| |
BGonline.org Forums
Bot Comparison after 60/500 games.
Posted By: Daniel Murphy In Response To: Bot Comparison after 60/500 games. (neilkaz)
Date: Thursday, 29 October 2009, at 3:47 a.m.
1. Neil writes: GNU 4 ply filter needs to be expanded to choose at least top 4 2 ply plays within .08 in order to be a better analysis tool, but further slowing it down.
What is Michael D using, add two plays? Perhaps Michael should use add-four in his comparison tests?
2. We know there are types of positions in which Gnubg's odd-ply evaluations beat even-ply. I'm wondering how much better Gnubg would be if it could do mini-mini-rollouts (short and truncated!) of its cube decisions whenever odd/even ply evaluations disagree. Looks like Gnubg 4-ply and XG would be neck and neck if Gnubg 4-ply's "Missed Double" score were better.
Is such a feature worth developing?
3. Phil writes: It's also only $50, and it is being continuously improved and upgraded (which is not the case with Snowie or GNUBG).
Click on my name , and be directed to the gnubg.org bug list discussion archive, where you'll get an idea of the ongoing work being done on GNU Backgammon.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.