UCI engine authors: Please add support for 'searchmoves'

Discussion of chess software programming and technical issues.

Moderator: Ras

Gian-Carlo Pascutto
Posts: 1260
Joined: Sat Dec 13, 2008 7:00 pm

Re: UCI engine authors: Please add support for 'searchmoves'

Post by Gian-Carlo Pascutto »

hgm wrote:You shouldn't feel inhipbited to develop new protocol. If engine and GUI authors would wait for each other, progress would stall forever in a deadly embrace...
I know, but it's best to do these kind of things in tight cooperation so there is feedback from both sides.

And I don't want to encourage the Aquarium programmers to add new features. The current ones aren't even halfway debugged yet :)
dadij
Posts: 40
Joined: Wed Jul 05, 2006 8:25 pm

Re: UCI engine authors: Please add support for 'searchmoves'

Post by dadij »

Gian-Carlo Pascutto wrote: I'm not going to implement this as there is no standard, no specification and no GUI for it. I'm just saying I wish UCI had forseen this.
Fair enough, but I think I've got the picture now. Here is the discussion in a nutshell:

1) Me: Please implement 'searchmoves'

2) Engine authors: We have a better idea which makes 'searchmoves' redundant.

3) Me: Will you then implement it instead of 'searchmoves'

4) Engine authors: No, there is no standard, no specification and no GUI for it.

So, we have a tough situation here.

On one hand 'go window' won't be implemented because there is no standard, no specification and no GUI for it. On the other hand 'searchmoves' won't be implemented, even if it is a part of the UCI standard, it has a specification, there are GUIs supporting it and there engines supporting it. The reason is that 'go window' is a better idea. Continue reading from the start of this paragraph.

Engine authors are clearly a tough crowd, but I wonder if they often talk themselves into a stalemate situation like this.
Because searchmoves is equivalent to feeding the child positions to the engine
OK, here you are encouraging adding this feature to the GUI. In another post you say you don't want to encourage adding new features to Aquarium.

I don't think this discussion is leading anywhere.
Gian-Carlo Pascutto
Posts: 1260
Joined: Sat Dec 13, 2008 7:00 pm

Re: UCI engine authors: Please add support for 'searchmoves'

Post by Gian-Carlo Pascutto »

dadij wrote: Engine authors are clearly a tough crowd, but I wonder if they often talk themselves into a stalemate situation like this.
It's not a stalemate: the end result is that we don't have to do anything, so we won :)

(Don't worry, I already re-added searchmoves to Deep Sjeng a while ago)
adieguez

Re: UCI engine authors: Please add support for 'searchmoves'

Post by adieguez »

no problem, multipv and searchmoves. In the next update Sir.
dadij
Posts: 40
Joined: Wed Jul 05, 2006 8:25 pm

Re: UCI engine authors: Please add support for 'searchmoves'

Post by dadij »

Gian-Carlo Pascutto wrote: It's not a stalemate: the end result is that we don't have to do anything, so we won :)
:)
(Don't worry, I already re-added searchmoves to Deep Sjeng a while ago)
In that case we can both celebrate victory :) I look forward to the new Deep Sjeng version.
dadij
Posts: 40
Joined: Wed Jul 05, 2006 8:25 pm

Re: UCI engine authors: Please add support for 'searchmoves'

Post by dadij »

adieguez wrote:no problem, multipv and searchmoves. In the next update Sir.
Thanks. Things are starting to move in the right direction! I already look forward to the upcoming versions of Amyan, Deep Sjeng and Pupsi2. Only 97 more to go! (If H.G.Muller's number is correct) :)