Analysis program woes

Originally posted by Regiscyde
OP here. I suppose the novice's lament is that, to coin a phrase, "you don't know what you don't know." I was looking to a program to enhance my post-game analysis in order to assist my understanding of my losses better. I understand it may not be as effective a method as I wish it'd be. Without the luxury of being able to join a club or obtain professiona ...[text shortened]... analyzer turn? Should I not try to analyze my own faults because of my own ignorance?

I bought every chess program I could get my hands on. I bought an automated chess board with 'analysis'. What you seek still doesn't exist.

It /could/ exist, but nobody has made it. People programming chess engines use the brute force method because it's easier than trying to program actual chess principles. Sounds like an opportunity.

Originally posted by Regiscyde
OP here. I suppose the novice's lament is that, to coin a phrase, "you don't know what you don't know." I was looking to a program to enhance my post-game analysis in order to assist my understanding of my losses better. I understand it may not be as effective a method as I wish it'd be. Without the luxury of being able to join a club or obtain professiona ...[text shortened]... analyzer turn? Should I not try to analyze my own faults because of my own ignorance?

am in similar situation as you...

but, due to lazyness and lack of time, i usually dont analyze my own games (i can here the outcry from the community - i just keep the games for the future analysis 😉 ).

the most helpful for me:
reading the blog of greenpawn
seeing some riddles as from swiss gambit (mate with enpassant, etc.)
following the tata steel chess with commentary (half i dont understand, but some points i got; it was live, so many big names, and small names and fun to see what happens)
reading weird statistics about chess (the longest game played, ... )
solving the newspapers chess puzzle
reading this forums discussions and annotations of games from time to time
reading funny anecdotes about chess
browsing the opening databases of standard openings

this is about it. the list has no particular order, but the 'analysis by box' was less helpful than any of the above points...

Originally posted by Regiscyde
OP here. I suppose the novice's lament is that, to coin a phrase, "you don't know what you don't know." I was looking to a program to enhance my post-game analysis in order to assist my understanding of my losses better. I understand it may not be as effective a method as I wish it'd be. Without the luxury of being able to join a club or obtain professiona analyzer turn? Should I not try to analyze my own faults because of my own ignorance?

in two words: master games.

it's very instructional to follow masters play, especially in real time as you're forced to wait for the moves for as long as it takes. go onto ICC, playchess, FICS (free), and watch the relayed tournaments online.

you'll run time and time again into moves which leave you scratching your head: "why didn't the master go there? why didn't he make the move that seems obviously the best to my amateur eye?". you even plug the move into an engine, which (sometimes) says "yes, it's a perfectly good move." and yet the master didn't go there. he won't snatch that free pawn, exchange, or sometimes even a full piece. he won't usually compromise his position even if it's "the best move", because he knows it's likely gonna bite him right back before long.

over time you'll realize that the good players never go there, in certain types of positions. only us amateurs do. and slowly you'll develop an understanding and intuition for WHY and WHEN that happens. the feel for danger, which is the result of years or decades of experience.


if you watch masters blitz, it'll be even more apparent, as they'll be playing almost exclusively based on general principles, experience, rules of thumb, and pure intuition based on those two. with little to no calculation or conscious strategic scheming.

as a rule of thumb: whatever you do, every time you end up scratching your head, and especially when you get to the painfull part, headache and nausea, that's when the real progress happens. no pain no gain.

there's no discomfort watching an engine spit out numbers. no progress either. it's as ineffective to your improvement as watching sports on telly is to your conditioning.

Originally posted by wormwood
...he won't snatch that free pawn, exchange, or sometimes even a full piece.

heh, when I wrote that, I thought "maybe I'm going a bit far with that full piece". then I checked back on the amber games on ICC I have open, and GM yermolinsky is speculating about the anand vs gelfand game that "...but then I have the knight on a6 hanging and need to somehow save it." -two second pause, then yermo continues: "nah, I don't wanna save it." and sacs the piece for two pawns, displaced queen and a strong pawn centre. 😀 he then sacs another rook, and it isn't really a mating attack. just a very strong positional compensation.

not that it was perfectly sound or anything, but it was sound enough to catch his eye first. simply based on his gut feeling.

and yermo isn't like larry christiansen or nakamura, he doesn't go around wildly saccing material. he's more the solid old school russian type, no wild stuff. yet this was what the positions implied to him, based on decades of experience.

Originally posted by wormwood
as a rule of thumb: whatever you do, every time you end up scratching your head, and especially when you get to the painfull part, headache and nausea, that's when the real progress happens. no pain no gain.

there's no discomfort watching an engine spit out numbers. no progress either. it's as ineffective to your improvement as watching sports on telly is to your conditioning.

Yes it has a certain hypnotic effect, like staring at a washing machine, the gentle hum of the roller drum, its easy to see why Steve Reich got into minimalism. Only recently I downloaded Arena and set up an engine tournament just to watch it. It has a soothing effect upon me, cant say why. I think its the comfort of watching an engine being beaten, after having suffered so many defeats in my earlier learning period when i played exclusively against the pc.

Was it a Colle thematic engine tournament?

tee hee.

Originally posted by nimzo5
Was it a Colle thematic engine tournament?

tee hee.

Lol, no actually it was the English attack for the Sicilian Najdorf thematic tournament, where rather interestingly almost all engines played the Najdorf grob, that is ..Ng4!? This is a really great idea against the Najdorf English attack if you have ever faced it, for it stops all of whites plans of f3, 0-0-0, g4, h4-h5 against the kingside. The problem is that it leaves ones own kingside looking like a Swiss cheese! but black gets great control over e5 and his kingside knight is rather safe as a consequence.

Originally posted by greenpawn34
And as for instruction...that's a joke.
Name the most famous and instructive game of chess ever played.

Most players would say it's Morphy at the Opera.
A box will never show you that game unless you nudge it towards it
because you know it's there.

...A box will always plays/suggests 8.Bxf7+ or 8.Qxb7 then consider 8.0-0 and 8.Be3.
The instructive beauty of 8.Nc3 (often given a ! by some writers)
would just pass you by.

Hi GP,
I pretty much agree with everything you said. But I will add that Houdini 1.5 doesn't take much of a nudge to come up with Morphy's game. It starts off like all other engines, but after about 5 min, 8.Nc3 pops up as the #2 choice, only a sliver of a pawn away from the #1 choice.

(Also noticed that Black's 9th and 10th moves were poor, allowing the ensuing fireworks.)

No 2 choice after 5 minutes.

How many million positons did it look at?

Morphy flicked it out after a few moments without even looking at the board.

Apparently most of his time was spent trying to watch the Opera whilst the Opera
singers were wondering what all the noise was coming from the Duke's private box.

http://en.wikipedia.org/wiki/Morphy_versus_the_Duke_of_Brunswick_and_Count_Isouard

"I would say opening study is the biggest enemy to beginner progress..."

Nimzo, I think that if you mean beginner <= 1600 (or so) I would agree with you. However, once one reaches Class B->A I think opening study is essential. By that I mean (not just memorizing lines - yet in some positions that is essential) but really learning the nature behind each particular opening - pawn structure, pawn imbalances, bishop vs. knight, understanding the short term advantages/disadvantages vs. long term. Learning opening -> middlegame -> endgame transitions taking into account Queen & Knight vs. Queen & Bishop, Rook & Bishop vs. Rook & Knight along with essential endings all should be studied at the same time. JMHO - since I'm not really sure what is meant by "beginner".

Originally posted by kbear1k
"I would say opening study is the biggest enemy to beginner progress..."

Nimzo, I think that if you mean beginner <= 1600 (or so) I would agree with you. However, once one reaches Class B->A I think opening study is essential. By that I mean (not just memorizing lines - yet in some positions that is essential) but really learning the nature behind each p ...[text shortened]... be studied at the same time. JMHO - since I'm not really sure what is meant by "beginner".

Yeah, I think I've taken this advice to heart a bit too well. I know, seriously, two openings.

For white:
1. E4, and I hope for E5, nf3, nc6, bb5-the classic ruy lopez. Most people don't play E5, so I make it up from there.

for black
1. I hope for 1. E4 so I can play C6. Otherwise I make it up from there.

(sometimes I'll follow the opening database, but I've lost that habit again).

I see threads about collie, and I think breed of dog, or french and I think resigning.

Originally posted by kbear1k
"I would say opening study is the biggest enemy to beginner progress..."

Nimzo, I think that if you mean beginner <= 1600 (or so) I would agree with you. However, once one reaches Class B->A I think opening study is essential. By that I mean (not just memorizing lines - yet in some positions that is essential) but really learning the nature behind each p ...[text shortened]... be studied at the same time. JMHO - since I'm not really sure what is meant by "beginner".

kbeark1k - I suppose when I say beginner I mean something to the tune of Class C or below but I guess it is a pretty vague term.

Originally posted by kbear1k
"I would say opening study is the biggest enemy to beginner progress..."

Nimzo, I think that if you mean beginner <= 1600 (or so) I would agree with you. However, once one reaches Class B->A I think opening study is essential....

kasparov thinks opening study is not relevant under 2000.



I kinda lean towards what you said though, about general understanding of typical positions. but that's not really opening study, is it? related and directly caused by the choice of opening, yes, but not really opening. more like general knowledge you should get down anyway.

Originally posted by wormwood
kasparov thinks opening study is not relevant under 2000.

And that's precisely why I keep my RHP rating under 2000.... I hate opening study. 😉