Go has a problem with "game tree complexity snobbery"

[cel70]

And yes I know chess has it's own problems with snobbery too! As a long term chess player I have branched out into Go in the past couple of years. I have enjoyed learning, but one thing I have noticed time and time again. When Go players talk about the complexity of Go, they always seem to delight in mentioning the fact that the game tree complexity of Go is seriously big, much bigger than chess. Numbers like 10 to the power of 360 (or even 10 to power of 700) appear time and again. The game tree complexity seems to take on some mystical property for some. Okay, it's big- really big- but in human terms these numbers are meaningless.

The number of chess games is usually considered about 10 to the power 120, or sometimes it's more conservatively pruned down to 10 to the power 40 or so. Which is a mere trillion trillion thousand billion quadrillion billion.....(.... etc... etc....) To visualize this, if everyone on Earth paired off with another person to play a chess game every day, it would take trillions of years to finally play every possible chess game. Considering that the universe is a mere 14 billion years old, you can see just how large 10 to the power 40 is. In human terms it may as well be infinite. Yet the fact that Go is even more complex than this seems to some people to mean it's a deeper game!

In my opinion Chess and Go are the two best games yet invented, but I feel that some Go players get overly excited about game tree complexity.

[RobertJasiek]

"The" go game tree is much larger but both games have trees that are sufficiently large for a player to take more than a lifetime to master the game.

[Bill Spight]

Well, what about 25x25 Grand Shogi?

Bingo!

[Jhyn]

but I feel that some Go players get overly excited about game tree complexity.

I agree with you. There's no reason to be proud about the large game tree complexity of go when this is unrelated, or only tangentially related, to what makes the game great. The main reason I was happy to see AlphaGo win the last computer-mankind match is that I hope we would lose the whole "chess is a puny game for computers, and Our Game is a noble endeavour of poetry and zen" attitude.

[RobertJasiek]

the large game tree complexity of go when this is unrelated, or only tangentially related, to what makes the game great.

No. The large game tree complexity is a requirement for the game to be great. (There are also other reasons why it is great but "unrelated" or "only tangentially related" are totally wrong.)

[Bantari]

the large game tree complexity of go when this is unrelated, or only tangentially related, to what makes the game great.

No. The large game tree complexity is a requirement for the game to be great. (There are also other reasons why it is great but "unrelated" or "only tangentially related" are totally wrong.)

No. It depends on your personal criteria.

Some people like games which are not very complex. I have friends who would rather play roulette or poker then chess or go. I know others who rather play backgammon or checkers. My niece loves snakes-and-ladders above all else. It takes all kinds.

So don't be rash in thinking that your personal criteria are the only valid ones.

[Bantari]

And yes I know chess has it's own problems with snobbery too! As a long term chess player I have branched out into Go in the past couple of years. I have enjoyed learning, but one thing I have noticed time and time again. When Go players talk about the complexity of Go, they always seem to delight in mentioning the fact that the game tree complexity of Go is seriously big, much bigger than chess. Numbers like 10 to the power of 360 (or even 10 to power of 700) appear time and again. The game tree complexity seems to take on some mystical property for some. Okay, it's big- really big- but in human terms these numbers are meaningless.

The number of chess games is usually considered about 10 to the power 120, or sometimes it's more conservatively pruned down to 10 to the power 40 or so. Which is a mere trillion trillion thousand billion quadrillion billion.....(.... etc... etc....) To visualize this, if everyone on Earth paired off with another person to play a chess game every day, it would take trillions of years to finally play every possible chess game. Considering that the universe is a mere 14 billion years old, you can see just how large 10 to the power 40 is. In human terms it may as well be infinite. Yet the fact that Go is even more complex than this seems to some people to mean it's a deeper game!

In my opinion Chess and Go are the two best games yet invented, but I feel that some Go players get overly excited about game tree complexity.

I think this is caused by the fact that, in western world, chess is a much more established game than Go.

So what do you say to a chess players when he asks you why do you not play chess instead? Go has no moving pieces, no kings and no horsies, no fast action, and no Kasparov or Fisher. The only thing a Go player can answer in defense is: "well, when it comes to complexity and game tree, Go can eat chess for breakfast."

[hyperpape]

the large game tree complexity of go when this is unrelated, or only tangentially related, to what makes the game great.

No. The large game tree complexity is a requirement for the game to be great. (There are also other reasons why it is great but "unrelated" or "only tangentially related" are totally wrong.)

The game tree must be practically large enough that humans can't exhaustively search it. But that is a bar that is quite easily met. There are few games that are widely played but fail that criterion. Tic-Tac-Toe is the only one that immediately comes to mind.

Some games that I find too sharp (play wrong in the first three moves and you have lost) still have very large game trees. It's just that a strong human can navigate that game tree in such a way as to preserve a win from that initial position.

[RobertJasiek]

I wrote "the" game, not "a" game, i.e. my requirement statement is meant only for go.

[Bantari]

I wrote "the" game, not "a" game, i.e. my requirement statement is meant only for go.

In other words, the requirement for Go to be great is that it is what it is (i.e. a highly complex game.)

No offense, but I don't feel that your statement is adding much to this (or any) discussion.

[hyperpape]

I wrote "the" game, not "a" game, i.e. my requirement statement is meant only for go.

You're making an ad hoc claim and not trying to generalize. Is it opposite day?

[RobertJasiek]

Bantari, ALA you do not read carefully what I write ("a" - not "the" - requirement), you do not understand what I write.

[Jhyn]

the large game tree complexity of go when this is unrelated, or only tangentially related, to what makes the game great.

No. The large game tree complexity is a requirement for the game to be great. (There are also other reasons why it is great but "unrelated" or "only tangentially related" are totally wrong.)

What you say is technically correct if you take my sentence out of context. We are talking about comparing go with other (2-player, no luck, combinatorial) games. To tell it in a more detailed way, if you compare go to e.g. chess, the complexity of both of these games is way above the threshold that is required to make them interesting to humans, so the relative difference in game tree complexity is not a compelling argument (as it would be compared to tic-tac-toe).

"only tangentially related" might be a bit strong since I actually think that some unique aspects of the game of go - our blind spots for "brilliant moves" and the mental processes to find them - are related with the sheer number of possible legal moves. But my point is about the belief that combinatorial complexity past a certain point is a relevant measure of the (subjective) quality of a game, or even that it is the only form of measuring its (intuitive) complexity. This is in my opinion wrong and the product of a misplaced credulity in all arguments (apparently) backed by mathematics.

[Bantari]

Bantari, ALA you do not read carefully what I write ("a" - not "the" - requirement), you do not understand what I write.

Hmm... 'a' requirement then. So let me rephrase: you say one of the requirements for go being a great game is go being the way it is. "A" requirement, yes? But you do not try to generalize and claim that as "a" requirement this also contributes to greatness of other games...

So still confused here... What exactly are you trying to contribute?
Or are you just saying your version of "me too" to what the previous poster said?

[illluck]

Could someone kindly help me understand why Robert is being questioned/mocked for stating that if Go were to be straightforward (not complex) it would cease to be a great game?