|
BGonline.org Forums
41$-44, data
Posted By: neilkaz In Response To: 41$-44, data (Nack Ballard)
Date: Friday, 25 December 2009, at 5:11 p.m.
After 8744 games (I stopped for a while to work on another project needing XG) D leads P by .011.
I still think D is correct and noting that it brings two new checkers into play strongly. Additionally, at high levels of tourney play, most players don't slot 41 when the score is close, but may slot it when trailing substantially. With a substantial lead at most match score,s cubeful considerations favor the play that wins the most games...ie D.
David Rockwell has stated that his research has shown GNU 2 ply passing too soon in variations after P. I did a quick check of XG 4 ply, and didn't find it making any serious cube errors after P but once I finish this, I'll have to spend a couple hours looking at some more variations (and doing some 3 ply RO's (not very long on this comp) to catch any big errors if there are any.
Since it takes a bit less than 18 hours of CPU time for me to get 5184 of these 4 ply RO's done for both plays, I'll end up doing 20k likely finishing this weekend.
Those with Snowie on new computers can feel free to compare that time vs the time needed for Snowie 4 to do 5184 3 ply precise RO's of both.
Merry Xmas .. neilkaz ..
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.