138 moves to finish only one stupid, boring, computer chess game.
Skippers most boring records
Moderator: Ras
-
- Posts: 739
- Joined: Thu Mar 09, 2006 2:56 pm
Re: Skippers most boring records
I don't understand how 3. Be3 can be attractive for a chess engine, even one of the stranger ones. Can you explain?
-
- Posts: 7251
- Joined: Mon May 27, 2013 10:31 am
Re: Skippers most boring records
I think it has a larger mobility on e3 assuming it might be that after some moves almost all pieces have been captured which blocked that bishop. Probably a bit too optimistic estimate.
-
- Posts: 243
- Joined: Tue Jan 31, 2023 4:34 pm
- Full name: Adam Kulju
Re: Skippers most boring records
I had this problem early on when I was just starting to add evaluation to the engine, if I were to guess it's one of two things:Peter Berger wrote: ↑Fri Feb 03, 2023 4:28 pm I don't understand how 3. Be3 can be attractive for a chess engine, even one of the stranger ones. Can you explain?
1. if Hank has a piece square table for the bishop, does it have larger bonuses for d3/e3 than f4/c4 or b5/g5? Mine does and if you don't compensate with other factors to take into account mobility/space/development the engine may just plunk the bishop there because according to the PST it's a better square for it than f4.
2. Does he have any sort of patch for central control? I used to calculate it by summing the number of pieces in the middle 16 squares. This led to the engine playing opening sequences like 1.d4 2.Nf3 3.Nc3 4.Be3 5.Qd3. I tried a few things to make it work such as giving penalties to pieces in front of pawns (which is a good start), but eventually scrapped it.
go and star https://github.com/Adam-Kulju/Patricia!
-
- Posts: 7251
- Joined: Mon May 27, 2013 10:31 am
Re: Skippers most boring records
Yes Skipper using piece square table only here. For bishop filled with too optimistic mobility values.
I don't think I can use piece square table to detect blocked (center) pawns for after some moves the blocking pieces might not be there who knows.
So to repair this I have to slow down the speed of the engine by adding code to eval.
I don't think I can use piece square table to detect blocked (center) pawns for after some moves the blocking pieces might not be there who knows.
So to repair this I have to slow down the speed of the engine by adding code to eval.
-
- Posts: 7251
- Joined: Mon May 27, 2013 10:31 am
Re: Skippers most boring records
Ok I might give penalties to pieces in front of center pawns where center pawns are at second row in piece square table
-
- Posts: 7251
- Joined: Mon May 27, 2013 10:31 am
Re: Skippers most boring records
Hmm there is a variation in the grunfeld where Be6 is a good move while pawn on e7 and other bishop on g7