Chess will never be solved, here's why

Sort:
playerafar

"Weird you should have imagined I asked something like that"

I'm referring to a possible future event.  Not a past event.
Should I be surprised that 'past' was read in?
No - but that doesn't have funny value.
What was funny to me was the preposterous idea of giving somebody advice about how to approach a family member in a single particular future context.
Which you prompted by choosing to mention such approach in the first place.

As was said - a 'rare' moment of direct communication.

LeoTSimoes26

One day i think it might be solved but can still be played by humans

playerafar
Optimissed wrote:

Thankyou for adding what you understand about the term. I agree with you that it seems to be a sort of techno-jargon, used to establish and maintain an illusion of knowledge. I'll look it up. I did the first year of a computing degree before changing to philosophy and can't remember nodes being mentioned. Top-down programming yes, nodes no. It was about 1992.

So I mentioned the idea of approaching your son.
Who you've touted many times here - and mentioned conversations about on this subject.
Should I indulge your obsession with credentials by reminding you that he 'knows much more than you' ?

Again - 'nodes' here isn't about credentials nor knowledge levels.
Nor about 'you and me'.
Its about a particular person using the term to defeat a much more realistic conversation about daunting computer projects in chess that would likely occur if the true technical speeds of computers involved was openly factored into the conversations.

playerafar


The posts of @Optimissed have disappeared from this forum.
Which suggests that he has been muted by chess.com.
When I went to the forum there were a great deal of empty pages at the end.
Mutes of a member often cause that.
As to whether its one of those one day mutes - I don't know.
But my post here should fix the empty pages thing.


taylorsinha

According to me, you should practice more and more.

Then you'll love it..

 

TeacherOfPain

@playerafar 

As long as the game is finite, the game of chess will be solvable.

It is not an infinite game, therefore the game will be able to solved, it just depends on which calculations and permutations you use to solve them. Statistically and realistically speaking it can be solved, and though movements are dependent on individual processes of logic(or creativity), dependent on a given position, only the moves that are best or good, will be played, in order to win or 'solve' a given game or position. 

From the perspective of a human, this is unreasonable, although a human being could good at calculating multiple lines to perfection, we all have our limits, as well as the GM's. So accordingly, the only way this game would be solved would be by a chess engine or supercomputer, which would no doubt take some time, but eventually could be done. 

For this matter, it seems clear, that the game is finite, and since it is, it can be solved, especially if the goal is aimed at objective piece placement between two engines, or higher level computers. 

The only way I would overturn this discussion to say that it would not be solvable, is if all of these moves were based on subjectivity, which then would add another layer of 'if chess is solvable'. Subjectivity in this sense, wouldn't allow the game to solvable(in my view), because without the the rigidity of optimal move placement, then perhaps no game(within our outside of chess) is solvable, because a person is given a choice based on intuition, human bias, error, and feeling, rather than structure, logic and order. 

Hopefully, this was informative or helpful in anyway, and can contribute to any further discussions. 

 

playerafar


Practice playing chess?
The chess I like the most on the site these days is the tactics puzzles.
I play them unrated.  For a while now in the 2600-3200 range.
(my actual playing strength probably well under 2/3 of that)

playerafar


Hi @TeacherOfPain
While a finite number of positions does indicate solvability in theory -
we've had a tremendous amount of spam about 'solved in five years'.
I'm predicting more of such spam shortly.
With the hardware speeds of computers concealed by using a misleading units system.
And that spam being doubled down by arbitrarily reducing the number of positions to be solved to a millionth of a trillionth of the actual 38 digit number of positions to be solved.

To get a much better conversation started - the realistic speeds of modern supercomputers needs to be presented in realistic units.
Then somehow an average time needs to be somehow figured simply to consider a position even just one move ahead ...
Then those values need to be compared to the upper bound of positions to be solved.
Then some measure of the number of billions of years for the task could be further discussed.

I know somebody on the site who's a programmer.
He might know how to present the hardware speeds of modern computers properly and to properly relate that mathematically to the real task.

chessisNOTez884

cant you all read the biggest ever novel made in history rather than typing this nuisance?

playerafar
sachin884 wrote:

cant you all read the biggest ever novel made in history rather than typing this nuisance?

You take the time to read it and then complain?
Lol.

TeacherOfPain

@playerafar 

Mathematically speaking, it would be difficult to convert those units because they would be very being in a quantifiable way. Supercomputers in theory would be able to solve these positions, however it would take a good while for there to be a solution.

When regarding the timeline, I have no idea when it would be complete, however I am confident to say that it would be solvable in time. I doubt that it would take billions of years, as you said before, but considering the progression of human and computational mathematics, it could take years to make progress to solve.

Yet, it would be largely determinant, on either if the game would be subject to an objective approach/perspective to the game, or subjective. If the game is reliant on objective move orders it would be much easier to find a solution. However, if the game is subjected to subjectivity in a position(aka. chess intuition, or tactics), then there would be a longer time. 

Of course multiple games would need to be played to verify this, for their to be a accurate answer to determine what works and what doesn't. 

TeacherOfPain

Grammatical mistake: "...those units because they would be very big, in a quantifiable way.[to add: In a human sense to be simplified or otherwise deductible by human logic.]" 

playerafar


" it would be difficult to convert those units because they would be very being in a quantifiable way"

@ TOP
You don't Need to 'convert' proper units of computer speed.
Speed speaks for itself.
Once the true hardware speed is established (should not be difficult for a properly informed computer scientist) - then the size of the task can be divided by that speed to produce various upper bounds (and lower bounds) on the time needed.

What would present some difficulty is - how many unit operations needed to evaluate a position?
It would take a certain number of operations just to present a position.
A minimal position like K vs K+pawn.
Then a certain number to present all possible legal moves available.
But lower bounds could be established.
Then - lower bounds could be assigned based on a 45 digit number of positions.

The times estimated will always be lower bounds - because the number of operations needed to properly evaluate a position could be Huge !!

Classic example:  the opening position.
You can't.  Too many operations needed.
So - the whole thing has to start from the other end.
But one could begin to establish lower bounds of time needed for simpler positions.
But you need that hardware speed and proper units.
In some cases - they're already known.
But unfortunately they had to leave out castling rights.

TeacherOfPain

@playerafar

Why do you highlight your points? Is this how you manage a discussion? It is rather odd, but ok, I accept it. 

But like you were saying, yeah, sure, speed is measured in speed, but if this is the case why did you mention mathematics in the first place? However, I see where you are coming from in terms of how units are measured. 

In this case, it would make sense, how a supercomputer would manage a chess position, from a hardware standpoint. 

In this case, reverting back to the discussion the overall way to determine if a position is to be solved quicker would be if, there is more or less subjectivity involved.

That's make take on it. If there is evidence I will change my mind, but otherwise, this would be the end-goal of chess being solved, albeit in a very simplified form. 

idilis
playerafar wrote:
sachin884 wrote:

cant you all read the biggest ever novel made in history rather than typing this nuisance?

You take the time to read it and then complain?
Lol.

The child is all contradictions

https://www.chess.com/forum/view/general/is-chess-an-sport

https://www.chess.com/forum/view/chess-com-community/pls-block-this-forum-70616775

Just needs some attention

playerafar


"But like you were saying, yeah, sure, speed is measured in speed, but if this is the case why did you mention mathematics in the first place?"

odd question.  You don't think speed is part of mathematics?
I think I'll ignore that one.
'Evidence'?  Math is evidence.  Real math.
As for 'subjective solution'  that's available right now.
Including with the Snake Oil of 'nodes per second' and 'taking the square root'.
'will change my mind'?
The actual math and the actual task wouldn't care about anybody's mind.

Example:  'The orbitting planets obey Kepler's laws' ...
No they don't.
Kepler's laws might approximate a good description of what happens - but its more complicated than that.
And they don't 'obey' either.  'Laws' don't define nor ordain ...  they attempt to describe.  And the planets couldn't care about Kepler himself either.
Its like anti-evolutionists trying to argue its all about Darwin.  Evolution couldn't care less about Darwin.

But perhaps some things are beyond 'Laws'.
Like causality.  Every effect has a cause.  Every cause produces an effect.
Conservation.  You can't get something from nothing nor nothing from something ... but some try to argue it happens in Quantum situations.
Reverse time travel:   You can't travel back in time.  But some might try to argue that Kurt Godel proved you could.

In math though - its tighter.
No greatest prime number.  Neat proof of it.
But it would be hard to prove it to a Flat Earther if he didn't want it proved to him.
'Evidence and proof and might change my mind' ...
if you want enough to find out about these things you will.

TeacherOfPain

@playerafar

I recommend that you help a player out by stop being so pretentious in your attempt in conversation my friend(no offense by any means, as you do make sense, it just doesn't seem necessary for the conversation we are having). I do admit that your information is helpful, but you don't say things in the right way, which makes me wonder what is the purpose of this forum in the first place, if you are just trying to show what you know, instead of having a discussion about opinions, and yes, at times, facts. We were talking about the game of chess being solvable, but now we are on a completely different thread, which is not why I came here, in honesty. 

I think you are a respectable player, no doubt as we've had our discussion in the past however, I just don't get why you are so uptight about these things. But then, I've said my piece, and I've learned some things on this forum, so I'd say, there is more to learn, in out and out of the chess board, so this forum is refreshingly different and exploratory as well. 

Whatever the case may be, good journeys to you and as always may we have good games! 

playerafar


"
so pretentious"
No pretense at all.
I meant every word.
" I do admit that your information is helpful"
then suggestion:  act like it.  Instead of confusing yourself.

to almost everybody else:
If the whole issue of basic units of computing speed is not brought into the discussion then there'll just be another 1000 posts of spam and another and another.
To do better - people needed who can quickly overcome any nonsense about 'nodes per second' and other spam.
For that - some credentials needed probably.
But many here might have friends who are more than versed enough about computer hardware.
There still might not be enough interest though.  Even then.
Usually - people go where the money is.  Especially when its 'professional'.

I know somebody who's a programmer. 
But I don't want to bug him about this right now.  grin
Tiny pun there.

tygxc

#3507
"When regarding the timeline, I have no idea when it would be complete, however I am confident to say that it would be solvable in time."
++ As the late GM Sveshnikov said: "Give me five years, good assistants and the latest computers - I will bring all openings to technical endgames and "close" chess."

Elroch

So, some finite time. Perhaps a billion years? A trillion?

Sveshnikov's statement clearly referred to a non-rigorous practical analysis where implausible moves are ignored. Enough to convince a chess player, perhaps, but blatantly not good enough for a solution.

Do you understand what a rigorous solution of a chess problem is?  (Genuine question, not rhetorical).