give up / propose draw

[flok]

Hi,

Is it possible for a chess engine to propose a draw and/or give up when using the UCI protocol?
If so: how?

[Ferdy]

Hi,

Is it possible for a chess engine to propose a draw and/or give up when using the UCI protocol?
If so: how?

Propose a draw is not, but give up is possible, just send an illegal move like
bestmove a1e8 and hope the gui will flag you of illegal move (often the case) and get a loss.

[flok]

Hmmm, a bit clumsy.

[Ferdy]

Hmmm, a bit clumsy.

Another option is to send the worst move and hope the opponent will mate you early or when gui adjudication is on, the game will be terminated early.

[Ferdy]

Hi,

Is it possible for a chess engine to propose a draw and/or give up when using the UCI protocol?
If so: how?

Still another method of giving up is by time forfeit. After receiving go ... command, search without limit, don't return bestmove, until you receive a stop command, by then you can send your bestmove.
In this case the gui had sent stop command because you have already exceeded the time allocation.

[flok]

I would like to propose an extension to the UCI protocol:
- draw-proposal by engine
- engine give-up.

My idea is that the "bestmove" stays mostly the same but gets an extra keyword after the move being "draw-req" or "giveup-req".
The GUI can then decide to go with the request and indeed propose the draw or give up the game to the human or ignore this and just pick the move that should also be included in the bestmove-output.

To make sure it doesn't give incompatibilities these enhancements could be default-off and be switched on using a "setoption"-command (in response to two new options "UCI_AllowDraw" and "UCI_AllowGiveUp").

[hgm]

Well, it is very easy to propose improvementes for UCI, and most of the UCI dialects (e.g. USI, UCCI) have indeed added such features. But, unlike WB protocol, which evolves with the needs of the community, UCI is essentially a dead protocol, which stands immutable in the face of time with all its defects...

[JoshPettus]

As such I enjoy this bit of irony on the chessprograming wiki

It has, by-in-large, replaced the older Chess Engine Communication Protocol (WinBoard/XBoard).

There really isn't a single UCI project that people can contribute and maintain, is there?

At least CECP has an opensource reference implementation and base in xboard.

Also upto date documentation. (Just as with so much else, huge thanks to Harm there.)

[flok]

Well, it is very easy to propose improvementes for UCI, and most of the UCI dialects (e.g. USI, UCCI) have indeed added such features. But, unlike WB protocol, which evolves with the needs of the community, UCI is essentially a dead protocol, which stands immutable in the face of time with all its defects...

I can't find the USI and UCCI specs?

[flok]

As such I enjoy this bit of irony on the chessprograming wiki

It has, by-in-large, replaced the older Chess Engine Communication Protocol (WinBoard/XBoard).

There really isn't a single UCI project that people can contribute and maintain, is there?

At least CECP has an opensource reference implementation and base in xboard.

Also upto date documentation. (Just as with so much else, huge thanks to Harm there.)

I'm more than willing to do so.
Question is though: are others willing to send the changes they've implemented to me?