View Single Post
Old 2005-06-28, 08:17   #32
garo
 
garo's Avatar
 
Aug 2002
Termonfeckin, IE

22×691 Posts
Default

I have a proposal to make. Since we are way ahead of the current doublechecks, we could divert a few resources to doing some P-1 for numbers that are ahead of the first-timers. There are two problems with the proposal as I can see.

1) A P-1 on a first-timer will take a long time. More than 4 times the time.

2) With factoring and LL edges being so close, there is a chance that something gets messed up and work gets wasted. Plus we should only do P-1 on numbers that have already had the required amount of trial factoring done on them which reduces our list of candidates drastically.
garo is offline   Reply With Quote