Can the world correspondence champion beat Stockfish

[CornfedForever]

Being a piece up against the naked king like a Knight or Bishop but insufficient to checkmate your opponent should be awarded 3/4 point
Conversely, the weaker side should be awarded only ¼ point in these cases

I honestly do not think any of the suggestions I've seen do anything. Even the suggestion above....in such a low variance game as top level "correspondence" chess, why should either side get 3/4 for the above...seems like Black should get more than White if he achieved this.

Regardless...stick a fork in it, erect a tombstone and...if people want to play this way, call it something else...and giving actual chess titles (GM, IM...etc, seems a bit silly when you think of it.

[Chessqueen]

Being a piece up against the naked king like a Knight or Bishop but insufficient to checkmate your opponent should be awarded 3/4 point
Conversely, the weaker side should be awarded only ¼ point in these cases

I honestly do not think any of the suggestions I've seen do anything. Even the suggestion above....in such a low variance game as top level "correspondence" chess, why should either side get 3/4 for the above...seems like Black should get more than White if he achieved this.

Regardless giving actual chess titles (GM, IM...etc, seems a bit silly when you think of it.

You are correct in the sense that 80% of their moves is probably suggested by top chess engines, therefore, the true correspondence Chess players were those that played back before 1980 when chess engines were NOT even IM rated like these 2 players

[Uri Blass]

Can the world correspondence champion beat Stockfish at 2 days per move and allowing stockfish 24 hours per move, and the correspondence world champion to use any chess engine besides stockfish

I believe not even if you allow stockfish only 1 hour per move.
It may be interesting to do a competition of the highest level that people can beat stockfish16.

People have 2 years to produce a win when the opponent is fixed and deterministic at x nodes per moves single core and 4096 mbytes hash tables and no book and no tablebases.

The target is simply to have the highest x.
I doubt if we are going to see x>10^9

[Uri Blass]

Without computer help. no.

The question is with computer help and I strongly believe the reply is also no.

[Chessqueen]

Without computer help. no.

The question is with computer help and I strongly believe the reply is also no.

Therefore. you are suggesting that Stockfish can solve chess at 24 hours per move and all chess positions will be solved

If that is the case give Stockfish 24 hours per move and see if it can win with this position
Despite the clear superiority of engines, there ARE positions which chess engines don't (and possibly can't) understand that are quite comprehensible for human players. Typically these positions showcase the human ability to think creatively and formulate plans and understand long-term factors in the position.

An example is this position by Nobel Prize winner Roger Penrose, brother to GM Jonathan Penrose. Humans will easily see that White can do nothing, and Black will have no plausible attempts to win, but chess engines will assert that Black is easily winning due to the numerous extra black pieces on the board.

[pgn][Event "Computer chess game"]
[Date "2023.07.09"]
[Round "?"]
[White "Human"]
[Black "Stockfish-windows-x86-64-avx2"]
[Result "*"]
[BlackElo "3550"]
[Time "06:02:10"]
[WhiteElo "?"]
[TimeControl "0+30"]
[SetUp "1"]
[FEN "8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8 b - - 0 1"]
[Termination "unterminated"]
[PlyCount "0"][/pgn]

[Chessqueen]

Without computer help. no.

The question is with computer help and I strongly believe the reply is also no.

Any such match is unfair to the human since computers have opening books and endgame table bases they get to use as reference. If the human was extremely focused and it was only calculation vs calculation I think the human (Magnus Carlsen) would still have a chance. If we bring reference materials into the game but only the computer could use them, like how they normally set up human vs computer matches, the human would lose. If they both had access to reference materials, but the human couldn't use an engine, I'll say Magnus barely wins.

[Robert Pope]

Without computer help. no.

The question is with computer help and I strongly believe the reply is also no.

Therefore. you are suggesting that Stockfish can solve chess at 24 hours per move and all chess positions will be solved

If that is the case give Stockfish 24 hours per move and see if it can win with this position
Despite the clear superiority of engines, there ARE positions which chess engines don't (and possibly can't) understand that are quite comprehensible for human players. Typically these positions showcase the human ability to think creatively and formulate plans and understand long-term factors in the position.

An example is this position by Nobel Prize winner Roger Penrose, brother to GM Jonathan Penrose. Humans will easily see that White can do nothing, and Black will have no plausible attempts to win, but chess engines will assert that Black is easily winning due to the numerous extra black pieces on the board.

[pgn][Event "Computer chess game"]
[Date "2023.07.09"]
[Round "?"]
[White "Human"]
[Black "Stockfish-windows-x86-64-avx2"]
[Result "*"]
[BlackElo "3550"]
[Time "06:02:10"]
[WhiteElo "?"]
[TimeControl "0+30"]
[SetUp "1"]
[FEN "8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8 b - - 0 1"]
[Termination "unterminated"]
[PlyCount "0"][/pgn]

You do realize that being likely able to draw at will isn't the same thing as solving chess, right? And even solving the opening position isn't the same thing as all positions.

[Uri Blass]

Without computer help. no.

The question is with computer help and I strongly believe the reply is also no.

Therefore. you are suggesting that Stockfish can solve chess at 24 hours per move and all chess positions will be solved

If that is the case give Stockfish 24 hours per move and see if it can win with this position
Despite the clear superiority of engines, there ARE positions which chess engines don't (and possibly can't) understand that are quite comprehensible for human players. Typically these positions showcase the human ability to think creatively and formulate plans and understand long-term factors in the position.

An example is this position by Nobel Prize winner Roger Penrose, brother to GM Jonathan Penrose. Humans will easily see that White can do nothing, and Black will have no plausible attempts to win, but chess engines will assert that Black is easily winning due to the numerous extra black pieces on the board.

[pgn][Event "Computer chess game"]
[Date "2023.07.09"]
[Round "?"]
[White "Human"]
[Black "Stockfish-windows-x86-64-avx2"]
[Result "*"]
[BlackElo "3550"]
[Time "06:02:10"]
[WhiteElo "?"]
[TimeControl "0+30"]
[SetUp "1"]
[FEN "8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8 b - - 0 1"]
[Termination "unterminated"]
[PlyCount "0"][/pgn]

being able to draw from the opening position is not equivalent to being able to play the right move in every position so the position you post is not relevant.

The point is not that there is no position that stockfish can make a losing blunder but that humans cannot get a position that stockfish make a losing blunder in a practical game.

[Peter Berger]

I believe not even if you allow stockfish only 1 hour per move.
It may be interesting to do a competition of the highest level that people can beat stockfish16.

People have 2 years to produce a win when the opponent is fixed and deterministic at x nodes per moves single core and 4096 mbytes hash tables and no book and no tablebases.

Your first challenge feels pretty doable though tough - the second one a little over the top.

Spontaneously I felt tempted to take the first one - I even did a few moves. Then I stopped to think a little. Even if someone can manage to do this (which I think is still possible), what is the point? It will take a few months - by the time the game is decided there will be further advancements in standard computer hardware and software ( like Stockfish 17).

I am under the impression that the computers still show minor weaknesses in the opening right now, if left on their own - but this clearly only is a very temporary thing, if even true.

I am not so convinced when it is about practically solving chess. I wouldn't be surprised if some engine in 10 years can clearly dominate current ones - it is just, that human opinions on the games will have become completely irrelevant.

[lkaufman]

Chess is a draw. So at the top level of ICCF play, all games are drawn, no matter how much effort we make to force victories.

Also, SF at 24 hours/move will surely avoid loss. So, I forecast 100% draws in this hypothetical match.

-Steve-

According to Dr. Jon Edwards who retired from Princeton University and is the current or 32nd Correspondence Chess Champion, he stated that human guidance is still the key to success in correspondence chess is claimed in Jon’s own words: “Computers today are tactical monsters and positional masters, but they fall short in positions that involve long-term planning within fixed structures...”

"According to Dr. Jon Edwards who...." won the title when someone blundered via a recording error....if my memory and look at the games doesn't fail me. The person who made the error could have as easily made it against one of the other players...in which can you would be referencing Dr. Jon Edwards, USCF "A" player who failed in his bid to win the ICCF Championship.

I don't mean for that to sound as bad as it does - heck, I have long owned his 'THE CHESS ANALYST' book, and loved it. But that was 25 yrs ago, when there was still life in the game.
Even little ol' me even has a 1.5/2 OTB score against a current ICCF GM.

All I'm really saying here is that Stephen Ham is right. "100%" draw (barring, ahem...transcription errors...). "Correspondence" is dead. What exists is...well, just something else.

Here are some recommendations to improve correspondence chess ==> https://en.chessbase.com/post/correspon ... aw-problem
Therefore, I specifically suggest enhancing the score system to include a ¾ point
for a performance with an added value as compared to a regular draw:

Being a piece up against the naked king like a Knight or Bishop but insufficient to checkmate your opponent should be awarded 3/4 point
Conversely, the weaker side should be awarded only ¼ point in these cases

And here is the best correspondence chess game when chess engines were just pawn pushers

Changing the scoring of minor vs. king and/or stalemate will help a little to reduce draws, but the one change that would make a really big difference without changing basic rules other than draw rules would be to forbid triple repetitions, rather than to call them draws. The game of GO added the KO rule (forbidding repetitions) hundreds or even thousands of years ago, it's time for chess to do the same. I would estimate that this would at least double the frequency of decisive games in high level chess, maybe much more in Correspondence chess where that percentage is miniscule now among the best players. It might not be enough to make Correspondence chess (or top engine matches on good hardware) viable with standard openings, but it has the best chance of any solution that might be acceptable to the chess community.