Cursed win at TCEC

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

zenpawn
Posts: 260
Joined: Sat Aug 06, 2016 6:31 pm
Location: United States

Re: fortress_draw_rule

Post by zenpawn » Sat Nov 19, 2016 1:33 pm

hgm wrote:Why would it be relevant what the Stockfish devs think, or what the engines see? Since when is the score earned by a Chess game determined by what the players think, rather than what happens on the board?
LOL. Never heard of a draw by agreement?

User avatar
hgm
Posts: 22187
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Contact:

Re: fortress_draw_rule

Post by hgm » Sat Nov 19, 2016 1:57 pm

Not after the game. No draw as offered during the game. It is also not possible for non-players to offer draw on behalf of a player. Not even during the game.
Last edited by hgm on Sat Nov 19, 2016 1:59 pm, edited 1 time in total.

syzygy
Posts: 4221
Joined: Tue Feb 28, 2012 10:56 pm

Re: fortress_draw_rule

Post by syzygy » Sat Nov 19, 2016 1:58 pm

The position reached at move 59 was a complete draw with and without the 50-move rule:
[D]4r3/K7/1P6/P2Q3b/7k/4b3/8/8 b - - 0 59
At move 64, black played Kg3 which is fine when using the 50-move rule, but which on my PC lets the eval rise to +0.88 when Syzygy50MoveRule is set to false. Probably still a draw, though.

At move 65, black "loses" the game with 65...Re5:
[D]K5Q1/4r3/1P6/P7/5bb1/6k1/8/8 b - - 12 65
With 65...Re6 the eval would have remained stuck at +0.88.

H5 played 65...Re5 only because it knew it would draw.

zenpawn
Posts: 260
Joined: Sat Aug 06, 2016 6:31 pm
Location: United States

Re: fortress_draw_rule

Post by zenpawn » Sat Nov 19, 2016 2:03 pm

hgm wrote:Not after the game. No draw as offered during the game. It is also not possible for non-players to offer draw on behalf of a player. Not even during the game.
In a sense, it was.

Think of it like an "instant replay" in tennis or football, where a call/adjudication is often corrected upon reviewing more detailed information.

User avatar
hgm
Posts: 22187
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Contact:

Re: fortress_draw_rule

Post by hgm » Sat Nov 19, 2016 2:07 pm

So that confirms what I stated earlier: the game was a dead draw under any rule, and Houdini did not care whether it would do moves that would keep the draw without 50-move rule as long as it would keep the draw with 50-move rule. So not paying any attention to it, it happened to do a move that would be losing without 50-move rule. Had the game continued, Stockfish could equally easily have played a move that would again 'blunder away' the win without 50-move rule, as it would not pay any attention to it either.

User avatar
hgm
Posts: 22187
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Contact:

Re: fortress_draw_rule

Post by hgm » Sat Nov 19, 2016 2:10 pm

zenpawn wrote:Think of it like an "instant replay" in tennis or football, where a call/adjudication is often corrected upon reviewing more detailed information.
I was not aware the players or the audiece determined the outcome of such a review. I thought they had video referees for that...

zenpawn
Posts: 260
Joined: Sat Aug 06, 2016 6:31 pm
Location: United States

Re: fortress_draw_rule

Post by zenpawn » Sat Nov 19, 2016 2:13 pm

Not really interested in joining the debate. I had legit questions. Sure, they implicitly revealed my opinion, but that's as much as I'd intended to say on the matter. Carry on. :)

Norm Pollock
Posts: 1011
Joined: Thu Mar 09, 2006 3:15 pm
Location: Long Island, NY, USA
Contact:

Re: fortress_draw_rule

Post by Norm Pollock » Sat Nov 19, 2016 3:27 pm

It is hard to believe that the 5-man gaviota tbs saw something that the 6-man syzygy tbs could not see. So it appears the gaviota tbs were not constrained by the 50-move rule while the syzygy 6-man tbs were.

So since the gui's tbs (gaviota-5man) are not constrained by the 50-move rule, the engine tbs should also not be constrained. But otoh, syzygy does not find the shortest mates and therefore could miss out on a mate that could be executed within 50 moves.

The other alternative is for the gui's tbs to also be constrained by the 50-move rule. That is the best solution going forward.

whereagles
Posts: 478
Joined: Thu Nov 13, 2014 11:03 am

Re: fortress_draw_rule

Post by whereagles » Sat Nov 19, 2016 3:31 pm

syzygy wrote:(...) The best we can do is run the position through SF with Syzygy50MoveRule set to false, which switches to the ICCF rule.

So I did that and the score gets stuck at +1.14. This suggests that since black cannot escape to a QvBB draw, it has to allow white slightly more space or something. But white still seems unable to make progress. (...)
This is interesting. It suggests the game would have been a draw even if engines were using Syzygy50MoveRule = FALSE.

At this moment the case for manually adjust to draw is very strong.

Next season should definitely either use adjudication TBs with 50 move rule, or allow explicitly for cursed wins and inform developers.

User avatar
Evert
Posts: 2895
Joined: Fri Jan 21, 2011 11:42 pm
Location: NL
Contact:

Re: fortress_draw_rule

Post by Evert » Sat Nov 19, 2016 8:41 pm

syzygy wrote: So I did that and the score gets stuck at +1.14. This suggests that since black cannot escape to a QvBB draw, it has to allow white slightly more space or something. But white still seems unable to make progress.
Interesting. It certainly looks that way.
However, I have seen a few cases where the computer is unable to see progress during analysis, but then if you let it play the position against itself, it still manages to win.

Post Reply