18-07-2013, 01:24 AM
Hi Andy
"But are we more interested in fun Chess or good Chess?"
Practical Chess.
We are not talking about the best players in the world.
We are talking about the under 2000 players who are sold these things on the pretence
it will give them a better understanding of the game.
They do more harm than good.
They leave huge gaps in a players development.
They shoot a players 3d vision to hell.
The player starts to fail recognise a human blunder.
They have no memory hook like a written note in a book can give.
"Computers could be programmed to look for cheapos like that too..."
Could they? I don't think so, not yet anyway.
If they could be made to play more like a human then could be more useful.
Using them to analyse what opening you may play is totally pointless.
Take a position we know, switch off the computers book then input this position.
[pos]r1bqk1nr/pppp1ppp/2n5/2b1p3/2B1P3/5N2/PPPP1PPP/RNBQK2R w KQkq - 0 4[/pos]
It won't play b4 as White, it won't even suggest it.
Yet we both know how tricky 4.b4 can be and it's been a relatively succesful human move for 170 years.
So if chummy is looking at a fresh position and hidden within is a shot only half as tricky as 4.b4
it won't suggest it. The human under 2000 won't suspect nothing till his human opponent plays it OTB.
Like all good Scottish Juniors I bet you can mate with a Bishop & Knight.
Good. (you will most likely get to use this once every 40 years - I'm still waiting my turn.)
Do it v a computer and it's a piece of cake.
Do it v a wriggling human and you will still do it, but it won't be as easy.
Humans very often in all aspects of the game do not play the best moves, they play the move
that gives them the best practical chance.
A computer does not understand what that is.
"But are we more interested in fun Chess or good Chess?"
Practical Chess.
We are not talking about the best players in the world.
We are talking about the under 2000 players who are sold these things on the pretence
it will give them a better understanding of the game.
They do more harm than good.
They leave huge gaps in a players development.
They shoot a players 3d vision to hell.
The player starts to fail recognise a human blunder.
They have no memory hook like a written note in a book can give.
"Computers could be programmed to look for cheapos like that too..."
Could they? I don't think so, not yet anyway.
If they could be made to play more like a human then could be more useful.
Using them to analyse what opening you may play is totally pointless.
Take a position we know, switch off the computers book then input this position.
[pos]r1bqk1nr/pppp1ppp/2n5/2b1p3/2B1P3/5N2/PPPP1PPP/RNBQK2R w KQkq - 0 4[/pos]
It won't play b4 as White, it won't even suggest it.
Yet we both know how tricky 4.b4 can be and it's been a relatively succesful human move for 170 years.
So if chummy is looking at a fresh position and hidden within is a shot only half as tricky as 4.b4
it won't suggest it. The human under 2000 won't suspect nothing till his human opponent plays it OTB.
Like all good Scottish Juniors I bet you can mate with a Bishop & Knight.
Good. (you will most likely get to use this once every 40 years - I'm still waiting my turn.)
Do it v a computer and it's a piece of cake.
Do it v a wriggling human and you will still do it, but it won't be as easy.
Humans very often in all aspects of the game do not play the best moves, they play the move
that gives them the best practical chance.
A computer does not understand what that is.