mclane wrote: ↑Sun Jul 29, 2018 12:42 pm
is the engine automatically switching the style when neither tal/capa/petr switch is on,
and i can manually set a switch to force a certain behaviour in the configuration ?!
If you don't select tal, capa or petrosian, the engine dinamically detects the position/algorithm type in its evalution and search, starting from a
capablanca's default type.
Conversely, if you choose a style, it bootstraps with this style.
This is usefu,l for example, when we know the strarting position's type to analyze.
What 's to be expected if you select all three styles at once? Is that advisable?
To me at least, it appears to play some very attractive chess when all three are selected, leaning more towards a Tal-style.
carldaman wrote: ↑Mon Jul 30, 2018 1:20 am
What 's to be expected if you select all three styles at once? Is that advisable?
To me at least, it appears to play some very attractive chess when all three are selected, leaning more towards a Tal-style.
Thanks for the release, btw!
CL
Hello!
As Shashin states in his book, this is the case of "total caos" positions:
the engine must try before Tal. If it doesn't work, Capablanca and so on until Petrosian.
I have a very old 32 bit computer (2008!) and I find the 32 bit compilation just doesn't install on a fritz (chessbase) or a shredder gui. The shredder gui states it is not a uci engine and the fritz 14 gui asks for the identity of the engine author and doesn't seem to accept any manual author information that I consequently type in.
Not a hugh problem , just seems that it may have an attractive style hence I would like to try it. Any tips?
Is it posible to implement Stockfish time management in ShashChess ( with Move Overhead, Slow Mover, and Minimum Thinking Time parameter)
I have experienced many time losses, where engine will run out of time....
It seems there is some kind of a bug when trying to run Shash engine on 32 cores. The kns are drastically reduced. Instead of a normal 50.000 kN/s, the engine would only give around 18.000 kN/s, and sometimes even much lower… I hope the author takes notice and fix this….