The chess.com analysis engine is a jerk

All it ever does is critisize.  "+1.32 ?? Your move sucks, you should have played our move which we rate at +1.34".

Is it too much to ask for a nice ! or !! reward once in awhile for extra shrap moves!?  I'm not a machine myself!  I need a little encouragement once in awhile!

lol I don't trust the engines on here anyway, maybe the one at your membership isn't too bad I wouldn't know, but the free one supposedly rated at 2000 in reality is only really probably 1600. Get yourself the latest Fritz or something... Then again that certainly won't be much kinder, just take it on the chin . Sometimes when I play a move it totally baffles the computer, but the computer doesn't understand that certain moves may only be made for phsycological reasons.

I like your idea a lot! and yes the chess engine is poor. I played a book move on move 5, and it said that i made a mistake!

The only problem is that what would a engine think a good move is? something that it missed, or a move that it would have played. Unfortunately, computers know what a bad move is, but they don't know what a good move is. because they have most variations covered.

Good idea though!

Ravi

I'm not crazy about Fritz's analysis either or Chessmaster.

If I was rich I'd spring for a coach.  It's not that I'm afraid to know I made a mistake, just that I want to know "why" & no analysis engine I've encountered can explain.

I do my own analysis on certain games but I know I'm still missing stuff, keep in mind this thread was meant somewhat humorously. ;)

Tell me about it!  I miss being 19, 20, 21!  Youth truly is wasted on the young!

How about 78?  One has to check, double check, and triple check.  But experience and study and practice helps out a lot.  Compensate, compensate.

Being limited by ply often creates the 'horizon problem' in chess programming.  The program checks X moves out, where X is based on your current membership level.  The program assumes each player makes the best move available and evaluates moves based on the position that arises X moves later.  However, it won't notice if that position allows a forced mate or crushing material loss in X+1 moves.

Those results often occur immediately after check or capture, and by default, Crafty extends its search in those situations.  The hard move limit on searches can negate that feature though.

If you download the open-source program (see http://www.craftychess.com/), you can run the same analysis for more moves than even a platinum membership offers, or just indefinately for some period of time, avoiding horizon problems.

I haven't used it in a while, but I belive the chess.com annotation is based on Crafty's annotate or annotateh commands, which produce annotation based on Crafty's analysis.

The chess.com interface is friendly, and I like the move parser for showing the position after each move, showing alternate lines, and the like, but the actual analysis is sometimes questionable.

If I had the code for chess.com's interface to combine with real Crafty analysis and annotation, I think it would be the best existing annotation tool around.

I've always wondered why it doesn't reward good moves (which might look far ahead in the game) with ! or !! as the original poster stated.

My guess is because the computer technically will play the "best" move, which is the move it thinks you should make, so if you make the "correct" move, it's expected, so to it nto a great move, and if you don't make the move it thinks you should make, then you made a mistake or a blunder. 

Think of it as Mr. Spock. Did Spock ever say WTG! or Gr8 Job! No. He only critisized. Your complaint is not logical.

Yeah but spock ussually lost to kirk. Cause he was too logical.

Looking through your games I think you are probably wrong there. You're poor in your openings, strategically and tactically you also struggle. Of course you're still learning, but I don't see any moves you make that can be considered better than a computer's. Give us an example of a move on this site which the computer considered bad, but you think was actually quite good.

ROFL You really think that?

That's a very crude analysis of the position and the way the game has gone so far. Yes, I was wrong to exchange my Bishop for your Knight earlier. But that has been my only mistake so far, while you have made numerous bad positional errors. In fact I'm not sure if any of your moves so far have actually been particularly good. So while I think I may probably only have an edge in the position, if you carry on playing like that I will win for sure. certainly a computer (even the one at around 1600 standard on this site) would have found better moves than you every time. Anyway we aren't really allowed to be discussing an ongoing game.

Really, you're just plain wrong, c3 is just one of the many terrible moves you've played so far:

So in summary, I don't care whether you still think you're better. I'm going to win this game and I hope it humbles you a bit.

Sorry to everyone else by the way that this thread has gone off topic, but this guy just keeps bugging me with his arrogance. Well I guess the trolls have again succeeded in dragging me into a pathetic internet squabble.

It's because it is so smart that it notices some stuff that happens in many moves which you don't see. Sometimes it makes mistakes but it is not programmed for every move. Only programmed most of the moves.

Or maybe it's just syndrome X...

Yes, going back on topic the advantage that humans have is that we don't have to do a brute force check in order to find the best move. Although the very top engines now have over taken humans probably, I think.