Hi there, I think it would be really cool if you included "good moves" and "great moves" to the computer analysis as a tally underneath the inaccuracies mistakes and blunders.
I feel this is a good idea as it gives positive feedback and a sense of accomplishment
This has been asked before, and I think it's a good idea in theory BUT.... how do you teach a computer what a good move is? The only "good move" a computer sees is its best move. The rest are inaccuracies, mistakes or blunders. I saw a request to indicate brilliancies in analysis, but that can't be done for the same reason - computers don't know anything about aesthetically pleasing moves or anything like that, and I would think that programming an AI to behave like that would be difficult, and probably lower its strength, because it would have to "think like a kuman" which isn't really the point of engines.
I agree completely. It would be near impossible to have an engine judge the aesthetics of a move when there are so many possibilities in chess. Perhaps a good move can be issued when a move is made that keeps the advantage in the position the same or minus 0.1 points in value. As for a great move..? perhaps something to do with depth of evaluation?.
Good movie would be "punishing their mistake" and great move "punishing their blunder".
Stockfish cannot see the difference between taking their hanging queen for free and an spectacular sacrifice, if the evaluation shifts for the same value.
Something might be interesting in the last sentence of message #3.
Let's suppose you foresee a brilliant combination of moves m1, m2, ... mp by your opponent.
Playing something against m1 or m2 isn't always easy, and in any case wouldn't deserve an exclamation mark.
However, suppose you play an apparently innocent move which will make mp to fail. Wouldn't that move deserve an exclamation mark ?
There are many examples of such "litlle" moves destroying a brilliant combination...
A sequence of moves forcing a piece to be overloaded would also deserve exclamation mark(s), IMHO.
And I don't see why a sac leading to mate wouldn't deserve one ! (or 2 for an unfrequent mate pattern after a fair number of moves).
While it's true that softwares dont "feel" anything, it remains true that they are designed to "talk" to human beings.
Just my euro... ;-)
