View Single Post
Old 2005-12-06, 19:44   #3
R.D. Silverman
 
R.D. Silverman's Avatar
 
Nov 2003

22×5×373 Posts
Default

Quote:
Originally Posted by MS63

(Now I am going to show my ignorance.)

Would it possible set the LL test running twice on the same prime candidate starting at each end, as it were, and meeting somewhere in the middle?
(1) Your question is not well posed. What do you mean by "meeting in the
middle"? Explain how you think one would or could "start at each end"?
And if your answer is "I'm not sure what I mean", then my answer is that
one should never ask a question that one does not understand. Which is
what I suspect that you are doing.

One does not need to be a mathematician to apply a little common sense.
I suggest that you do so.

We have an initial starting value: L_0 = 4. We are trying to compute
a final value: L_n. How do you propose we "start at each end" if we don't
even know what the ending value is???

And it certainly is possible to do FFT's in parallel with little loss in efficiency.
However, we can do Mersenne prime testing with ZERO loss in efficiency
simply by giving separate exponents to separate machines. This is what is
done now.
R.D. Silverman is offline   Reply With Quote