Komodo strikes first at the TCEC Superfinal

[Modern Times]

Adjudication is always a small risk, but usually a necessary one when running a decent number of longer time control games.

Personally I think 6.50 is not high enough. I normally use 8.50 in my matches and for 8 consecutive moves. If I'm watching, I like to see more of the game.

I think that the problem is that a high eval score could be because the engine is seeing too deep but they have not reached a point in which a normal human would resign. Even if it is true, the spectator would miss the chance to see the execution.

I always said that what is necessary is both a high score and a clearly positive material value. For instance, +6.5 AND at least two pawns up or equivalent. Here, there would not have been any adjudication.

Miguel

An adjudication setting of 8.5 almost always means the game goes on longer than with a setting of 6.5, so the spectator does in fact see more of the execution. A setting of 6.5 chops the game off too early for me.

[M ANSARI]

I think most can agree this is a bug as Bc2 seems quite trivial for either engine considering the time control and the hardware. I would like to know if this bug happens more often and is not detected or if it happens only on very rare occasions.

[Hai]

It's simple Stockfish hadn't won game 22 because he evaluated Kh3 as a draw by 3-Fold Rep but it was only a 2-Fold Rep.
But Stockfish evaluates 2-Fold Rep = 3-Fold Rep.
And that's why Stockfish played Kf3??.

As a result you need to fix 2-Fold Rep and evaluate this correctly.

No, the problem is simply that it missed Bc2 and played Kg4 instead (due to a bug).

After Kg4 Rg2+ most engines, including SF and Komodo, evaluate Kh3 as a draw because it allows Rb2 resulting in a first repetition. This has always tested better in terms of Elo, so that's why they do that.

When Stockfish evaluates a FIRST repetition in a +30.00/+250.00 POSITION as +0.00 then he is a NOOB.
And the developers who missed that needs a lot of ass spankings.

And what is the secret better elo gain?
I know it's + 0.00000000000001 ELO.

[rabbits23]

Thanks for the analysis site Sean. I thoroughly enjoyed this guy's
explanation of the game.
Allan

[Norm Pollock]

What I like about TCEC, specifically the 100 match LTC final, is that it pushes the engines to their max. The very rare bugs you would never see otherwise, become palpable.

Entertainment, drama, not so enticing.

From my POV, it looks like SF developers have become aware of several positional bugs, and time wasting and time forfeit issues. Komodo developers don't seem to have become more aware of any faults, although I believe they have noticed that Komodo does not use enough of its available time.

I would appreciate it if one of the experts following the match would make a listing of the different "bug" issues that have been exposed.

[Damir]

Is TCEC chat working ? I can not log in, neither can I read what others are writting...

[Eelco de Groot]

In my case there was a popup screen, saying my e-mail needed to be verified. Press button to receive a verification code, then enter that code.

[Eelco de Groot]

I think most can agree this is a bug as Bc2 seems quite trivial for either engine considering the time control and the hardware. I would like to know if this bug happens more often and is not detected or if it happens only on very rare occasions.

Maybe someone can confirm this, but Lucas proposed a bugfix 7 days ago for a TT issue that looks like it could have caused the TCEC missed winning move. Probably there was a discussion on Fishcooking but I have not read the threads about this yet. But on Rybkaforum they said this was fixed with the new patch. Too late for a possibly vital point, that could have given Stockfish some confidence back that it can also win games from Komodo, not only lose from Komodo. We can only hope that in turn Komodo becomes overconfident and starts making silly mistakes near the end.

[Damir]

same happened to me, although no verification code received... how do we get one ?

[syzygy]

But on Rybkaforum they said this was fixed with the new patch.

The problem is that everything, including doing nothing at all, seems to "fix" a problem that is not easily reproducible.

And even if it were reproducible and went away after a change, it is likely that the problem is only hidden / moved to different positions.