I could also give a go for your engine :)
Hm, you could last.
Beat me I had a temporary brain departure and blundered a knight.
It's okay dude it was a learning experience for the both of us
Beat me I had a temporary brain departure and blundered a knight.
Well, don't be sad. It happens to all of us! Great engine by the way, I would really like to try it!
I sent you a challenge
I'll take a crack at it. I have no idea of my actual chess strength, honestly, play too much over at lichess where the ratings are skewed heavily.
But chess programming interests me, so taking on an 'infant' engine sounds like an extra sort of fun.
I'll take a crack at it. I have no idea of my actual chess strength, honestly, play too much over at lichess where the ratings are skewed heavily.
But chess programming interests me, so taking on an 'infant' engine sounds like an extra sort of fun.
Sorry, I have been very busy with work, which is why I have not sent you a challenge. I will do so now!
Nice enterprise by the OP, I would surely like to face the engine as well. Challenge me anytime, if you're interested.
Good luck, but you'll probably lose. Just a thought.
Way-to-be a downer, lol
OK, sending everyone challenges now. the main problem in its evaluation is king safety - also, DavidIreland, comparing its moves to top engines helps a lot, and trying to reason through (using Stockfish, because it is open source) the engine's moves. Which is pretty easier with my engine, harder given the Stockfish code, even though Stockfish is open source. To me, the engine's wins can help in two ways: Mainly with its shortcomings (i.e. how it could have improved) and also, how "creative" it was compared to other engines. Engines tend to have an..."enginey" playing-style, and so does mine. Adding in human elements to the evaluation not only improved its play, but made it less engine-like, although it is still very much so (given that it values material over king safety...). I have not had time to make it UCI-compatible, as I have been busy with family and work. I can still, however, have it play against other engines. I will have it play against both Stockfish 6 and Houdini 4 Pro (birthday gift), maybe even the Tarrasch toy engine for giggles. perhaps 40|40 games. Anyway, I learn more from its losses than its wins, however the wins do help. OK, sending the challenges now.
Could you post little you chess code here? I'm just a little curious how it looks like. Or even lots of it. There is now upper limits in lines one post can have. I have studied programming but I do not know nothing what kind of data structures and algorithms is used in chess programming, but I'm interested in.
Hi all, I'm back and looking for more players (old and new welcome!). Inspired by the Giraffe chess engine, I took a look at creating a neural network-based engine evaluation with a standard, but low-depth, search. As such, my bot has a learning mode in which it learns through self-play as well as games of experts I can feed it. The goal is to mimic human-like, expert-level play. If you would like to participate in playing against the engine, in an unrated, marked game, please let me know. Thank you.
I could also give a go for your engine :)
Great, going to send the challenge now!