Author: Dann Corbit
Date: 14:48:45 02/23/04
Go up one level in this thread
On February 22, 2004 at 17:25:55, Ratko V Tomic wrote: >While the draw is not 100.000...% guaranteed, extrapolating from the endgame >table-bases and theory -- any endgame with symmetrical material & position and >far apart, mutually isolated forces is a draw -- overwhealming odds are it is a >draw. Another extrapolation indicating a draw for the perfect play is that the >percentage of draws increases as the players' strengths increase. The >diminishing returns indicate the same (the rise in draw percentage) for the >computers as the processing power increases. It would be interesting to see if it is due to evenly matched skill set or simply with increasing strength. What I mean by that is to: 1. Match together programs that are very strong and evenly matched. 2. Match together programs that are medium in strength and evenly matched. 3. Match together programs that are not very strong and evenly matched. Then, looking at the result, do we see a higher and higher proportion of draws as a function of Elo? If the draw ratio stays the same, then it would seem that draws are simply the most probable output between evenly matched opponents. But if the draws increase in frequency then we can extrapolate that they may become more probable with an increase in strength.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.