If it is clear that the engine does not to lose on time then it is logical for it to go for the longest possible draw and if the engine needs only 0.01 second to make a non losing move in tablebase position then even 1 second left on the clock is enough for it.Chessqueen wrote: ↑Thu Jun 29, 2023 10:46 amIt all depend on the time left for both engines, if the engine that can draw with 2 moves only has 1 second left on his clock it is logical to draw with 2 moves, but if it has more than 1 minute left and the opponent only has 1 second left it is logical to make the game as long as possible.Uri Blass wrote: ↑Thu Jun 29, 2023 10:24 amIt may be better from the engine's point of view to make the game as long as possible when it is a draw because the opponent may lose on time so drawing in 2 moves by this reason is worse than drawing in 50 moves.Chessqueen wrote: ↑Thu Jun 29, 2023 6:19 amMr. Dann, Most top GMs will select Rh3+ since it draws in 2 moves, therefore, it is NOT a matter of how many different moves can lead into a drawDann Corbit wrote: ↑Thu Jun 29, 2023 5:04 am The following moves draw:
Rf5
Rh3
Ng7
Nc5
Kf6
Kf5
Kf7
as proven by the nalimov tablebase files at:
https://www.k4it.de/?topic=egtb&lang=en
Therefore, it is completely irrelevant which of the above drawing moves is played.
All lead (provably) to a draw.![]()
NOTE: The reason why I believe it is illogical to prolong the game to 50 moves is because I do NOT use tablebases and never ever had, and for some engines to select the longest draw is very illogical![]()
Why do some top engines select Kf5 ?
Moderator: Ras
-
Uri Blass
- Posts: 11089
- Joined: Thu Mar 09, 2006 12:37 am
- Location: Tel-Aviv Israel