... each year there were fewer and fewer decisive games in the ICCF World Championship Finals and now there are none.
The thing is, this can have multiple causes.
The conclusion of "Therefore, we have reached perfect play, or close to it" (or some similar conclusion) isn't the only conclusion we can draw from the results.
Some other possible conclusions:
Possibility 1: Previous ICCF championships may have had greater disparities between competitors, in terms of hardware and software being used, and/or being available - thus leading to differences in playing strength among the competitors.
Possibility 2: ICCF competitors may have believed, in past years, that their human decisions were more valuable - thus they may have made more human choices in their games. But as the strength of engines took a large leap forward in recent years (due to AlphaZero's rise and the community acceptance of NNUE, for example), ICCF players may have learned to trust their hardware and software more, and to rely on their human input less and less - thus leading to a greater amount of draws, due to competitors all pitting the same hardware/software against each other with minimized human interference.
Possibility 3: ICCF players may have learned to embrace less volatile openings/defenses and have thus narrowed their repertoires to the safest options, to minimize their chances of losing ... resulting in a greater number of draws.
... We can consider more possibilities as well, if we just sit and think about it. And this is one of the necessary steps to avoid leaps in logic - we have to consider other possibilities and examine them accordingly ...
@12647
"I am aware of two examples from master praxis where an underpromotion to bishop was the only winning move." ++ So that makes 1 position in 10^6 games.
Now of how many examples from master praxis are you aware where a pawn promoted and both bishops were still on the board? ++ That is maybe also 1 position in 10^6 games.
Guessing - your modus operandi - is not good enough. This is a very reckless guess, without even shoddy reasoning behind it.That gives 1 position in 10^17 positions.
Based on a made up number, so worthless.
To estimate the number of relevant positions to weakly solve Chess (10^17)
Another made-up number where the reasoning never came close to making sense.
that is negligible indeed.
As well as guessing, you are indeed big on neglecting things.
"So it suffices to generate drawing strategies for the similar game to chess where underpromotions to bishop and rook are forbidden" ++ No they are allowed, but underpromotions to pieces not previously captured not.
Now try reading what I wrote. I referred to generating drawing strategies for a MODIFIED game where underpromotions to bishop and rook are forbidden. These strategies would do for chess as well due to my valid reasoning. You should be pleased with this!
"The number of legal chess positions without ANY promotion is around 4 x 10^37."
++ Correction: without any promotion to a piece not previously captured.
No. My statement is true AND your statement is also true. They are the SAME set of positions.
For brevity the article title does not mention that, but it is obvious. Yes, to everyone.
"strongly believed that chess is a draw" ++ To me it is proven.
That is because, as has been widely observed, you don't know what a proof is (a key piece of knowledge early in a mathematical education).
"It's safe to say chess still can't be rigorously solved."
++ No, on the contrary it is safe to say chess can be rigourously weakly solved.
The 17 ICCF WC Finalist and their servers are doing it now.
110 games out of 110 that redundantly link the initial position in average 39 moves to certain draws.
You are remarkably clueless about what a weak solution is: it involves generating a complete proof tree. In a proof tree not a single legal move for the opposition is left unanalysed. Ponder on that a while.
Essential reading about the definition of proof tree (it applies to all of mathematics, not just weak solutions of games), In the weak solution of games the key step is that the game theoretic value (W/D/L) of a position is the maximum of the game theoretic values of the positions reachable by a legal move from that position.
And of course, you have no understanding of what rigor is. It's a shame this will never be fixed.
"Today's top chess engines will probably lose matches to those of the future."
++ At short time control. In TCEC they have to impose 50 unbalanced openings to avoid all draws and they do hit the 7-men endgame table base.
Current chess engines beat those of only a short time in the past in some games. It is absurd to suggest this must be over. The ICCF is a competition between Stockfish 16 and itself running on high end hardware. Even using significantly inferior hardware is enough to lose games. There is no reason to believe Stockfish 16 is invincible, just that there is nothing to beat it right now. Stockfish 17 will come, I assure you!