SEE test suite for Xiangqi (Chinese chess)

Discussion of chess software programming and technical issues.

Moderator: Ras

User avatar
phhnguyen
Posts: 1524
Joined: Wed Apr 21, 2010 4:58 am
Location: Australia
Full name: Nguyen Hong Pham

SEE test suite for Xiangqi (Chinese chess)

Post by phhnguyen »

I have just created a new test suite for testing Static Exchange Evaluation (SEE) for Xiangqi (Chinese chess).
- It is in EPD format: bm (bestmove) is actually the move to test and ce is the gaining score in centipawn
- Piece values: Rook: 1000, Cannon: 500, Horse: 400, Elephant: 200, Advisor: 200, Pawn: 200
- The test considers that a King can capture the other King (it is not obvious as standard chess)

https://github.com/nguyenpham/XiangqiTe ... e/main/see
https://banksiagui.com
The most features chess GUI, based on opensource Banksia - the chess tournament manager
User avatar
hgm
Posts: 28353
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: SEE test suite for Xiangqi (Chinese chess)

Post by hgm »

Is it wise to have different value for Cannon and Horse? Late in the game the Horse is worth more than a Cannon. In Chess we also count B=N in SEE, even though usually B > N (e.g. because of the Bishop pair). It seems more natural to have R=8, C=H=4, A=E=passed P=2, unpassed P=1.
User avatar
phhnguyen
Posts: 1524
Joined: Wed Apr 21, 2010 4:58 am
Location: Australia
Full name: Nguyen Hong Pham

Re: SEE test suite for Xiangqi (Chinese chess)

Post by phhnguyen »

hgm wrote: Fri Sep 09, 2022 7:18 am Is it wise to have different value for Cannon and Horse? Late in the game the Horse is worth more than a Cannon. In Chess we also count B=N in SEE, even though usually B > N (e.g. because of the Bishop pair). It seems more natural to have R=8, C=H=4, A=E=passed P=2, unpassed P=1.
You are totally right! Thanks!

Fixed as your suggestions. Added more tests for Pawns
https://banksiagui.com
The most features chess GUI, based on opensource Banksia - the chess tournament manager