Go back
Can you improve by playing against computer?

Can you improve by playing against computer?

Only Chess

Vote Up
Vote Down

Originally posted by EmLasker
Can you improve by playing against computer? for example, I play several rounds of blitz chess games against computers(such as GM Lasker, Alekhine, Morphy etc) in CM Grandmaster, can you improve that way?

any comments and suggestions are appreciated.
In my opinion - yes. For example engine will punish you for your blunders and tactical mistakes better than humans and it can learn you not to make such a mistakes again.

On the other side - playing against humans may learn you how to exploit tactical mistakes of your opponent, how to refute incorrect sacrifices, etc.

1 edit
Vote Up
Vote Down

Originally posted by diskamyl
sorry GP but again, you're just imposing your style, and with employing the concepts inaccurately too. Dynamic chess is something different than setting up traps and going for cheapos whenever you get "bored" of the position, and believe it or not, modern engines are best at it. you should try playing against Rybka some time.
Hmmmm....

My conbinations are not 'cheap traps' but deeply calculated precision
crafted works of art - they do however have a slight trace of innacuracy
in them.....cheap traps is about right.

I am fully aware of what dynamic play is. It's creating an inbalance
in the position so the path is not clear. This will involve an element
of risk but as I'm playing a human this is justified.

Most players are not as good as what they think they are

1800 on the grading list,
2000 in their minds
1600 OTB.

Street Fighter has written an excellent book showing humans
collapsing in won positions because they failed to solve the OTB
problems he set them.

These are human v human games. If he had been playing a box it
would have won 99% of these games.

Perssitant box playing will destroy your 3d vision.
You are under no pressure at all v a box - how many of you take
a move back after the box has spotted an elementry blunder.

I don't resort to my complicatiions tactics when I get 'bored' I
resort to these complications tactics when ever I can.
Why?
Not because I'm good in these 'only move 'positions' but because
the majority of players I have beaten are not.

Playing a box thinking it will assist you in human v human encounters
is deluding.

Computer players....biggest feet in the jungle....drones.

Edit 1: Show me Rybka setting a trap when there is a better move
in the position. Setting the trap knowing the human player will
not see the refutation.

Vote Up
Vote Down

Originally posted by EmLasker
Can you improve by playing against computer? for example, I play several rounds of blitz chess games against computers(such as GM Lasker, Alekhine, Morphy etc) in CM Grandmaster, can you improve that way?

any comments and suggestions are appreciated.
You can improve by playing against the computer, but human opponents are a better option for you.😏

Vote Up
Vote Down

Yes you will improve by playing against a computer.

1 edit
Vote Up
Vote Down

Originally posted by greenpawn34
Hmmmm....

My conbinations are not 'cheap traps' but deeply calculated precision
crafted works of art - they do however have a slight trace of innacuracy
in them.....cheap traps is about right.

I am fully aware of what dynamic play is. It's creating an inbalance
in the position so the path is not clear. This will involve an element
of risk but a the position. Setting the trap knowing the human player will
not see the refutation.
again, you're asking for examples of traps. I insist that dynamic chess is something different.

here's an example I found in 2 minutes. Rybka 2 vs shredder, where rybka sacrifices two pawns and a piece for a serious attack:

Vote Up
Vote Down

Originally posted by diskamyl
again, you're asking for examples of traps. I insist that dynamic chess is something different.

here's an example I found in 2 minutes. Rybka 2 vs shredder, where rybka sacrifices two pawns and a piece for a serious attack:
[pgn]
[Event "15th World Computer Chess Championship"]
[Site "Amsterdam, The Netherlands"]
[Date "2007.06.18"]
[Round "11"]
...[text shortened]... 2 Ne5 32. Qe8+ Kh7 33. Ke3 b5 34. Kf4 Bb7 35. Qe7 Bd5 36. Kxe5 1-0
[/pgn]
funny. if I didn't know better I would've thought it was a game between two 1200s.

Vote Up
Vote Down

Originally posted by EmLasker
Can you improve by playing against computer? for example, I play several rounds of blitz chess games against computers(such as GM Lasker, Alekhine, Morphy etc) in CM Grandmaster, can you improve that way?

any comments and suggestions are appreciated.
Can you improve by playing exclusively against computers? Yes. However, the key point is that you will improve faster by playing both computers and humans and using the engine to aid your post-mortem analysis in your games against humans.

Vote Up
Vote Down

Originally posted by wormwood
funny. if I didn't know better I would've thought it was a game between two 1200s.
I kind of agree with that. I don't enjoy that much complications either. but it's just a good example of dynamic play (I think).

4 edits
Vote Up
Vote Down

Originally posted by diskamyl
again, you're asking for examples of traps. I insist that dynamic chess is something different.

here's an example I found in 2 minutes. Rybka 2 vs shredder, where rybka sacrifices two pawns and a piece for a serious attack:
[pgn]
[Event "15th World Computer Chess Championship"]
[Site "Amsterdam, The Netherlands"]
[Date "2007.06.18"]
[Round "11"]
2 Ne5 32. Qe8+ Kh7 33. Ke3 b5 34. Kf4 Bb7 35. Qe7 Bd5 36. Kxe5 1-0
[/pgn]
The box sacced the piece because it thought that was the best move.
It would not play a 2nd best move.

(are you really watching two comouters play each other - have to get you out more?)

I want a position where the computer plays a move knowing
it's NOT the best move and is unsound but figures/gambles
that the refutation is so unclear the human won't find it.

Tonight I'll give you human example(s)

3 edits
Vote Up
Vote Down

Originally posted by greenpawn34
The box sacced the piece because it thought that was the best move.
It would not play a 2nd best move.



I want a position where the computer plays a move knowing
it's NOT the best move and is unsound but figures/gambles
that the refutation is so unclear the human won't find it.

Tonight I'll give you human example(s)
computers play moves that they think they are inferior to their first choice only in positions that are drawn, if they have a contempt factor greater than 0, and there exists such a move that could prevent the draw and has an evaluation between 0 and contempt factor.

that means, if rybka's contempt is, say, 0.5, and the opponent threatens perpetual check, if there exists such a move, it will play the inferior move that prevents the perpetual but puts rybka behind somewhere between 0 and -0.50.

but this is not what you're looking for of course. I know computers don't set up traps. my example was for dynamic play by computers.

quote: (are you really watching two comouters play each other - have to get you out more?)

I'm not, but I can give you many reasons why I could. you should have known better (not to judge people by such interests) as a dedicated chess player.

Vote Up
Vote Down

Originally posted by wormwood
computers make unprincipled moves. because they can, due to sheer number crunching brute force power. a human can't learn from an unprincipled move, because it doesn't connect to anything we understand from before. it doesn't make sense to us in the big picture, so the best we can do is memorize it. but memorizing isn't sufficient for being able to [i]apply ...[text shortened]... s simply play differently. but who's gonna defend KNB a hundred times against you in a row?
I think you're under estimating the potential gains from being exposed to these 'unprincipled moves'. People find these tactical moves too. If you play a computer regularly, then you are exposed to deep tactics which a human opponent wouldn't always find. There's nothing like taking a few heavy spankings to ground your perception of you're own ability. That said, playing strong human opposition would be just as useful...

Vote Up
Vote Down

(are you really watching two comouters play each other - have to get you out more?)

It was a wee joke.

Relax we are talking about a game and a pocket calculator with an attitude.

To continue....

Playing and getting trashed by a box is not teaching you a thing.
Where did you go wrong?

You will have work out it for yourself and if you get that bit wrong
then the original error will go un-noticed.
Then the bad move/idea in certain positions becomes a habit.

Be very wary about playing continuously v a box.

Your chess playing becomes a habit - dry, dull and dreary.
One trap and 'Kapow' game over.

You must learn how to swindle in bad positions - at least put
up a fight.

1 edit
Vote Up
Vote Down

Originally posted by greenpawn34

I want a position where the computer plays a move knowing
it's NOT the best move and is unsound but figures/gambles
that the refutation is so unclear the human won't find it.
As we all know, it's very difficult to program that into a chess engine. Probably the chess program that came closest to that goal was the Chess System Tal II program, by Chris Whittington. The program is very difficult to find these days (programmed for Windows 95/98, I think). Unfortunately, I don't have a copy of the program, and I've never played against it.

But if you're interested in reading about it, here's a LONG thread on another forum:

http://www.talkchess.com/forum/viewtopic.php?t=14284&postdays=0&postorder=asc&topic_view=flat&start=0

Also, here's a link with a description of the CSTAL program and a pic of Chris. (Note the link to the CSTAL Home is broken.)

http://www.rebel.nl/authors.htm

1 edit
Vote Up
Vote Down

Originally posted by Mad Rook
As we all know, it's very difficult to program that into a chess engine. Probably the chess program that came closest to that goal was the Chess System Tal II program, by Chris Whittington. The program is very difficult to find these days (programmed for Windows 95/98, I think). Unfortunately, I don't have a copy of the program, and I've never played agains of Chris. (Note the link to the CSTAL Home is broken.)

http://www.rebel.nl/authors.htm
for the record, I have created a "Tal-Wannabe" personality in chessmaster that plays around 2000-2100 level, with a very sacrificial style. you can be sure your kingside is going to be bust. you just have to survive with your extra material. (I never could.)

Vote Up
Vote Down

Originally posted by EmLasker
Can you improve by playing against computer? for example, I play several rounds of blitz chess games against computers(such as GM Lasker, Alekhine, Morphy etc) in CM Grandmaster, can you improve that way?

any comments and suggestions are appreciated.
Yes, you can.I suppose you're asking if you can improve more by playing against a computer than a human?

I'd bet playing the human is better, although there are some good points to make about playing against computers. For example, they're mostly gonna punish you hard for every inaccuracy, so you'll get in habit to calculate for "backfires" since you know it will happen if you forget about it.

Blitz probably won't improve your chess until up a certain level.

The best about playing humans though is that they expose themselves to attack, and the better you are, the better you can exploit inaccuracies. A computer though, won't make any tactical mistakes, you're essentially just praciticing positional play.. which is good, in a way.