Page 2 of 30

Re: Cursed win at TCEC

Posted: Tue Nov 15, 2016 10:52 pm
by Uri Blass
I read in the rules that
Cutechess will also adjudicate 5-men or less tablebase endgame positions automatically.

I did not read in the rules that Cutechess is going to adjudicate them based on distance to mate and not based on distance to conversion.

I think that it is TCEC's fault for not having a clear rules.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 12:27 am
by MikeB
Dan Cooper wrote:
MikeB wrote:
Dan Cooper wrote:
zenpawn wrote:
Dan Cooper wrote:It should be changed to a draw. Not much discussion needed.
Indeed. Is the TD aware of the situation?
Martin is aware but not sure Anton is. Martin just said the result would stand because rules are rules.

I don't think they intended for this to happen though.
If true and they stick by their "rules are rules" and "our head is in the sand" mantra, they can kiss their little tournament , as it stands today , "good bye". Idiocy at its finest - if true. Hopefully it's not true and the results are "fixed", make that "corrected". 😊
This is a bit of an overreaction.
Haha - was waiting to see who would pick up on that - just kidding - I really don't care one way or the other.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 1:51 am
by Houdini
whereagles wrote:Have a look:

http://tcec.chessdom.com/archive.php?se=9&sf&ga=17

Engines showing 0.00 due to 50-move rule, but position was auto-adjudicated as an M72 TB win :D

Discuss :)
Not much discussion possible.
Both engines know that it's a draw (0.00) and play accordingly.
Suddenly the GUI decides otherwise and is clearly not following the rules of chess as implemented in the engines.
It's kinda ridiculous, but not very important.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 2:07 am
by JJJ
I also think it's should be a draw, not a win.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 4:42 am
by gladius
Houdini wrote:
whereagles wrote:Have a look:

http://tcec.chessdom.com/archive.php?se=9&sf&ga=17

Engines showing 0.00 due to 50-move rule, but position was auto-adjudicated as an M72 TB win :D

Discuss :)
Not much discussion possible.
Both engines know that it's a draw (0.00) and play accordingly.
Suddenly the GUI decides otherwise and is clearly not following the rules of chess as implemented in the engines.
It's kinda ridiculous, but not very important.
Agreed, it should be a draw. As I posted in the TCEC chat, the adjudication should match the result if the engines had played the position out. In this game, it was a 50 move draw.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 5:33 am
by S.Taylor
whereagles wrote:Have a look:

http://tcec.chessdom.com/archive.php?se=9&sf&ga=17

Engines showing 0.00 due to 50-move rule, but position was auto-adjudicated as an M72 TB win :D

Discuss :)
I used to argue very bitterly against this, but EVERYONE ALWAYS said back to me, that if the tb says it is a draw or a win, then that is what TCEC will listen to.
And regarding the 50 move rule, it does not apply in a position where it is proven that it can be a win after a larger amount of moves, and if TB says it is, then it is.
I didn't use to agree but i was told every time i complained, that this is how it is. This is the TCEC rule.
So why does everyone just wake up now? It has happened many times and everyone agreed to it.
No one was moved by the fact that the engines played accordingly to a different reality, as they were wrong.
And Cato too, said that "if we don't listen to the TB's then why would we need them? Now we have them, we should only say thank you! Don't start saying they are wrong!"

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 5:52 am
by Evert
S.Taylor wrote: And regarding the 50 move rule, it does not apply in a position where it is proven that it can be a win after a larger amount of moves, and if TB says it is, then it is.
Check section 9.3 of the FIDE rules. There are no exceptions in the current rules.

What needs to be clear is what tablebases will be used to adjudicate a position. Clearly the newer DTZ50 tables are superior to DTM tables.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 7:43 am
by Jouni
Note, that in TCEC Rapid there was also same ending round 20 Arasan - Protector. No discussion then. But engines were not reporting 0,00 - does it change situation?

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 8:01 am
by Evert
Jouni wrote:Note, that in TCEC Rapid there was also same ending round 20 Arasan - Protector. No discussion then. But engines were not reporting 0,00 - does it change situation?
I don't see why it would.
Of course if the engines don't recognise that the position is a draw (either by using a tablebase of through knowledge/search), then adjudicating the game may distort the result. That can be said about GUI-side tablebase adjudications either way, of course.

If done correctly:
  • Both engines use tablebases: there are no swindles possible, so the result of the game will be what the tablebase indicates. Safe everyone's time and adjudicate the game.
  • One engine uses tablebases, the other does not: if the engine with tablebases is in a won position, adjudicate the game. The other engine can only make things worse. Otherwise let the game play out as normal, the engine without tablebases may yet mess up an lose the game.
  • Neither engine uses tablebases: don't adjudicate the game.
Of course the referee and the engines should use compatible tablebases, otherwise you end up with situations like this one.

Re: Cursed win at TCEC

Posted: Wed Nov 16, 2016 8:04 am
by syzygy
Jouni wrote:Note, that in TCEC Rapid there was also same ending round 20 Arasan - Protector. No discussion then. But engines were not reporting 0,00 - does it change situation?
Arasan reported 0.00. And Arasan had committed to this path to a draw 15 moves earlier or so. So Arasan made a bet which was fully supported by the rules of chess but still lost it.