Checkmate

Originally posted by ThudanBlunder
Obviously you can't prove it but what evidence do you have for this claim?

common sense. it wouldn't be an advantage if perfect play couldn't win.

but what if that advantage alone cannot be converted into enough to win the game, with perfect play from black?

Originally posted by mazziewag
but what if that advantage alone cannot be converted into enough to win the game, with perfect play from black?

then it is not an advantage and both players are equal.

it might still be an advantage- if the misstake needed by black for white to win was smaller than the one needed by white to commit for black to win then white has an advantage

Originally posted by omulcusobolani
I'm pretty sure the game hasn't been solved yet, so it could be either way...maybe white is in zugzwang?

Yes - it's perfectly possible that White is in zugzwang and that chess is a win for black assuming best-case play from both sides.

I would say that unless something magical happens in the realm of quantum computing, chess will never be solved.

Ever.

If it were solved, of course, my life as a chess programmer would be easy - absolutely no need for qualatative evaluation at the specified nodes of a search tree - just look up the correct move in a database...

Originally posted by tomtom232
common sense. it wouldn't be an advantage if perfect play couldn't win.

Common sense? If it is common how come only you have it?

What is your definition of 'advantage'?
What is your definition of 'perfect play'?

In noughts-and-crosses (tic-tac-toe), the 1st player has the advantage of moving first. Because it is a much simpler game than chess, this advantage is greater than it is in chess. Yet the 1st player cannot force a win. even though he plays perfectly.

Originally posted by tomtom232
Actually with a perfect game white should win.

No, that is simply incorrect. If that was the case then there would be no point in playing the game!

Originally posted by Chris
Yes - it's perfectly possible that White is in zugzwang and that chess is a win for black assuming best-case play from both sides.

I would say that unless something magical happens in the realm of quantum computing, chess will never be solved.

Ever.

If it were solved, of course, my life as a chess programmer would be easy - absolutely no need for qual ...[text shortened]... uation at the specified nodes of a search tree - just look up the correct move in a database...

Opening lines come and go. Some are analysed to death and prove winning for one side. When that happens people stop playing it and move onto something else. It is impossible to have an opening that wins in every variation. There has been too much analysis over the last couple of hundred years for this "One hit kill" opening to be discovered.

If you look a players like Kasparov, do you see his opening repertoire slowly diminishing into a set of winning lines over time, that he simply repeats for a garunteed win? No, of course not! His games are analysed by people the World over. The people he played (top 20 in the World strength at least...) will play through all of his games and find refutations that his opponents missed OTB. "Solving" chess is like finding the answer to "Life the Universe and Everything". Chess is 'cause and effect', you can never solve it unless you literally calculate a winning continuation from every legal position (hundreds of billions of them as i understand it). If one was to attempt to do this it becomes clear very quickly that it isn't possible. for example...



This is a legal position. How long would it take to assess this as winning for either side? It's been around for about 400 odd years and has been analysed by some of the sharpest minds the World has to offer, not to mention the computing power of countless thousands of computers. So after 400 years we have hundreds of thousands of players who will quite happily play as white or black in this position! Now times that by the number of legal positions and it is quite clearly impossible to create such a database, not simply because of the vastness of the data, but mainly becuase of the nature of the data, which is based on argument and counter argument, going off into infinity....

Originally posted by Freddie2006
How can anyone play the perfect game? Is that possible?

It is possible since the possibilities are finite. Non-simultaneous games with finite possibilites are solvable.

Originally posted by Marinkatomb
Opening lines come and go. Some are analysed to death and prove winning for one side. When that happens people stop playing it and move onto something else. It is impossible to have an opening that wins in every variation. There has been too much analysis over the last couple of hundred years for this "One hit kill" opening to be discovered.

It's doubtful, but not impossible.

As someone else pointed out there are a finite (albeit incredibly huge) amount of possible positions in chess, so a perfect solution, or more likely, perfect solutions, given best play by both sides, do exist.

That's not to say these solutions will be simple, they most certainly won't be. They could be something absurd like mate for black in 5553 moves!


Whether or not we could ever find a perfect solution, or reckonise one when we see it, on the other, is incredibly doubtful, but until we do, we will never know what the outcome of a perfect game is.

Originally posted by Marinkatomb
"Solving" chess is like finding the answer to "Life the Universe and Everything". Chess is 'cause and effect', you can never solve it unless you literally calculate a winning continuation from every legal position (hundreds of billions of them as i understand it). If one was to attempt to do this it becomes clear very quickly that it isn't possible. for example...

And oh yeah, there are *a lot* more positions than hundreds of billions. If there were "only" hundreds of billions of positions, chess probably would have been solved by now.

Anyways, solving chess may be a practical impossibility (but who really knows?), but it's definitely not a mathematical impossibility.

Perhaps back to the subject of the original post, would I be the only one interested in seeing a tournament where the resignation option is turned off? Being a low rated player, I can't always pick out why a position is lost, and I think it'd be really cool to see some of those endings in action as opposed to theory.

Originally posted by Fat mans revenge
Perhaps back to the subject of the original post, would I be the only one interested in seeing a tournament where the resignation option is turned off? Being a low rated player, I can't always pick out why a position is lost, and I think it'd be really cool to see some of those endings in action as opposed to theory.

Yeah, it would be nice, but GMS, particularly in a tournament where they may have to play many games, probably don't waste energy playing out a lost game.

One suggestion in your own analysis is just run the remainder of the game through an engine.

Originally posted by Marinkatomb
No, that is simply incorrect. If that was the case then there would be no point in playing the game!

people don't play perfect so it can't be proved end of story