TalkChess.com

macOS - ditto .. for me anyway ..

Oops - ignore .. I goofed. Silly goose ..

Tried compiling latest dev for my Mac and got this .. somebody help me here please .. :D Been compiling for my Mac for a while now without issues - incidentally Brainfish compile was unsuccessful too - cant remember error now. Whatever happened with/to Mac compiled uploads of SF and BF ?? T -msse -m...

dkappe wrote: ↑

Tue Jul 23, 2019 1:50 pm

48x5, q-ratio 0.75 (more stockfish, less weak humans), trained on the same data as Bad Gyal 5.
https://github.com/dkappe/leela-chess-w ... i/Bad-Gyal

Thanks !

dkappe wrote: ↑

Mon Jul 22, 2019 2:04 pm

https://github.com/dkappe/leela-chess-w ... i/Bad-Gyal

Bad Gyal 5, 128x10, 1.3M games, q-ratio 0.5

I will release Good Gyal (q-ratio 0.75) as a 48x5 later today.

Thanks !

Waiting for Bad Byoy

feral, aggressive ……

dkappe wrote: ↑

Sun Jul 14, 2019 8:34 pm

Bad Gyal 4a

The endgame data was problematic. This is a rerun without that data at 128x10. This should be the strongest net tactically, including T40, LD2.

+1

And then there's Beckham which means (using English words no longer in fashion) - 'from the farm with the brook running through it' - seriously ..

https://database.lichess.org/ I downloaded several 2019 PGN's. They contain about 30 million games each. When I make a selection on elo=2200+ (both players) the output is below 1%. Makes me wonder, either there are not many strong players or they maintain a non-standard elo rating. Need to be caref...

I learned in one-two hours of analysis more than they have learnt in decades of study and practice. I guess they are natively not that gifted. Lack of vitamins. (apart of course from top talmudic scholars). Might be an idea for you to take that non-self-evident 'truth' and shove it up your south fa...