I think XG has gained in popularity over gnubg mainly because it is faster, and only marginally because of the slight playing strength advantage.
And Xavier has done a great publicity job. Xavier had the advantage of being able to design for speed and multi-threading from the outset. Gnubg’s multi-threading has been patched in afterwards, where it would make the biggest effect. For, example, if I recall correctly, someone once did an analysis that showed gnubg spent a lot of time decoding and encoding the position id. The id is wonderfully compact, but saving a bit of memory probably no longer has the priority it once did. Ian From: Bug-gnubg [mailto:[email protected]] On Behalf Of Joseph Heled Sent: 02 December 2020 02:31 To: Timothy Chow <[email protected]> Cc: [email protected] Subject: Re: Snowie and GNUBG https://katgammon.com/blog-post/which-backgammon-program-is-strongest/ On Wed, 2 Dec 2020 at 14:12, Joseph Heled <[email protected]<mailto:[email protected]>> wrote: Thank you, Timothy! So not enough data to know one way or the other, then. -Joseph On Sun, 29 Nov 2020 at 07:48, Timothy Y. Chow <[email protected]<mailto:[email protected]>> wrote: Joseph Heled wrote: > My recollection is that GNUBG was much stronger than Snowie even 15 > years ago. And 2020 GNUBG is stronger than 2005 GNUBG. > > Am I missing something? According to Frank Berger, Torsten Schoop conducted a "Big Bot Shootout" back in 2006. http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=7784 GNU came out slightly ahead but arguably not enough to be statistically significant. Backgammon Galore records some earlier results, also inconclusive. https://www.bkgm.com/rgb/rgb.cgi?view+1040 https://www.bkgm.com/rgb/rgb.cgi?view+1086 Michael Depreli also did a bot comparison back in 2012, but this wasn't a head-to-head duel; rather, it was an error-rate comparison. I am not sure which version of GNU was used. http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=114355 Tim
