English Chess Forum

I was looking through some other chess forums recently (yes, such heresy, I know..) and found the subject of engines "solving" chess to be a recurring theme.

It made me wonder if they did reach a stage where it was possible to analyse the entire game to completion, would any further research into the field now be pointless as they could make no more progress, and perhaps finally kill off the industry that has plagued our game for two decades.

Physically impossible to solve the game 'properly'.

They've already basically reached a point where further progress is somewhat pointless - even the free engines are comically strong.

MartinCarpenter wrote:Physically impossible to solve the game 'properly'.

What could happen, as I think it has in some forms of Draughts is that opening theory, perhaps from apparently forced simplifications, meets up with n man tablebases where the outcome is known for certain.

The other approach which worked for Go, but perhaps not chess, is for the engines to run an immense number of simulated games and use this to recommend moves rather than evaluation functions.

They are solving chess backwards via Table Bases.

I think they are now up to 7. The computer knows and can
instantly display any result from any 7 piece position.

Geoff Chandler wrote:They are solving chess backwards via Table Bases.

I think they are now up to 7. The computer knows and can
instantly display any result from any 7 piece position.

That's 100000000000000 positions covered

only 9999999999999999999999999999999999999999999999999900000000000000

to go

IMHO,

Tablebases, game databases and chess engines are wonderful tools.

I guess there are plague like consequences, like cheating and the fear of cheating, with chess engines. However they have enriched the understanding of many players and the educational benefit of chess for children.

Long live chess engines1

They are not good at explaining. As for the consequences there is nothing better than science fiction. 2001: a space odyssey is such a classic because of what happens when you try to turn the computer off. Made in 1968.

For all practical purposes, chess has been solved. Computers beat the best of us quite easily.

Chess began as a human invention, utterly pointless to start with; and utterly pointless at every point thereafter. It's a game. We find it pleasant, intriguing, frustrating, complex, demanding. Hence, it's a good game. We find it vastly difficult whereas computers find it fairly trivial. So what! Everytime we open our eyes each morning, our brain instantly solves several billion times more problems than any chess engine ever will. So live fully. Play chess from time to time.

Geoff Chandler wrote: I think they are now up to 7. The computer knows and can
instantly display any result from any 7 piece position.

Is that true? I thought it was still only some of them, but I my not be well informed.

It is true. I have them as an app on my phone

For those of you who want to give the matter further serious thought, this Wiki on Moravec's Paradox might be a starting point. In a nutshell, Moravec's Paradox states that what we find easy, computers find hard; and vice versa

A no less serious, but more literary 10-minute point of view from Howard Jacobson with a breakfast club (if you listen carefully you'll also hear hon. mention of the Athenaeum) tacked on -

http://www.bbc.co.uk/programmes/b08ns2m6

More on clubbing can be found at -

http://www.standard.co.uk/lifestyle/lon ... 14871.html

NickFaulks wrote:

Geoff Chandler wrote: I think they are now up to 7. The computer knows and can
instantly display any result from any 7 piece position.

Is that true? I thought it was still only some of them, but I my not be well informed.

They did not do positions with six men versus bare king. Little point: either immediate stalemate or a trivial win.

So where do you find them?

NickFaulks wrote:So where do you find them?

https://play.google.com/store/apps/deta ... omonosovtb