Cursed win at TCEC

[Evert]

Concerning the real TCEC game, this is quite similar to FIDE rules concerning the situation when a player has been flagged, but he has just delivered his opponent a mate. In this case, although the flag of the player is already down, thus losing on time, he still wins the game, fully according to FIDE rules, as there is mate on the board, which overrides flagging.

Correct: I cannot claim a draw by 50 move rule if I have just been mated. Mate tumps the 50 move rule. Conversely, you cannot claim a win by mate-in-two if I can claim draw by 50 move rule first.

Very similarly, based on TCEC special draw rule, the game was declared drawn, but, when Cutechess adjudicated the tablebase position, it adjudicated a win, as the position is simply won. An existing win on the board, which is the case in current game, actually an existing mate, certainly overrides the special TCEC draw rule.

This, however, completely misses the point of what happened here.
The game was not declared a draw, and there was no mate on the board. The game was halted in a position that is drawn under the 50 move rule, but adjudicated as a win because of the existence of a mating line that ignores the 50 move rule.

Now, it could be argued that the referee has no business adjudicating a game as drawn under the 50 move rule: it is up to the player to claim the draw in that case. That is a separate discussion. Of course the UCI protocol lacks a way for engines to do this anyway, which is why the referee makes the claim on behalf of the player.

[hgm]

You could also adjudicate this position as a win for white:
[d]7q/8/5k2/8/8/4K3/8/8 w - - 0 1
After all, it is a forced win in 3 (1. Kd4! Kg7 2.Ke5! Kf6 3.Kxf6 Qxf6 1-0). It would be a bit strange to do that in a game of orthodox Chess, however. Even if you announced that 3-men positions would be adjudicated.

[Dirt]

It is not an opinion but a fact. There have not been any 1-1 openings so far.

Rounds 17 and 18 were the same opening and each was won by white. Not unexpected, of course.

[Guenther]

It is not an opinion but a fact. There have not been any 1-1 openings so far.

Rounds 17 and 18 were the same opening and each was won by white. Not unexpected, of course.

uhhmm... game 17 is the cursed win which was wrongly adjusticated

[MikeB]

Yea it pretty funny all over a 1/2 point between two chess engines and the only thing at stake is bragging rights. we need to move on. Oh. .. and do me a favor , next time try to get it right 😉

[Norm Pollock]

The fundamental unresolved issue has been around a long time, and it will be around long after tcec. What it comes down to is why should computer engines that see a mate in 51+ moves be penalized and forced to accept a draw. The 50 move rule was established solely for humans and it was established long before computer engines. Should it be changed to accommodate technological progress and if so, how?

As for this tournament and superfinal itself, no big deal, but the incident illustrates the bigger problem that is unresolved.

[hgm]

Actually something like the 50-move rule is essential in computer Chess, even more so than in human Chess. Humans will tire, and eventually agree a draw. Engines would continue for many millions of moves.

[syzygy]

It is not an opinion but a fact. There have not been any 1-1 openings so far.

Rounds 17 and 18 were the same opening and each was won by white. Not unexpected, of course.

Well.... I consider game 17 to be a draw

[duncan]

Actually something like the 50-move rule is essential in computer Chess, even more so than in human Chess. Humans will tire, and eventually agree a draw. Engines would continue for many millions of moves.

I agree something like the 50-move rule is needed in computer Chess

eg. 50 moves except where there is a tablebase win. this will cover the 1000 move win for 8 piece tablebase which we may have in 7 years, but will disallow pointless prolongation of the game where there is no tablebase win

[Evert]

eg. 50 moves except where there is a tablebase win. this will cover the 1000 move win for 8 piece tablebase which we may have in 7 years, but will disallow pointless prolongation of the game where there is no tablebase win

This is too messy and inconsistent.
There was an exception in FIDE chess, and it was removed because as time goes on there become more and more exceptions and the rule becomes unwieldy.

I think it's better (and easier) to just accept that under the rules of chess there are some positions that cannot be won under the 50-move rule, that could be won otherwise. Variants such as Makruk and Sittuyin have "counting" rules that are in some ways similar to the 50 move rule (but more punishing). These alter the nature of the game in end games, making positions that could be trivially won drawn instead. It's the nature of the game.

I suppose one could have an option to ignore or change the 50 move rule, if desired. That way people can decide to ignore it (or change it) if they want for their own games/tournaments. That's a different discussion though.