Chess Engines

Originally posted by tvochess
Let's pull this piece out of context and into the spotlight..

Originally posted by LittleDonkey
[b]Personally I think reading (...) is a far better way of sharpening one's tactical skills.

On the contrary, reading helps very little to sharpen tactical skills. You have to practice, even if against a computer! In fact, it ...[text shortened]... ant to become skilled in tricks and traps. Then you hope your opponent doesn't see what you see.[/b]

You edited out the important bit! You admit that you have to be aware of the possible attacks and defences, and I agree with you. An engine might show you the best move but it only shows you the best move in a given position, it doesn't tell you the thinking behind it. Greenpawn shows you the position, the mating patterns, the very human thinking behind the moves... I reach a totally different conclusion to you, I think it does train you to find those tactics yourself.

It raises an interesting point on learning. Greenpawn v Engine. Who is the better teacher?

Originally posted by LittleDonkey
You edited out the important bit! You admit that you have to be aware of the possible attacks and defences, and I agree with you. An engine might show you the best move but it only shows you the best move in a given position, it doesn't tell you the thinking behind it. Greenpawn shows you the position, the mating patterns, the very human thinking behi ...[text shortened]... f.

It raises an interesting point on learning. Greenpawn v Engine. Who is the better teacher?

I'm sure a human would be a better teacher, however a tactically superior engine can be useful in analysis of completed games, and in analysis of middle and endgame positions. We don't all have access to a human teacher, so must make use of the tools we have.

Originally posted by tvochess
Let's pull this piece out of context and into the spotlight..

Originally posted by LittleDonkey
[b]Personally I think reading (...) is a far better way of sharpening one's tactical skills.

On the contrary, reading helps very little to sharpen tactical skills. You have to practice, even if against a computer! In fact, it ...[text shortened]... ant to become skilled in tricks and traps. Then you hope your opponent doesn't see what you see.[/b]

Yeah, but...only the reader completely devoid of curiosity looks at an article on chess and does NOT play the moves / variations out on the board.

Once moves are played and questions are asked, then practice has occurred.

I especially like "on the contrary" in your post. How can you be so sure that the reading method won't work for anyone?!!?

Better teacher for who?

Would a trout breathe better under water or out of water?

Does a rabbit breathe better under water or out of water?

It depends

Engines are very easy to found and install, today free engines like stockfish are better than any human.
.
Lucas chess, for example, has a package of 30 different engines.
.
So the temptation to use one is very high.
.
Even GMs has been cheating because the iphone chess app is stronger than Magnus

I see I've been mentioned.

The engine will show you tactics that have been missed but
only if the missed tactic gave the computer the highest evaluation.

It will never show or play a flawed combination even if the combo is
20 moves deep requiring the human to play some extra ordinary difficult
and only moves. It will see it but keep it to itself. I often wonder what wonderful
things it sees that would in all probability work v a human.

One day I'm sure they will figure out a way to tell a chess computer what is a
difficult position for a human to fathom and add a Human Difficulty rating into
it's evaluation. AT the moment these things don't even know they are playing a game.

They will never set traps., never, even if it has a 99% chance of success.
They have no hope, they cannot anticipate a blunder, when they are lost they sulk.

Because of this flaw there are thousands of games humans have won that a computer
would have lost because the concept of swindling a win (a fine human art form) is beyond them.

I'm sure we have all set a trap in a lost position that worked. Put the same position
into a computer it will not play it (unless for a completely different reason it considers
it is the best move.) The blog has loads of examples of 1400 players winning games
even the best computer on the market would have lost. The human set a catch a human trap.

----

Karjakin when interviewed about his missed chance in Game 10 where Carlsen blundered.
Carlsen went on to win that critical game, tie the match and retained the world title.

Karjakin was not expecting the blunder, he sais:

""You might say I fell victim to my own respect for Magnus."

Anand said something like that when he missed an excellent chance in the last match after a Carlsen blunder.

My response.

'Too much computer in there Sergey and not enough human. You never missed it
because of respect for Magnus, you missed it because computers do not make those
type of blunders and you are slipping when it comes to spotting a human inexact move.

Switch off the box, let your seconds fanny about with that, you study without one, you will be playing without one.'

----

The Blog.

I do not assume to be the teacher on RHP. I enjoy spotting shots and seeing
them come off. I enjoy trying to figure what went wrong and seeing things missed.
I enjoy writing about chess and having fun. Hopefully there is instructive value in there
and somebody will use an idea or two. But the main object is very selfish.

I enjoy doing it. It's not a chore or something I have to do. I really do enjoy it.
(the world championship coverage was a bit of a rush and a hassle - I suddenly
found myself with deadline, I won't be doing that experiment again .)


----

I agree with Knight Stalker,

A lot of the accusations are false, I do get a few PM's telling me so and so is cheating.
I look at the game and see the lad was winning after some bad play by his opponent,
(definitely no computer there) they messed it up, lost. and fired off a heat of the moment PM.

Originally posted by greenpawn34
I see I've been mentioned.

The engine will show you tactics that have been missed but
only if the missed tactic gave the computer the highest evaluation.

It will never show or play a flawed combination even if the combo is
20 moves deep requiring the human to play some extra ordinary difficult
and only moves. It will see it but keep it to i ...[text shortened]...
(definitely no computer there) they messed it up, lost. and fired off a heat of the moment PM.

The blog has taught me a thing or two Jeff, although I still sometimes forget to "check all checks" though not as often as I used to.
Well done and keep it up.

Originally posted by LittleDonkey
You edited out the important bit! You admit that you have to be aware of the possible attacks and defences, and I agree with you. An engine might show you the best move but it only shows you the best move in a given position, it doesn't tell you the thinking behind it. Greenpawn shows you the position, the mating patterns, the very human thinking behi ...[text shortened]... f.

It raises an interesting point on learning. Greenpawn v Engine. Who is the better teacher?

My statement is that reading does very little about one's tactical awareness. Playing in any form that forces you to think, is going to be more valuable.

I agree, a chess engine is not going to tell why something is good or bad. But I didn't say an engine would be a good teacher. It's just a sparring partner to practice with, if you need one. It'll definitely punish you when you make tactical mistakes. So it will be showing you ideas. But you'll have to go back and work on a position yourself if you want to understand it better.

Greenpawn's blogs are a valuable resource, for the reasons you mention. For me personally, I learn a little chess history, some famous games, common blunders and some tricks and trap ideas. And above all maybe, the blog keeps chess fun, which is sometimes necessary after all the hard work for making moves in my games. But I don't think the blog does much for the tactical skills of his readers. He says himself about every time that players on RHP keep falling for the same traps and blunders that he constantly mentions!

Originally posted by BigDoggProblem
Yeah, but...only the reader completely devoid of curiosity looks at an article on chess and does NOT play the moves / variations out on the board.

Once moves are played and questions are asked, then practice has occurred.

I especially like "on the contrary" in your post. How can you be so sure that the reading method won't work for anyone?!!?

Well, and the readers with limited available time, no immediate access to a chess board, are mentally exhausted from a long day at work... which are situations I'm often in. Whenever possible, I try to study the variations in my head, but rarely can spend more time/effort.

"Once moves are played and questions are asked, then practice has occurred."
100% agree.

On you last question, maybe it works for some people. But learning rarely comes from passive methods, in chess and everything else.

It seems there are actual two issues in play.

1. Passive learning vs. Active
2. Human vs. engine teaching

On issue 1, much depends upon the disposition of the learner. I see no reason, for starters, why reading must be a passive activity. A reader can stop, ask questions of themself and the text, refrain from moving on until the point is resolved. In the case of GP's blog, one can send a PM directly to the blogger and get clarification! So, there is no excuse for a reader's passivity.

Equally, with engines, I see no reason why the student must sit idly by, looking only at the engine's top suggestions. A curious analyst almost cannot help but ask, "hey, why didn't Ne4 show up!?" when that move looks promising to them. It is ridiculously easy to play that move on the board and force the engine to show them how it would respond. Or, simply take a few moments to figure it out for themselves, then continue with the analysis.

Perhaps the hard truth is that people want to improve in the most rapid way possible with the least amount of work. But there is no avoiding the work. The passivity is usually the fault of the student.

I'll write some about issue 2 later.

Big Dog,

I can only agree.

Our earlier misunderstanding is probably due to the interpretation of what 'reading' means.

My point throughout the discussion is the same, that one should be actively involved in order to learn.

Originally posted by tvochess
My statement is that reading does very little about one's tactical awareness. Playing in any form that forces you to think, is going to be more valuable.

I agree, a chess engine is not going to tell why something is good or bad. But I didn't say an engine would be a good teacher. It's just a sparring partner to practice with, if you need one. It'll defi ...[text shortened]... me that players on RHP keep falling for the same traps and blunders that he constantly mentions!

I am somewhat inclined to disagree with the first sentence, as I am a counterexample. The second sentence is probably true more often than not, but I believe it depends on the person.

Most of the basic tactics I know either came from reading about them in a book, seeing the tactic in a game, or having it played against me.

In my particular case, I tend to recall tactics from games I have played through, such as Petrosian's queen sac against Pachman in Bled 1961 or Vasiukov's sac on d5 against Uhlmann in 1962 ( I forgot where).

I'm not a big fan of combo books or computer tactics programs, but I love playing through annotated chess games, which are the poetry and prose of our game.

I suppose I would get better if I altered my method of study, but I mostly play for fun, so my work on the game is geared in that context.

In any event, I sometimes win games at my club or in tournaments simply because I saw the tactical idea in a game I read, and recalled it for the game I was playing. It happens.

I do not use bots during my games, but I do afterwards for games I loose, always trying to find where I might gone wrong.