169 is extraordinarily high...i should say. and id never-ever dispute this w/ opti. he's exceptionally well-rounded imo. and my guess is hes a scary sponge in anything he wansta learn abt. retention is...well lets just say he probably hazza a memory like a salmon.
The phrase is "like a goldfish".
Regarding the terminology 'weak solution' I'm tempted to say 'I told you so' ...
----------------------------
Its too bad really because surrounding valid concepts with jargon - blocks progress and provides opportunities for those wishing to so block progress.
The jargon is there to allow progress.
The term "weakly solved" is quite apposite. Several possible meanings of "solved" have been identified and they fall into an order according to their comprehensiveness, so "ultra weak", "weak" and "strong" are natural epithets.
This is not even in any sort of conflict with common English usage. The fact is the term "weakly solved" wouldn't be used in common English.
It would be used only by game theorists, people who have an interest in game theory and have taken the trouble to understand the definition and people who have heard the term but not taken such trouble.
The last block progress on a thread like this, but a thread like this is realistically not the place where progress in the theory is made.
The idea of computer projects seeking to solve chess by shortening the process with no further handling of positions that are obviously wins (corresponds to 'resign' in chess games) is a valid one.
Obviously not. You immediately decide these positions are Black wins?
(The last is an Ottó Bláthy win in 50 by White.)
Dubro's non-receptivity to this idea seems to serve notice that he's not interested in forward progress.
Or just the idea is duff, maybe.
-------------------------------
@Elroch - the so-called 'weak solution' of chess (poor terminology to refer to a valid idea)
does have a coupld of other bugs in it though besides its poorly-coined name.
One is that even if proper table based projects (and forward 'wide game tree searches') are both launched with that idea of skipping further analysis of lopsidedly-won positions ...
then that's still not a big enough slice of the task - even though that 'slice' leaves a small percentage of what the task was.
See above two points.
That slice is still just too big. Too many trillions of years of 'slice' left.
The second is that that slice also still has to avoid stalemate situations.
Further increasing the difficulty.
---------------------------------------------
One could conjecture that software engineers of chess-solving projects are very aware of the idea of skipping lopsided solving positions.
And very quick to reject it in connection with solving chess.
Perhaps a big reduction of John Tromp's number is already out there.
An estimate of how many positions are left after factoring out lopsided situations.
---------------------------
tygxc took a little run at that but in a hopeless way.
Like for example wanting to reject all further analysis after e4 e5 Ba6.
It 'looks' valid. But the plus of a bishop isn't always enough to win.
Even with 'ceteris paribus' factored in.
------------------------------
I like this next idea:
In all the solved tablebase positions (in other words all 7-piece or fewer legal chess positions) that have been found to be wins for white or for black -
Each won position has a further algorithm run on it -
where adding more pieces to the winning side in all ways that do not interfere with that side's win - is considered - but with no further evaluation - they are simply counted and added and a number of such is determined.
You refer to "that side's win" (singular).
Here is Syzygy playing twice from a winning position.
Notice that the wins are not identical.
The second is probably far short of the longest Syzygy win, but the average product of the number of Syzygy optimal moves for a White and Black move pair seems to be roughly 90 so it would be a reasonable guess that Syzygy has over 10^120 winning lines for that position.
What would be your practical algorithm for determining the number of ways of adding White pieces in such a way that they interfere with none of those lines? (A reasonable guess at the number would be 0, but a guess is not an acceptable algorithm.)
A huge amount of work could then be saved.
How? What use would your number be in cutting down the work in tablebase generation?
As to adding more pieces to the lost side - that's harder to do without interfering with the win but could also be done.
The point is - that when then plunging into 8-piece tablebase projects - the number of positions to be considered is already much reduced.
And also 9 and 10 and so on - without even reaching those projects yet.
Guess: they've already done that - and the John Tromp number reduced - is out there somewhere.
--------------------
And no - I haven't asked the AIs about that yet.
Just FYI your "I like this idea" is basically done in real time by most of the best chess engines for years.
@playerafar's idea is inchoate and would never have been incorporated in any chess engine ancient or modern, especially since they're not designed for tablebase generation.
It is called tablebase probing.
And it's not called tablebase probing, whatever Grok the world's smartest AI might tell you.
It is no wounder this thread is filled with nonsense. As these people have no clue about computer chess.
A high percentage by yourself of late.
And I never laught so hard in my life. As hearing their nonsense that chess could be weakly solved exactly like checkers.
And Grok the world's smartest AI sold you a heap of crap about what a weak solution is, so perhaps you should work out what you're laughing about before you start laughing.
I guess this was their logic.
Checkers is played on a 8x8 board.
Chess is played on a 8x8 board.
And no more thought was needed to justify their nonsense.
... I'm one of those people who had Covid at least 5 or 6 times and who refused the vaccines. There aren't all that many of us ...
I wonder why.
Optimissed didn't understand Martin's reply there.
It's worth mentioning one of the key nuances in the project to solve checkers - which involved 50 machines running in parallel for months, so efficiency was a big issue - is that rather than simply using what the engine chose as the best move as the candidate for the proponent of a strategy, they somehow estimated the amount of computation that would be necessary for each of the candidate moves, and used one that appeared both optimal and to have relatively small computational demands. I don't know how they did this, but this is mentioned in the paper I linked a while back.
Clearly the same concept would be very useful for solving chess if it wasn't so far from possible for the forseeable future.
And we are still waiting for you to REFUTE any of this from Grok, or myself.
But you can't......You no nothing but bad time stories. And not what is needed. Knowledge of chess, computer chess, and game theory.
You don't understand enough to see when Grok gets it wrong, and you (inappropriately) worship its infallibility so are blind to intelligent input from people who know much more than you.
Fairytale time is OVER!
I have to point out that spouting a childish mantra every time your errors are mentioned is not the behaviour of an intelligent person, but you seem to have already given up any claim to that by deferring blindly to what you perceive as the infinite wisdom of what is in truth a useful but imperfect tool.
I didn't read Opto's reply yet. Maybe later I might read part of it.
He reads all of my posts. He often agonizes.
Usually I prefer to read after posters much better than Opto have posted - and there are lots of those people.
In fact - I'll check back now - it'll be easy to find a more worthwhile post by somebody else to read.
And we are still waiting for you to REFUTE any of this from Grok, or myself.
But you can't......You no nothing but bad time stories. And not what is needed. Knowledge of chess, computer chess, and game theory.
You don't understand enough to see when Grok gets it wrong, and you (inappropriately) worship its infallibility so are blind to intelligent input from people who know much more than you.
Fairytale time is OVER!
I have to point out that spouting a childish mantra every time your errors are mentioned is not the behaviour of an intelligent person, but you seem to have already given up any claim to that by deferring blindly to what you perceive as the infinite wisdom of what is in truth a useful but imperfect tool.
Its not clear yet though as to whether Dubro's just a bit over-enthused about Grok plus having a teenage 'rant party' - or he's got some other agenda (returning account? 'Jalex'?)
Will Dubro keep making the same mistake of posting the same goblin image from Grok over and over again? Does the 'suspense' build?
Dubro not as robotic as C-Rat (ibrust) ... but D good at getting attention!
---------------
Dubrovnik posted three 'water is wet' arguments about checkers just now that Elroch would be well aware of.
But the discussion about chess-solving being expedited by dismissing superflous positions - has never been very thorough here.
That's what players do in their 'personal solving' of chess both at the chessboard and their studies of the game.
Superfluous positions are skipped.
For over 500 years now.
Just FYI your "I like this idea" is basically done in real time by most of the best chess engines for years.
@playerafar's idea is inchoate and would never have been incorporated in any chess engine ancient or modern, especially since they're not designed for tablebase generation.
It is called tablebase probing.
And it's not called tablebase probing, whatever Grok the world's smartest AI might tell you.
It is no wounder this thread is filled with nonsense. As these people have no clue about computer chess.
A high percentage by yourself of late.
And I never laught so hard in my life. As hearing their nonsense that chess could be weakly solved exactly like checkers.
And Grok the world's smartest AI sold you a heap of crap about what a weak solution is, so perhaps you should work out what you're laughing about before you start laughing.
I guess this was their logic.
Checkers is played on a 8x8 board.
Chess is played on a 8x8 board.
And no more thought was needed to justify their nonsense.
And we are still waiting for you to REFUTE any of this from Grok, or myself.
If by "this" you mean your post I was responding to above, the refutation is the stuff in orange (maybe you missed it).
If you're referring to the Grok dump you've previously pestered about so often, my answer is the same; I'll do it in my own time. Try a little patience.
If you get bored you could try answering #19606 as requested several times.
But you can't......You no nothing but bad time stories. And not what is needed. Knowledge of chess, computer chess, and game theory.
In such clear contrast to yourself. I'm sure we're all in awe of your prodigious erudition.
Fairytale time is OVER!
Not really possible to refute silly pictures, only to point out that someone who keeps posting them must have a serious personality disorder.
Why Not Solve Chess Like Checkers?
Scale: Checkers’ weak solution relied on exhaustively solving a subset of positions (endgames) and connecting them to the start. Chess’s endgame subset is too small (7 pieces is a tiny fraction of the game), and the middlegame is too chaotic to bridge the gap.
Heuristics Fall Short: Chess engines like Stockfish approximate perfect play with evaluation functions and pruning, but they don’t prove an outcome—they just win practically. Checkers’ simpler rules made exhaustive search viable; chess’s complexity defies this.
Starting Point: Checkers’ initial position was directly linked to solved endgames via search. Chess’s opening tree is too broad, and we lack a critical "solving position" to anchor a weak solution.
For now, chess remains a mystery. Checkers’ weak solution was a triumph of computation over a manageable problem; chess demands a fundamentally different strategy we haven’t yet cracked.
Ah! If only you could read! Nobody on the thread has ever suggested that a solution of chess along the lines of the checkers solution would complete in the foreseeable future. Not even @tygxc.
Regarding Lola's post (she finally criticized O- took awhile)
O has been spamming his 'verdict' on himself for ten years now.
But his posts always need some translation.
In other words he usually means an opposite of what he's spamming.
You've called everyone here stupid but what's your verdict on yourself?
theyre just wishing they had the memory of a squirrel like u opti. just thinka all the nuts youve found here.
Researchers tagged squirrels in New York City's Central Park and set up cameras to record their activities several years ago. It turns out that squirrels have no memory of where they buried nuts, but if lots of squirrels bury lots of nuts any squirrel will later be able to find a buried nut by scent.
@MARattigan, wouldn't you agree that saying "a fundamentally different strategy" to solve chess is much like saying "we need faster than light drive to travel around the galaxy"?
i.e. the truth is that there is almost certainly no such strategy that will take us a large fraction of the way to a solution being possible (on a log scale). Rather the large majority of what we need is an increase in computatational resources, whatever method is used.
When comparing here, I am thinking on a log scale. Say we need 10^15 more computing power (not a precise number), we might pull that down by a factor of, say, 10^2 with brilliant advances in strategy, but the large majority of the gap remains (on a log scale).
That refuted nothing.
All you can do is tell fairytales.
You can not even get the tygxc information right without lying.
tygxc said chess could be weakly solved in 5 years many times, and that started 3 years ago moron.
But not with Schaeffer's procedure O highly intelligent being.
That refuted nothing.
All you can do is tell fairytales.
You can not even get the tygxc information right without lying.
tygxc said chess could be weakly solved in 5 years many times, and that started 3 years ago moron.
But not with Schaeffer's procedure O highly intelligent being.
Yes. @tygxc was more saying that it would be possible by five years of effort to become more convinced about the answer we already believe, without getting anywhere near a true solution. He obfuscated this by obstinately misusing the term "solve".
I am confident it doesn't exist. If it did, it would probably be publishable and, as you say, it is likely that someone else would spot it - there are some very smart and hard-working people out there.
@MARattigan
spotted your latest diagrams - but i think you missed my point about 'obvious positions'.
Because you displayed a position with a lot of bishops moving on the same color squares.
I'll read back some more - but the point isn't about such types of positions.
Its about lopsided material advantage where the other side has no stalemate or other draw refuge like perpetual check or other counter-play to rescue the position.
This idea of computer-solving projects (not Stockfish not Komodo) skipping such positions - hasn't really been discussed in the forum yet.
But how much interest can there be in any forum ?
The people doing the tablebase projects have an obvious 'interest'.
Its called a paycheck.
Stockfish evaluates this position as +13.6 for white. Should be enough, surely? (Chess.com analysis doesn't correct it, but uses some heuristic to conclude what the value is).
It doesn't matter for practical chess.
@MARattigan
spotted your latest diagrams - but i think you missed my point about 'obvious positions'.
Because you displayed a position with a lot of bishops moving on the same color squares.
I'll read back some more - but the point isn't about such types of positions.
Its about lopsided material advantage where the other side has no stalemate or other draw refuge like perpetual check or other counter-play to rescue the position.
This idea of computer-solving projects (not Stockfish not Komodo) skipping such positions - hasn't really been discussed in the forum yet.
But how much interest can there be in any forum ?
The people doing the tablebase projects have an obvious 'interest'.
Its called a paycheck.
So what is your suggested algorithm for not discounting the Bláthy diagram? Solve chess and see if it can be discounted?
@MARattigan, wouldn't you agree that saying "a fundamentally different strategy" to solve chess is much like saying "we need faster than light drive to travel around the galaxy"?
i.e. the truth is that there is almost certainly no such strategy that will take us a large fraction of the way to a solution being possible (on a log scale). Rather the large majority of what we need is an increase in computatational resources, whatever method is used.
When comparing here, I am thinking on a log scale. Say we need 10^15 more computing power (not a precise number), we might pull that down by a factor of, say, 10^2 with brilliant advances in strategy, but the large majority of the gap remains (on a log scale).
I would say - by much more than 10^2 ...
but an improvement of 10^15 times as fast - doesn't look good enough against that 5 x 10^44 upper bound on the number of legal chess positions.
There hasn't been much good discussion of what actual ratio of chess positions are superflous to solving chess. If any.
Certainly tygxc's claims weren't 'good discussion' of that.
There was a 'tablebase project' here to 'table' his claims.
Was that 'weakly solved'? Or strongly?
---------------------
So many lopsided positions that are superflous.
Six knights on their back rank against lone king on his back rank.
Or their second ranks and so on.
Would it be that hard for the computer projects to count up obviously lopsided positions and see how much they impact John Tromp's number?
Hey maybe Grok knows ... !
169 is extraordinarily high...i should say. and id never-ever dispute this w/ opti. he's exceptionally well-rounded imo. and my guess is hes a scary sponge in anything he wansta learn abt. retention is...well lets just say he probably hazza a memory like a salmon.