff error detection and recovery, then a program cannot recover from a roundoff error above 0.5. >There is however a problem with some exponents when switching between >computers - even with the same software version. If you switch from a >non-SSE2-capable processor type to a SSE2-capable processor type, and the >exponent is between the SSE2 and non-SSE2 run length limits, you will almost >certainly end up with excess roundoff errors, and very likely with a wrong >residue - even if all the roundoff errors were apparently recovered. Prime95 should handle this OK. It will convert the save file to the FFT length appropriate for your machine. _______________________________________________ Prime mailing list [email protected] http://hogranch.com/mailman/listinfo/prime
