At the start I must admit that I am not the brightest chess player, so this just might be a stupid question but: Calculated Chess moves have been determined to be infinite, as I am given to understand. But of what use is this knowledge to the game, or player when the guestimate (this can probably be calculated) is half or even far less than half would constitute legitimate moves or moves that make no sense at all (a move because it can be moved to "x" without reason. The question that should be quired is, "how many legitimate moves are there in chess that lead to a legitimate conclusion 'win, draw, or forfeit'?" Who knows that actual legitimate moves that can be made which evolves into a legitimate conclusion?
5
-
Related: chessprogramming.wikispaces.com/Type+B+Strategy– Jossie CalderonJun 8, 2017 at 18:36
-
Thank you for that lead. I missed in clarity because I intended it to refer to human subjects playing another human subject, that is why I included sentence two. The lead was most interesting, especially for machine chess and the skills required to program said.– Glock27Jun 8, 2017 at 18:40
-
I don't understand what you are asking. The number of possible moves at any point depends on the position of the pieces on the board. At the start of the game it is 20 (16 pawn and 4 knight moves) and iit s typically around 30 possible moves in middle game positions. Any chess game ends in win or draw or forfeit, so I don't understand your requirement ("which evolves into a legitimate conclusion", etc).– user1583209Jun 8, 2017 at 19:10
-
Actually, each pawn has 2 moves, so that is 16 pawns x 2 = 32, + each knight has 2 moves (a/c file and h/f file) = 8 total of 40.– PriyomeJun 8, 2017 at 19:27
-
@Priyome: I meant to say the number of moves for either side (black and white) at the start of the game is 20. Of course the sum of white's and black's first move options is 20+20=40 which is what you calculate.– user1583209Jun 8, 2017 at 21:14
Add a comment
|
2 Answers
From Shannon's Programming a Computer for Playing Chess:
From these remarks it appears that to improve the speed and strength of play the machine must:
- Examine forceful variations out as far as possible and evaluate only at reasonable positions, where some quasi-stability has been established.
- Select the variations to be explored by some process so that the machine does not waste its time in totally pointless variations.
The first can be done by evaluating all checks, captures, and threats - a true and tested method to avoiding blunders.
But notice "some process" in the second bullet point has to be set by you. There's an easy way to choose this, and it's also noted in Vukovic's Art of Attack:
If the threat move is the same on two consecutive plies (with one ply in between them to account for the side to move), the threat move is a serious one and the search is extended.
Threat moves are moves that, if the other side were to (illegally) 'pass' on this move, they would be mated, material loss would occur, or some other favorable advantage would occur.
If I attacked your queen with my knight, and your queen moves to some spot but is still attacked by my knight, that is a serious threat move. Then we check the other possible squares that the queen can move to, and if any resulting position is unfavorable, then the position of my knight is superior and considerable for my next move.
So what else?
If a side makes a move and it leads to a mating attack in one variation without compromising your position in all variations (one of the variations is a serious threat), it is a good candidate for your next move.
We can also turn this around.
If we attack the king and the defending side has to compromise his position (by making his surrounding squares weak, for example), that is a considerable candidate move. Because it's more forcing, this is of a higher category.
Of the highest category is attacks on the enemy king with certainty of mate. Huge compromises like a queen sacrifice would have to occur to stave off this attack.
-
Jossie, an interesting point you make here and I can see how this would benefit a player, if he or she had the knowledge and skills to operate such a system. However, in the moment of play I cannot see how an opponent would agree to such analysis. This would have to occur at a break and it would be of limited use would it not. Please note I am trying to see the relationship of these staggering numbers to the actual on going game. As an amusement I think it is fine but otherwise how is it beneficial. You might have addressed this but I must have missed it as I am not that computer savy.– Glock27Jun 10, 2017 at 0:56
-
@Glock27 You can rule out all of the uninteresting variations by looking at the advice after "So what else?". Anyways, Bobby Fischer actually said “People think there are all these options, but there is only one right move." You have to play with that conviction. Jun 10, 2017 at 4:50
-
@Glock27 Yes, there can be four or five positionally sound moves - but there's only one plan. The move that can accomplish that plan flexibly is the correct move. Jun 10, 2017 at 4:55
-
"Ypu have to play with conviction". This is probably the most accurate statement relative to the "Possible Chess Moves Calculated". Certainly there are other answers that are excellent. But I think playing with confidence or conviction just may be the most significant even is one were to loose every game. If you don't loose how do you learn. Is it not accurate to say that these calculated moves role plays a role in the development of Analysis engines? Can an analysis engine actually give you the best move?– Glock27Jun 11, 2017 at 18:28
-
@Glock27 Yes. Turn everything I said around, and you'll see the engine plays non-losing lines. Then the human could make a mistake and take advantage of it. That's the difficulty in beating super engines like Stockfish: They appear to make no mistakes. Jun 12, 2017 at 0:54
Granted that there might be an infinitesimal number of possible moves in a chess game, in reality only a few really make sense on any given turn, so the vast majority can be excluded. Nevertheless, the tree of the total number of reasonable moves in a game would probably still be beyond calculation, but I can't think of any good reason to know what that possible number might be anyway.
-
Yea, I agree, but I wonder or ask to what value this actually has for anyone who plays the game. I guess it is a curiosity more than anything else to know the overwhelming number of moves, but is it not profitable moves we are looking for?– Glock27Jun 10, 2017 at 0:50