how to improve endgame play?

Discussion of chess software programming and technical issues.

Moderator: Ras

Whiskers
Posts: 243
Joined: Tue Jan 31, 2023 4:34 pm
Full name: Adam Kulju

Re: how to improve endgame play?

Post by Whiskers »

turned out after i made all the changes and pitted them against the old version (from about a week ago), it was actually a regression :cry: Trying again...
User avatar
hgm
Posts: 28354
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: how to improve endgame play?

Post by hgm »

What always served me well for programs at this level was looking at the games. To see what they actually did wrong to turn a drawn situation into a loss, or aspect of the position they overlooked so that they believed a lost position was a draw.
Whiskers
Posts: 243
Joined: Tue Jan 31, 2023 4:34 pm
Full name: Adam Kulju

Re: how to improve endgame play?

Post by Whiskers »

hgm wrote: Mon Feb 06, 2023 4:57 pm What always served me well for programs at this level was looking at the games. To see what they actually did wrong to turn a drawn situation into a loss, or aspect of the position they overlooked so that they believed a lost position was a draw.
I traced it to razoring, my quiescent search is minimal and doesn't bother analyzing quiet checks or losing captures, which explains why it missed Bf7+ in this game (I wondered about this for a while because other than that I don't ever reduce checks except with razoring):