----- Original Message ----- From: <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Thursday, November 15, 2001 5:38 AM Subject: Mersenne: Fw: The Mersenne Newsletter, issue #18
> Brian Beesley wrote: > > Eh? Doesn't it make more sense to concentrate on factoring > > Mnumbers that haven't yet been L-L tested? That way "success" in > > finding a factor reduces the number of LL tests, as well as > > (eventually) the number of double checks. Not necessarily. The marginal benefit/cost ratio of doing factoring work on exponant x awaiting a first time test, is twice what it would be if exponant x were awaiting a DC. This does not mean that it is greater that doing factoring work on exponant y which is awaiting a DC. > Surely you don't mean to suggest that someone who receives a PrimeNet > double-checking assignment that starts with some trial or P-1 > factoring > should stop, return that DC assignment to PrimeNet, then specifically > request an assignment of factoring a Mnumber that hasn't yet been L-L > tested, because that would make more sense than doing the second round > of factoring on the once-L-Led Mnumber, do you? :-) Ideally, the program 'should' only do factoring work up to the point where the benefit/cost ratio equals 1, which means that it 'should' factor to a lower level for DCs than for comparible 1st time tests. The amount of factoring that actually has been done might be lower still. Whether the program actually works like this is another matter entirely. > Another way of looking at what those to whom I referred are doing is > that we (I'm there)... As am I. > ...are performing the extra-factoring portions of > potential future DC assignments. If we don't do that, whoever gets > the future DC assignment will do it, Perhaps they won't. What I'm doing at the moment is collecting heaps of DC exponants, P-1 factorising them, then returning them un-DCed to the server. My reasoning is that many of the machines doing DCs will be older computers with relatively small amounts of memory. These will not be able to do stage 2 P-1 factorisation as effectively as I can with 512MB RAM, if they can do it at all. Therefore I am doing work which might not otherwise get done. [...] > As an old punchline goes: "If you don't, someone else will!" Ahem. Or not. > Richard B. Woods Daran G. _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
