Will computers ever solve chess?

Checkers was a lot harder to solve than tic tac toe [although we knew the answer ahead of time]

However one cannot really compare solving chess with solving checkers because of the math.

One can stick to their guns but they will not be correct--chess is not going to be solved even if we wait until the time our sun explodes [really expands]  

We already know chess played correctly is a draw but this knowledge has little effect on chess. 

@JeffGreen333:  That would not surprise me.  The other thing I wonder about is: how long would the final winning for white or drawing for black combination be?  Would it be a 400 move monster game with lots of maneuvering that finally either gets white the breakthrough or guarantees black the draw?  That actually would be kind of a best case scenario, guaranteeing that chess would ALWAYS be viable for human play, with computers being used mainly to make it easier for humans to learn the various parts of the game.

@ponz111:  You are EMPHATICALLY sticking to your guns.  Ok I get it, you BELIEVE STRONGLY that chess will never be solved.  We'll know in the year 2063 or so, and if I am still around at age 103 and it turns out that chess is still not be solved, I would be DELIGHTED to have you come over to tell me I told you so!

Actually my belief is based on evidence.

However i will be delighted to tell you chess cannot be solved--tell you this in the year 2063--as i will be 122 years old then.

My gut feeling is that such a game would end in a perpetual check draw, long before 400 moves were made.   Especially if it's two computers playing each other.   I base this on the fact that a computer will always make the best move, even if it leads to perpetual check, rather than making an inferior move in order to avoid a draw, like a stubborn human being might.

I dont know what Jeremiah was smoking either, but I do know walking on the moon was impossible. Then. Solving chess is impossible, now. But I am curious what historical reference says walking on the moon was obvious.

When did you know walking the moon was impossible?  For my whole life of 76 years--i never knew walking on the move was impossible.

Regarding historical references that walking on the move was obvious--there probably is none. People learn things at different rates. Scientists may know things which most people do not know. Very educated people know things others do not know.

Some folks say it did not happen anyway. But, I will not go there.

OK, I am going to say right now that going to the centre of the sun is impossible.

As long as when it happens, I get invited to the barbecue pre-lunch, make that pre-launch.

No human know for sure the outcome of this endgame position. 

The complexity level is enormous, only accurate play on both flanks may cause zugzwang for white. 

Despite black's material advantage, white's central knight is posted well and covers most vital squares. 

It is not easy for black to cause breakthrough.

What is your say on the likely outcome?

If I can see it with my naked eye, then I can get there eventually.   No historical reference needed.   It's just common sense.

This, I agree with.  In fact, walking on Venus is probably impossible, since every probe that they've tried to send there burned up or crashed on the surface.   But the moon is much closer to us and doesn't have the high temperatures that Venus has.   I also believe that walking on Mars is possible, although it would be a very costly and dangerous trip.  

There is no real reason that walking on Venus will not be possible in the future. We may not want to do it but it could become possible.

I have solved positions which the best chess engines could not solve.

Strong chess engines are usually stronger than humans but they are not gods. 

Saturn's

I also wondered whether any major proved mathematical result had ever been found to be wrong and asked one of my math professors about it.  He looked puzzled for a moment and said "None I've ever heard of.  I'd be surprised if there are any.  After all, if the proof is solid then it's a proof."  So...up to you what to make of that.

You can think that the complexity difference between Tic Tac Toe and Checkers must be about the same as that between checkers and chess.  Or you can look at some numbers.  Tic Tac Toe is on the order of 10^5.  Checkers is on the order of 10^20.  Chess is on the order of 10^120.  So, Checkers is 10000000000000000 times bigger than Tic Tac Toe and Chess is 100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 times bigger than Checkers.  Not at all proportional.

As to Jeremiah Washington, had he known how much fuel it would take to propel a person from Earth to the Moon, based on the distances and forces involved, and had he pointed out that even if the whole Earth were used for the task it would not be sufficient, then he'd have been right.  But he didn't know what it would take, and the Earth does happen to be sufficient.  So he was giving an opinion, not making an argument.

As to the thing that is 10^100 times faster than current computers, I would say that too is impossible but it would become a circular argument.  So, what I already said about a similar thing is if God Himself (or your 10^100 machine if you like) were to give you the answer, you could not store it in a memory made of the whole universe.

The concept of combustion engines hadn't even been invented yet, in 1783, let alone any conceivable notion of rockets.   So, he wasn't basing his claim on the size of the rocket needed.   He was wondering how a horse could carry a human being to the moon.   lol   I'd imagine that a visionary type, like Leonardo Da Vinci, envisioned aircraft or rockets though, well before 1783.   Btw, who the heck is Jeremiah Washington anyway?   Why are we basing this on some unknown person's inability to foresee what the future might hold?   Or did I skip that class in history?   lol   Da Vinci was already drawing sketches of flying machines in the 1500's.   

Well...I think my point is not far from yours.  He wasn't saying "we know what is required and we know it exceeds all available materials and therefore it is clearly beyond us."  Rather, he was saying "we don't know what is required but it is so vast we will clearly never know."  And that is not the same thing.

Two more points as I think about this.

To the person who asked whether anything considered mathematically proven had ever turned out to be not true, I said that I had asked a math professor and he couldn't think of any.  But here is something close.  Euclid's Fifth Postulate.  Euclid made five postulates which he considered to be true without proof.  The first four are to this day considered true but unprovable (which is what makes them postulates), and are taken as starting positions in geometry.  The fifth one, Euclid also regarded in this way, but countless mathematicians over 2,000 years believed it was not a postulate at all (because it could be proven).  And countless of them tried to be the one, because it would bring immortal fame.  But then, in the 1700s/1800s, several mathematicians actually proved that it is not true at all in curved space.  Which is the origin of non-Euclidean geometry.  So that is the closest thing I can think of.

Second, someone asked what is a quantum computer.  So this is pretty cool.  Assume you have a 64 bit byte:

00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

represents zero.

00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001

represents one.

11111111 11111111 11111111 11111111 11111111 11111111 11111111 11111111

represents 2^64 - 1.

In a regular computer, each bit is either 0 or 1.  In a quantum computer, each bit is 0 and 1 AT THE SAME TIME.  This might be theoretically possible, using quantum uncertainty.  As a result, for a regular computer to check each of these values, it would have to do 2^64 - 1 operations.  A sufficiently powerful quantum computer can do it all in one operation.  If it isn't powerful enough, then it can't do it at all.  So a sufficiently powerful quantum computer COULD solve chess...it just can't exist in the universe because there isn't enough material to make it.

Now, if there are infinite universes, and you could devote 10^100 whole universes to being your quantum computer, then you could solve chess at that point. (Taking than number because we have solved checkers in one universe, and chess is 10^100 times bigger.  You actually probably wouldn't need quite that many universes.).

But what I'm trying to show here is the level of difficulty of the problem, because the question was "will we ever" and the answer is "no" in my view, but it is at least "no, unless you think we will someday build a quantum computer out of 10^100 universes."  If you think we are going to do that (and use said computer to solve chess), then I guess it's possible.