Two Things make the Engine ibn-kahba

big part of getting better at chess is just being able to tell when you're winning and when you're not, my compass or sense of goodness or pretty much my evaluation bar is not the best because I often play into positions that I think I'm better but then I'm actually worse and sometimes I play into positions I think I'm worse but I'm actually better and I just can't convert, I feel like the ability to determine whether you're better or you're worse is very important

-last year I had a very good friend I met in the very club where I played my last tournement, we were working (we haven't made anything, just pies in the sky ideas) on this like one pro coding project we were trying to create like an AI that could play, not chess of course, I don't know like Tic Tac Toe or some number games or like you know, just an engine in general, let's just say hypothetically like checkers

-the AI, the way you create the AI, you need two parts really, you need to have branching, so you need to be able to go to a certain depth to calculate all the possible variations from this position and then the second the second part of the AI is probably the more important one is the evaluation or like the estimate value of a position, so if I were to give this AI a position, just a position nothing else and I were to say okay it's a white to move or in checkers it's like oh it's this side to move, give me an exact evaluation of who do you think is better, and that's probably the most important part of the AI in general, because you can't go to an infinite depth in theory.computers can't calculate infinitely deep they must stop at certain depth, the strongest chessengine stockfish is estimated at around 3600 elo, chess is not mathematically solved, you could go to certain depth and then see who won at the end of that depth and just that's your evaluation, so the evaluation would always be either white wins black wins or a draw, that's it. but you can't because computers can't do that, so they have to stop at some point and that's the furthest depth,for example you're at depth five, you're at depth six and so on, but at that depth how do you know who's better, you can't just go to depth seven or depth eight, you have to make your judgment too, and making that judgment is what makes an AI good and what makes an AI bad, it's just that simple, if your judgment is very on point you will play to the correct positions and you will win more, otherwise you just lose, I think that's very analogous it's identical to how you should play chess, you have two parts, you have your calculation ability, how deep you can go, like tactics related. and how far can you think, can you think four moves, in five moves, in six moves, in the more moves the better because if you see one move ahead and your opponent sees one less, you're two steps ahead.

-the second most important thing as I mentioned in the begining is being able to tell if you're better or not or if this position is good and then the scale on which it is good, because at that point once you get to like depth five in your calculations, you just need to be able to say am I winning here, am I better, is my opponent better, and then you need to be able to make that judgment and that's what really makes a player good player so there's just really two parts how deep you can go and how accurate you can judge the position, I love the super GrandMaster Hikaru Nakamura, I watch his youtube channel alot, pretty much in every game he plays, he'd look at a position and be like "oh this this position feels like a minus 6 or minus point 2 and he would be like spot on ! like not 100% but like 90% of the time he' be spot on with the engine and that's what makes a grandmaster really good, of course I'm not a Grandmaster but this is like my observation only, so I feel tactics can train your depth right and then theory and game understanding would train your evaluation ability and I feel those two are the most important

- playing chess, in general is, of course you need a ton of theory and a ton of opening prep, a ton of like ( I don't know the word it's just like the pattern) you learn tactics that will help you calculate better, you learn opening this and that " like put your Rook on the back seventh rank" " keep attacking the direction of your Pawn chain" " don't push your kingside pawns" you know those are all like like rules of thumb, they're correct most of the time, those are like shortcuts, the engine doesn't really go by those rules, not like oh if I put my Rook on the seventh rank it'll give me extra blah blah blah.... no they just do all the math, they do everything. they just do all the calculations, and the human has something that the engine doesn't which is like the ability to have tips and tricks and traps, we can come up with those kind of things, it makes chess fun for such a boring game, I think chess boils down to those two factors the ability to think far and the ability to evaluate accurately if you can get those two things down I think you're going to be a decent chess player