Go back
Chess Engines...Again

Chess Engines...Again

Only Chess

Small Animal

Joined
11 Jan 09
Moves
66098
Clock
19 Apr 09
Vote Up
Vote Down

I am looking for opinions on chess engines , and specifically if there is anything of value you can learn from them , or does one develop bad habits playing against an engine.

MCA
TokerSmurf

Bonnie Scotland

Joined
16 Jun 04
Moves
15683
Clock
19 Apr 09
Vote Up
Vote Down

Originally posted by dirtysniper
I am looking for opinions on chess engines , and specifically if there is anything of value you can learn from them , or does one develop bad habits playing against an engine.
i have used chessmaster to play the tricks and traps stuff - the little exercises to help with patern recognition and stuff (find mate in x number of moves, find pins etc) and i think it helped my game (a bit)

couldnt say much about playing against an engine - i dont do it often because i always get my ass kicked lol

MR

Joined
19 Jun 06
Moves
847
Clock
19 Apr 09
Vote Up
Vote Down

Originally posted by dirtysniper
I am looking for opinions on chess engines , and specifically if there is anything of value you can learn from them , or does one develop bad habits playing against an engine.
Stay away from them - They're pure evil. (Greenpawn can vouch for what I'm saying.)

OK, all kidding aside, they're good for quickly finding any tactical mistakes you made in a game. They're good for solving puzzles if they're not too deep. Some engines and/or GUIs can dumb down the playing strength so you have a fighting chance, so if you can't play a person for some reason, it's better than nothing.

greenpawn34

e4

Joined
06 May 08
Moves
43363
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by dirtysniper
I am looking for opinions on chess engines , and specifically if there is anything of value you can learn from them , or does one develop bad habits playing against an engine.
Hi DS.

I'll see if I can get '31 or Fat Lady to answer your question in your Bates Motel box.

You know what I am going to say on the subject.

I know '31 is of the same opinion as me regarding box use, he also
a lot younger, possibly the same generation as the guys who swear
it does them good.

he won't come on here and argue and with them''cos he thinks...

Well I can't tell you he thinks about them, but he thinks I'm crazy
for getting embrolied in it all.
But there again he thinks I'm crazy anyway.

So I'll prepare another post to give these tin can users another shock.
You can watch your box for a reply.

Played over one of your game per chance a few hours ago - will come
back to you on that - lots of good things to say.

How did you find the puzzles - too easy, too hard - just right?
(that bit sounds like Goldilocks tasting porridge).

I'll be back in this thread in about 30 minutes.

greenpawn34

e4

Joined
06 May 08
Moves
43363
Clock
20 Apr 09

Hello Again.

Yes in my opinion a computer will help you progress provided you use it correctly.

Using it simply to play a game.

Stick the thing on a low level have a standard set next to you,
transfer the moves from the screen to the board and study the board.

Nothing will shift me on this point, playing a computer constantly
without a full set will destroy your 3d vision.
You must learn how to roam and read a chess board with your eyes.

If you can beat it with White and Black on that level on a regular basis
take it up a level.

Using a computer to study
A smashing tool for endings.

Put in some of Capa's endings, from well annotated games,
you take Capa's side and try and implement the plans Capa used.
(Again you must use the full set method as described above).

I've lost some of Capa's famous endings doing this and learned
a whole bucket loads of endgame tricks from a computer.

What happens is the box comes up with moves that are not
considered in the game notes. You are on your own.

I can tell you this some of Capa's opponents rolled over.
If he had been playing one these exact monsters then he would
have been faced with obstacle after obstacle.
Try it yourself.

The box is a good ending tool, possibly it's best feature.
I would even go so far as to say that a good box with a standard
set of databases is better than any endgame book.

I got more from a computer regarding endings that I did from any book.

Let's take Rook Endings
Books give you all the standard ways to win from the same
standard positions - but you cannot ask a book questions.

You can ask a computer questions on the ending.

Does the same idea with work with the Rook a1 instead of c1.
The computer will show you.
What happens if the King is on c6 instead of c7 as in the book.
Again a reasonable box will show you.

Using a computer to analyse your games

This is where it fails. It cannot instruct you on how to play
a middle game, and the middle game is where games are won or lost.

It will never take the place of a good human going over your game

Yes it will show you a tactical trick you may have missed.
But it would never show you an instructive idea if that idea
happens to have a flaw in it.

This is a position I discussed in another thread. Black to play.



One human move here is 1...Nb4 going after that Bishop on d3.
White now wins with 2.Nh5. A nice idea that 2.Nh5.
If you have not seen it before then there is a new string to your bow.

But 1...Nb4 is unsound.
You can enter that position into a box and it won't even show 1.Nb4
in it's top 10 choices.

A plausible human move hiding a nice instructive idea and if you
used a box to analyse that position you won't even know it was there.

Be very wary of what analyse the computer gives you.
Try some ideas of your own and see what the computer thinks of those.

You will have to poke the computer into finding mates, tricks and traps
in positions it did not analyse because the play preceeding it
was not what it considered the best moves.

I recently followed a game CHESS where the annotator had to
force Fritz to look at variation that was played in the game
but it did not rate it all until suddenly 5 or 6 moves later
it decided it was the best move.

There is more about the down side of a computer looking at
your games in the next section.

Playing a computer constantly will seriously damge your chess

It does not know what a human move is.
It does not play cheap shots/swindles or traps.
It does not leave a piece hanging - never.

Humans do set traps, create swindling chances and
leave pieces hanging all the time.
playing a box constantly and you forget what it's like to
play a human. You fail to spot the human errors because you
are so use to playing aginst blunder free chess.

A box has no King fear.
In the middle game a box will gladly grab a pawn or piece
and merrily send his King for a walk knowing it cannot get mated.

Humans will suck out a King hoping there is a mate there whilst
defenders will rarely send there Kings for walks fearing there is a mate there.

Given a choice of moves the computer will never pick a move based
on the fact that it present a human with a major problem to solve.
It does not know what positions a human will find difficult to handle.

It will always play, or suggest you played if analysing your game,
what it considers is the best move.

And every experinced OTB player on here will tell you.
Sometimes the best move is NOT the best move to play.

And it is those 'sometimes' positions where the box falls down.
Why? Because it's a machine.

b
Enigma

Seattle

Joined
03 Sep 06
Moves
3298
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by dirtysniper
I am looking for opinions on chess engines , and specifically if there is anything of value you can learn from them , or does one develop bad habits playing against an engine.
Chess engines can be helpful as training partners, especially if they have education programs on tactics, endgames etc. If your engine has a hint feature, you can go over your finished games with the engine and use the hint key to suggest stronger moves. Engines however are not as necessary as many people think. You can get the same information out of a few well selected books, and Red Hot Pawn can provide all the compitition you'll ever need. 😏

DF
Lord of all beasts

searching for truth

Joined
06 Jun 06
Moves
30390
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by greenpawn34
Hello Again.

Yes in my opinion a computer will help you progress provided you use it correctly.

[b]Using it simply to play a game.


Stick the thing on a low level have a standard set next to you,
transfer the moves from the screen to the board and study the board.

Nothing will shift me on this point, playing a computer constantly
without ...[text shortened]... ose 'sometimes' positions where the box falls down.
Why? Because it's a machine.[/b]
Good post. I agree 101% with everything you say here.

This is perhaps most relevent when you are "losing" as the best chance to get back into the game is to take chances to create counter play. This often involves playing an "inferior" move taht allows your opponent a chance to screw up whereas an engine would not play this simply because it "loses" quicker than the "best" move.

In other words once an engine is down it would simply lose because it cannot see those moves that complicate the position and create chances.

C
Strategos

Ancient Greece

Joined
12 Apr 09
Moves
2375
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by Dragon Fire
Good post. I agree 101% with everything you say here.

This is perhaps most relevent when you are "losing" as the best chance to get back into the game is to take chances to create counter play. This often involves playing an "inferior" move taht allows your opponent a chance to screw up whereas an engine would not play this simply because it "loses" q ...[text shortened]... lose because it cannot see those moves that complicate the position and create chances.
Advice for "once an engine is down it would simply lose because it cannot see those moves that complicate the position and create chances" guys - try to play against strong engine, starting from position with decisive advantage (extra pawn for example) for you.

DF
Lord of all beasts

searching for truth

Joined
06 Jun 06
Moves
30390
Clock
20 Apr 09
2 edits
Vote Up
Vote Down

Originally posted by Cimon
Advice for "once an engine is down it would simply lose because it cannot see those moves that complicate the position and create chances" guys - try to play against strong engine, starting from position with decisive advantage (extra pawn for example) for you.
You will lose of course!

Please don't misinterpret the key point here. The key point is that in a "lost" position an engine will not play an "inferior" move that creates counterplay but will simply play the "best" move so, all other things being equal, it will lose.

Obviously an engine in an inferior position playing most people here will simply win but that has nothing to do with creating counterplay. It will win because its opponents are, relative it it, useless.

Set that position up with an engine vs a 2800 GM or 2 2800 GMs then you will see what I mean. The engine will play the "best" moves and lose. The human GM will try and find "inferior" moves that create counterplay. Simply really! That is one of the key differences between human play and engine play!

V

Joined
21 Sep 05
Moves
27507
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by greenpawn34
Interesting post. Any example middlegames where the human doesn't know what to do, and the engine is of no help?

I understand your point about computers not being able to make "human" suggestions. e.g. cheapos in a lost position. But let's face it, most of the time we are interested in ideas that don't have flaws, with the interesting flawed examples being in the minority.

A plausible human move hiding a nice instructive idea and if you
used a box to analyse that position you won't even know it was there.


For the Nh5 idea, try pressing 'X' in the Fritz GUI to see the threats in a position. I agree that Nh5 is not highlighted prior to doing this. Or have the "Explain All Moves" pane open and you'll see it littered with comments such as "Black doesn't see the mate".

MR

Joined
19 Jun 06
Moves
847
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by greenpawn34
Hello Again.

Yes in my opinion a computer will help you progress provided you use it correctly.

Using it simply to play a game.

Stick the thing on a low level have a standard set next to you,
transfer the moves from the screen to the board and study the board.

Nothing will shift me on this point, playing a computer constantly
without a ...[text shortened]... s those 'sometimes' positions where the box falls down.
Why? Because it's a machine.
Hi GP. Good post, I enjoyed reading it. But I do have a minor quibble.

"It does not leave a piece hanging - never."

For some engines, not true at all. On the lower levels, some engines have a built-in blunder factor that makes them occasionally hang a piece. Heck, the Hamsters engine even has a dedicted blunder setting in its configuration options!

greenpawn34

e4

Joined
06 May 08
Moves
43363
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by Mad Rook
Hi GP. Good post, I enjoyed reading it. But I do have a minor quibble.

"It does not leave a piece hanging - never."

For some engines, not true at all. On the lower levels, some engines have a built-in blunder factor that makes them occasionally hang a piece. Heck, the Hamsters engine even has a dedicted blunder setting in its configuration options!
Hi - the early ones did as well.

I'm talking about the leve you use to play it to get a reasonable game.

C
Strategos

Ancient Greece

Joined
12 Apr 09
Moves
2375
Clock
20 Apr 09
3 edits
Vote Up
Vote Down

Originally posted by Dragon Fire
You will lose of course!

Please don't misinterpret the key point here. The key point is that in a "lost" position an engine will not play an "inferior" move that creates counterplay but will simply play the "best" move so, all other things being equal, it will lose.

Obviously an engine in an inferior position playing most people here will simply wi Simply really! That is one of the key differences between human play and engine play!
I agree that most of us will lose. Because players of our level have many things to learn from engines. That`s the reason why claims like "Playing a computer constantly will seriously damage your chess" seems to be absurd.

Talking about defence - active defence is not always the best way to survive. After "trying and finding "inferior" moves that create counterplay" many players have lost very fast while many have survived after making patient defence. Take a look on games played by good defenders like Ratmir Kholmov and Ulf Andersson.

DF
Lord of all beasts

searching for truth

Joined
06 Jun 06
Moves
30390
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by Cimon
I agree that most of us will lose. Because players of our level have many things to learn from engines. That`s the reason why claims like "Playing a computer constantly will seriously damage your chess" seems to be absurd.

Talking about defence - active defence is not always the best way to survive. After "trying and finding "inferior" moves that create cou ...[text shortened]... ence. Take a look on games played by good defenders like Ratmir Kholmov and Ulf Andersson.
I totally agree.

What I am saying (which is what Greenpawn states) is that faced with a almost certain loss a human player would seek out "inferior" moves that give his opponent the maximum opportunity to make a mistake rather than simply make the "best" moves.

Sometimes as a result of this the human pulls a "cheapo" and wins. Other times he loses faster than would have been the case but the point is the game was lost anyway so any moves that create chances, even desparate ones are preferable to nothing at all.

greenpawn34

e4

Joined
06 May 08
Moves
43363
Clock
20 Apr 09
Vote Up
Vote Down

Originally posted by Cimon
I agree that most of us will lose. Because players of our level have many things to learn from engines. That`s the reason why claims like "Playing a computer constantly will seriously damage your chess" seems to be absurd.

Talking about defence - active defence is not always the best way to survive. After "trying and finding "inferior" moves that create cou ...[text shortened]... ence. Take a look on games played by good defenders like Ratmir Kholmov and Ulf Andersson.
You will learn more from humans that you ever will from a box.

"Playing a computer constantly will seriously damage your chess" seems to be absurd."

A lot of good players agree with this absurd notion.
Playing a human at chess if totally diffenrent from playing a box.

There will be things you pick up from playing box but you will
learn how to play Chess by playing humans.

You get beat by a computer - big deal. What have you really learned?
Will you worry about it. Did the lose smack you in the face?

You lose to a human. These loses stick, the mistakes you make v
human can haunt your mind for years and years.
Believe me.

Kholmov and Andersson learned their trade v humans.
And that is what we are talking about here - learning how to play Chess.

I've given the pro's and cons about box use.
Used properly (see above post) I really do believe a computer has
a lot to offer.
But you are not going to get all the answers.

One day you are going to have to sit opposite a human on a full
sized set who will use all the tricks posted in another thread
'Shrewd OTB tactics'. and more that are not mentioned.

They will sit poker-faced and play cunning traps, unsound tricks
and outright blunders. If you have never seen them before.......

Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.