We have a winner!
Click on the name at the bottom of the picture to read the story.
https://medium.com/@polypragmosyne/chess-master-2000-on-the-amstrad-cpc6128-beated-stockfish-b4bb99bec48d (https://medium.com/@polypragmosyne/chess-master-2000-on-the-amstrad-cpc6128-beated-stockfish-b4bb99bec48d)
Interesting. The CPC has the better algorithm for the beginning level (calculates slower). I think it's not a matter of age. Could you do the game play on the highest level on both sides so we will see who definitly the winner is or how good their math implementation are? I assume that the game will last then a couple of days? (...on CPC side) :)
So when will the CPC be taking on Deep Blue? :D
Bryce.
First try to beat Magnus Carlsen. When that is done, try to beat the best computers
About CM2000 derivated from MyChess
QuoteMYCHESS uses an iterative (https://chessprogramming.wikispaces.com/Iterative+Deepening) Type A search (https://chessprogramming.wikispaces.com/Type+A+Strategy), with alpha-beta pruning (https://chessprogramming.wikispaces.com/Alpha-Beta) as well as the killer (https://chessprogramming.wikispaces.com/Killer+Heuristic) and capture heuristics (https://chessprogramming.wikispaces.com/MVV-LVA). It will predict its opponents best move, and start, analyzing replies (https://chessprogramming.wikispaces.com/Pondering) while the opponent is still thinking. One extra ply (https://chessprogramming.wikispaces.com/Extensions) is examined before backing up from a best variation (https://chessprogramming.wikispaces.com/Principal+variation) if the side to move (https://chessprogramming.wikispaces.com/Side+to+move) can have anything captured.The desirability of a possible position is "scored (https://chessprogramming.wikispaces.com/Evaluation) " on the basis of material (https://chessprogramming.wikispaces.com/Material) strength, using a "swap off" evaluator (https://chessprogramming.wikispaces.com/Static+Exchange+Evaluation) to resolve situations where something is under attack. If a decision can not be made using this score, then a secondary positional score is generated, which takes into account such features as pawn structure (https://chessprogramming.wikispaces.com/Pawn+Structure), piece placement (https://chessprogramming.wikispaces.com/Piece-Square+Tables), and mobility (https://chessprogramming.wikispaces.com/Mobility). When a possible position is found which is better than the current best variation, it is saved in the ply table (https://chessprogramming.wikispaces.com/Triangular+PV-Table) ; otherwise it is discarded.
Back in the days i always prefered Cyrus II, a very good engine
My favourite (non-CPC) was Battle Chess. I was crap at Chess, so at least the animation kept me entertained while I was being thrashed by a computer.
Bryce.
beated ???
on topic: isn't the first level meant to make mistakes? Else what's the point of levels and a beginner to choose the easiest one?
The only things that tells me is that StockFish is easier for beginners.
Go compare the highest difficulty setting, if you want to prove anything.
Quote from: khaz on 13:46, 21 November 17
beated ???
Yup, it's a Michael Jackson song as far as I know. :)
Bryce.
Quote from: khaz on 13:46, 21 November 17
beated ???
on topic: isn't the first level meant to make mistakes? Else what's the point of levels and a beginner to choose the easiest one?
The only things that tells me is that StockFish is easier for beginners.
Go compare the highest difficulty setting, if you want to prove anything.
I agree about the 'beated' thingy - even my browser's spell correction points it out :D /grammar_nazi
About easiness and mistakes: I'm definitely not an expert in chess AI but I'd think that when you choose an easier level it just limits the decision trees and time available for the computer to think. I wouldn't think you can introduce 'mistakes' as such.