23 Aug
2006
23 Aug
'06
4:51 p.m.
At 04:33 PM 8/23/2006, David Wilson wrote:
The problem arose in trying to determine whether a two-computer chess program could be played where the computers transmitted one 8-bit byte per move.
A note about a related problem. In the generation of optimal-play endgame tablebases, they assumed that no forced win would take more than 255 moves. Well, a few months ago they found one that took (I think) 312 moves.