It's well established that chess is solvable, and I think it's also well established that the ability to do so is well outside of our current computing capabilities.
It's well established that chess is solvable, and I think it's also well established that the ability to do so is well outside of our current computing capabilities.
Maybe I'm just being naive, but modern computers have enough power to simulate incredibly complex systems like the weather, and the human body. I find it hard to believe that a Cray supercomputer or something wouldn't have the capacity to solve chess, if it were possible. Maybe its just wishful thinking on my part.
The complexity of the game is quite simply out of reach -- not only for our current computing capacity, but for classical computing in general.
We've only managed to solve the game for seven pieces (total) or less and each piece you add increases the complexity exponentially. Solving the game is equivalent to establishing a 32 piece table-base.
The only problem I can see is that if a solution is found, it probably won't be too hard (over time) to memorise the moves of that solution. No need to understand the rationale behind the moves, just memorise...
At which point - welcome to the stage Chess960!!!
Hmm, I don't think so, Mike :)
There are probably a Gazzillion solutions, not only one. If the position is a draw, there are possibly so many moves that keep the draw, and then many replies, etc.
The only problem I can see is that if a solution is found, it probably won't be too hard (over time) to memorise the moves of that solution. No need to understand the rationale behind the moves, just memorise...
At which point - welcome to the stage Chess960!!!
Um, no. Even if you can vastly cut down on the complexity by only memorizing the best response to each of your opponent's candidate responses you only cut the tree you need to memorize in half (because the tree grows exponentially with every full move instead of with each ply). This is still easily outside the realm of what it's possible for a human to memorize within a small handful of moves.
No.
A computer cannot think like a human, and neither can we think as a computer. Computers can do calculations in 1 second that would keep a human like me, or even Kasparov, busy for days! Even if a computer would "solve" chess (what -in my humble opinion- is impossible), there would be no reason to quit playing the game, as nobody would profit from that!
My explanation about the calculating doesn't mean that computers can't be beaten in a same time span.. Let's say that -hypothetically - a computer can do 1000 calculations per seconds, and that we can do 1 per 3 seconds. Gives us that a disadvantage? Depends on how you look at things, and how the computer is programmed, etc. If you play a computer like rybka on the highest difficulty level ( so the most calcs/sec), it's impossible for us, hobby-players, or even IM to beat him, but he CAN be beaten ( look at the match Kasparov - Deep Blue). Therefor having time at your side (or the chess computers' ) doesn't mean that you are "discriminated".
A computer calculates every single legit move, and we filter out a lot of those moves! If we want to talk how good these calculations are, we are speaking A.I. (Artificial Intellegence), which is a totally other subject than what we're discussing here.
Like I said from the start, computer can't think as humans, they are only as good as the persons who made him, only faster.
I hope this clears out a bit of your issues =)
Ofcourse, all of this is hypothetical, I don't think it's possible to solve chess, but for the believers, have a nice read ;)
Imagine a tournament in the future, the 11 best players in the world, and also the computer beast. All players memorize the "perfect game", and draw against the computer, which scores 50% for the tourney. But the humans win and lose against each other, and the computer finishes halfgway down the field... Yeah for the humans 
It's well established that chess is solvable, and I think it's also well established that the ability to do so is well outside of our current computing capabilities.
Maybe I'm just being naive, but modern computers have enough power to simulate incredibly complex systems like the weather, and the human body. I find it hard to believe that a Cray supercomputer or something wouldn't have the capacity to solve chess, if it were possible. Maybe its just wishful thinking on my part.
maybe i'm just being naive, but modern computers don't have enough power to simulate incredibly complex systems like the weather...
(they get the weather wrong all the time, and they can at best offer a suggestion of what it may be like, and let's not even get into the butterfly effect because we have absolutely no idea how that works, and it directly dictates weather patterns.)
...and the human body...
(the most complex part of the human body, the brain, is not even close to being fully understood yet, for example how electric impulses translate into thoughts, or how our unconscious minds work and influence our conscious minds.)
...I find it hard believe that the Cray supercomputer or something would have the capacity to solve chess, if it were possible...
(because the complexity of all the possible moves in chess is absurdly huge, allegedly more positions than atoms in the known universe. so perhaps once computers can conquer the universe, they can move on to more important things like chess.)
...Maybe it's just pessimistic thinking on my part.
It's well established that chess is solvable, and I think it's also well established that the ability to do so is well outside of our current computing capabilities.
Maybe I'm just being naive, but modern computers have enough power to simulate incredibly complex systems like the weather, and the human body. I find it hard to believe that a Cray supercomputer or something wouldn't have the capacity to solve chess, if it were possible. Maybe its just wishful thinking on my part.
Weather is not complex to a computer. Heck, even humans can predict the weather.
I wasn't talking about predicting the weather. I was talking about computer simulations of it, where trillions of molecules are involved.
Anyway, it is a moot point imho. Even if chess is solveable, it still won't matter on the human level.
I mean it is theoretically possible to create a robot golfer that has the perfect swing every time. Doesn't mean that you can do it.
I don't care about human v. machine. I care about human v. human.
BigPoison's story about Alekhine and Lasker is hilarious.
I believe it is widely accepted that given our current state of technological development, chess is not capable of being solved at this point in the time line.
Thus, assuming that chess is solvable, I would submit that it would be possible to solve chess if a sufficient enough technological development takes place....it may not be in our live time...maybe 1000 years from now...maybe even longer...who knows.
Let us now enter out time machine and travel say 1000 years into the future...to the point when chess was eventually solved. We would be introduced to the super-computer which did this feat. Clearly, this would be a very unique piece of hardware and software, too expensive for for anyone individual to own (more than likely, a combined-international cooperation)...and the results of this equation would require a huge storage capacity.
With all this being said, this result may be relevant for computers playing chess, but not really relevant for human vs human chess play.
Computers weigh more heavily on material advantages, as opposed to positional play, they cannot plan, they cannot formulate strategy...and this is the main reason Kasparov was able to defeat Deep Blue...sacrifice material for a positional advantage.
In terms of play, there is hardly and difference for a human. Soon engines will be like 3500, i.e. it won't be possible even to draw them. And then only a fight vs an even more stronger engine could prove something...
Well, of course, it would be fun to learn, for instance, that some lines lose by force. 
However, no one will be able to memorize why that is so.
That's what I meant Natalia :) You won't win, but he doesn't play perfect chess!
They solved baseball!?
Let me guess: It's always a loss for the Cubs?
and the most money wins.
Not true. The most money wins often, sure, but how do you explain the two Marlins World Championships, not to mention the upcoming 2010 Tigers world championship?
i can't explain that because i was joking.
LOL!
So perhaps logic will win out over memorization in terms of playing chess when the futility of memorizing craploads of lines becomes insanely depressing even for the genius starting at age 2?
What if chess went the path of starting from random positions considered equal (as determined by the future 3500-level engines) and players were forced to accept whatever "system" happened to present, thus forcing players to either learn all or no opening systems, in essence a version of "random" chess, yet still chess?
What if man solves chess? Will the computers quit playing?
EXACTLY GOOD POINT
I used to run from 30 to 34 miles per week. Unfortunately I had to quit due to some physical issues. I could run a 10K in About 42 minutes. Respectable but not good enough to even win local events. Didn't matter because I really enjoyed the activity of running. If its even possible to solve the game of chess, I don't think this event would cause many people to quit playing for the same reason. And can't we find a more productive use of computer time than chessbrain? Medical research maybe?
Yeah but can that perfect robot golfer anticipate when a live bird will collide with his golf ball?
Perhaps not the best analogy.
i remember when i pegged a bird while golfing. it was a beautiful drive off the first tee. nice and straight (which i'm not accustomed to hitting). stupid bird flies right into the ball and drops to the ground. the ball bounces back towards me and almost rolls up back onto the tee area.
i decide to hit another ball. slice it off to the right obviously. we then start walking up the fairway to the bird that's just sitting there. as soon as we get close to it it flies away. i think it was just trying to screw me over and faked its injury.
I wasn't talking about predicting the weather. I was talking about computer simulations of it, where trillions of molecules are involved.
Sorry, but computers can't simulate systems containing trillions of molecules. I perform molecular dynamics simulations for a job, and I can tell you with full authority that the size and time scales that can be simulated within a reasonable time with one new CPU are of the order of thousands of atoms for a simulated time of nanoseconds. With parallelization methods (linking a number of CPUs together to calculate separate regions of the system simultaneously), the number of atoms can be increased to several million, but the time scale remains the same. In other words, real-time simulations of macroscopic systems cannot be performed.
Meteorological simulations are based on approximations that have been proven to work for most of the time - the wilder the approximations, the less accurate the results. "Solving" chess would require exact calculations with no approximations whatsoever, which means that such an endeavor is way out of our reach.
If both sides play perfectly, it will always end in a draw.