Rotor 0.7a released

[Jan Brouwer]

Hi,

This release corrects a castling bug in version 0.7, and is otherwise identical.
I expect no change in playing strength, so testing results of 0.7 and 0.7a can probably be merged.
The new version can be found at https://sites.google.com/site/chessrotor/

Jan

[Graham Banks]

Hi,

This release corrects a castling bug in version 0.7, and is otherwise identical.
I expect no change in playing strength, so testing results of 0.7 and 0.7a can probably be merged.
The new version can be found at https://sites.google.com/site/chessrotor/

Jan

Thanks Jan. Not sure that I've seen the castling bug. Can you explain what happens?

[Jan Brouwer]

Hi,

This release corrects a castling bug in version 0.7, and is otherwise identical.
I expect no change in playing strength, so testing results of 0.7 and 0.7a can probably be merged.
The new version can be found at https://sites.google.com/site/chessrotor/

Jan

Thanks Jan. Not sure that I've seen the castling bug. Can you explain what happens?

Sure. While refactoring some code I thought that I had discovered a bug (which wasn't the case), and I "corrected" it. As a consequence, long castling was no longer considered legal if B1/B8 was attacked. If the opponent would castle under those conditions, rotor would ignore the move, and consequently possibly make an illegal move itself.

I checked the games played by CCRL, and did not find any problem. In theory, rotor could have played an inferior move because of this, but in practice I think that in most games castling is no longer possible once out of the opening books. So I don't expect a noticeable change in playing strength because of this.

Jan

[Graham Banks]

Hi,

This release corrects a castling bug in version 0.7, and is otherwise identical.
I expect no change in playing strength, so testing results of 0.7 and 0.7a can probably be merged.
The new version can be found at https://sites.google.com/site/chessrotor/

Jan

Thanks Jan. Not sure that I've seen the castling bug. Can you explain what happens?

Sure. While refactoring some code I thought that I had discovered a bug (which wasn't the case), and I "corrected" it. As a consequence, long castling was no longer considered legal if B1/B8 was attacked. If the opponent would castle under those conditions, rotor would ignore the move, and consequently possibly make an illegal move itself.

I checked the games played by CCRL, and did not find any problem. In theory, rotor could have played an inferior move because of this, but in practice I think that in most games castling is no longer possible once out of the opening books. So I don't expect a noticeable change in playing strength because of this.

Jan

Thanks for the explanation.