Analysis with an engine - How useful do you find it to be? • page 1/2 • General Chess Discussion • lichess.org

goose_1989

I've been trying to use the engine lines to evaluate where I made mistakes:

https://losingatchess.wordpress.com/2019/11/22/a-mad-chigorin-variation/

Do you guys use them much? And if so, how could my own analysis be improved?

As ever, any tips from players on my general play / the blog would be appreciated.

I've been trying to use the engine lines to evaluate where I made mistakes: https://losingatchess.wordpress.com/2019/11/22/a-mad-chigorin-variation/ Do you guys use them much? And if so, how could my own analysis be improved? As ever, any tips from players on my general play / the blog would be appreciated.

Onyx_Chess

#2

Analysis with an engine - How useful do you find it to be?

(added) - - - > And can you provide evidence of something that it's taught you that you can take away and apply in subsequent games?

Analysis with an engine - How useful do you find it to be? (added) - - - > And can you provide evidence of something that it's taught you that you can take away and apply in subsequent games?

goose_1989

#3

@Onyx_Chess that's a much better question, thanks.

Personally, the engine analysis I've done has persuaded me to stop playing c5 with white in the queen's gambit. Though I needed a human to explain the engine analysis to me before I took it on board. Anyone else?

@Onyx_Chess that's a much better question, thanks. Personally, the engine analysis I've done has persuaded me to stop playing c5 with white in the queen's gambit. Though I needed a human to explain the engine analysis to me before I took it on board. Anyone else?

MU4321

#4

I find them very useful and use Stockfish 10 with 20 CPU core Dual Xeon E5-2680 v2 engine. Houdini, Komodo. Shredder and
Deep Fritz are all strong too as are some other engines. I try not to spend too much time on them, playing is better!

I find them very useful and use Stockfish 10 with 20 CPU core Dual Xeon E5-2680 v2 engine. Houdini, Komodo. Shredder and Deep Fritz are all strong too as are some other engines. I try not to spend too much time on them, playing is better!

Benedictine

#5

For the vast majority of time I just use the engine to do a quick blunder check of blitz and bullet games. I hardly even need the engine for this but it speeds up the process. I do take on board more minor fluctuations out of interest but I'm more likely to act on the database percentages and the games following then. For example if I didn't know why the computer disliked ...C5 in the QG I might follow the engine analysis a bit but more useful perhaps would be to flick over 5/6 games (maybe more) from that position and look for common themes against it.

kramnikstudent1

#6

I would like to divide my answer into sections
a) I lose the game badly out of the opening: In this case I just check what I thought was the blunder against what the engine says. I then proceed to spend some time in the opening part and learn from my mistakes. There is no point in analyzing the middle game.
b) I lose a close game with no opening catastrophe (So loss in middle game and endgame): I run the engine to check the crucial points to see where I could have improved. Usually if the opponent attacked me and I failed to find the right defensive resource. Or see where I could have defended better in the ending. Again this section is analyzed to learn. I also try to find missed tactics if any by me and my opponent.
c) I win a game in the opening due to a tactic: I personally skip analyzing this
d) I win a game which was closely fought: In such cases I run the engine and the emphasis is more on the graph I see how many times the game swung to and fro and places where I had an advantage and didnt capitalize. Also I look at the way the game graph moves to determine if I managed to convert my advantage effectively. It many times happens that I think I played a very good quality game only to realize that there were simpler ways to win.

I would like to divide my answer into sections a) I lose the game badly out of the opening: In this case I just check what I thought was the blunder against what the engine says. I then proceed to spend some time in the opening part and learn from my mistakes. There is no point in analyzing the middle game. b) I lose a close game with no opening catastrophe (So loss in middle game and endgame): I run the engine to check the crucial points to see where I could have improved. Usually if the opponent attacked me and I failed to find the right defensive resource. Or see where I could have defended better in the ending. Again this section is analyzed to learn. I also try to find missed tactics if any by me and my opponent. c) I win a game in the opening due to a tactic: I personally skip analyzing this d) I win a game which was closely fought: In such cases I run the engine and the emphasis is more on the graph I see how many times the game swung to and fro and places where I had an advantage and didnt capitalize. Also I look at the way the game graph moves to determine if I managed to convert my advantage effectively. It many times happens that I think I played a very good quality game only to realize that there were simpler ways to win.

h2b2

#7

It's probably not useful to go over a game with engine analysis without understanding why a move is bad, or the strongest move, etc.

If you can understand why a move was bad or the strongest move in terms of fundamentals, tactics, strategy, etc, and put a label to it, maybe it's a bad move because it permanently weakens dark squares, or whatever, then sure. as long as you look out for those things while you're playing. which I never do.

It's probably not useful to go over a game with engine analysis without understanding why a move is bad, or the strongest move, etc. If you can understand why a move was bad or the strongest move in terms of fundamentals, tactics, strategy, etc, and put a label to it, maybe it's a bad move because it permanently weakens dark squares, or whatever, then sure. as long as you look out for those things while you're playing. which I never do.

will_is_myth

#8

I like engine analysis when I wonder what the fuck is the correct play in some specific position. In others words, when I need a concrete understaing...

For thematic attacks, common pawn-breaks, endgames I dont want reach, etc.. I preffer a strong player saying for me.

I would learn alone using engines, by testing each possibility, but that waste a huge amount of time!!! One would start engine in some critical positions and play any kind of ideas making anotations on curious and smart answers from engine. "oh, in many variations I have troubles with my a-pawn" etc.. But again, that waste much time. A human would be direct to the point, in 30 minutres iot is possible learn the main issues in a verbal way...

I like engine analysis when I wonder what the fuck is the correct play in some specific position. In others words, when I need a concrete understaing... For thematic attacks, common pawn-breaks, endgames I dont want reach, etc.. I preffer a strong player saying for me. I would learn alone using engines, by testing each possibility, but that waste a huge amount of time!!! One would start engine in some critical positions and play any kind of ideas making anotations on curious and smart answers from engine. "oh, in many variations I have troubles with my a-pawn" etc.. But again, that waste much time. A human would be direct to the point, in 30 minutres iot is possible learn the main issues in a verbal way...

CM Sarg0n

edited

#9

Sure!

My advice:

-playing whole games
-thinking about good moves and actually taking responsibility for them i.e. moving/playing
-analyzing the game with an engine

Since you have contemplated about every position you have a certain feeling for the position and you will understand the engine‘s suggestion better.

As easy as that.

Sure! My advice: -playing whole games -thinking about good moves and actually taking responsibility for them i.e. moving/playing -analyzing the game with an engine Since you have contemplated about every position you have a certain feeling for the position and you will understand the engine‘s suggestion better. As easy as that.

Kilbroney

#10

I never analyse my finished games - don't see any sense in it. The game is over, it can not be re-won. Ye can't change the past. Play the game, if you win, lose or draw, fine. Don't beat yourself up over it. You can easily get too wrapped up in thought storms, such as "..oh-what-if-I-had-played-this-or possibly-this.." it's just a board game. It's not the-be-all-and-end-all of your day. Move on. Get over it - it's just a game to while-away a few hours when you want a bit diversion and entertainment.