Welcome to this website. We are always happy with new chess programmers. Maybe first step is to create a working chess engine using alpha beta search. And perhaps next step is to beat Fairy max or if that is too difficult Skipper chess engine in online chess tourney.
Henk wrote:Welcome to this website. We are always happy with new chess programmers. Maybe first step is to create a working chess engine using alpha beta search. And perhaps next step is to beat Fairy max or if that is too difficult Skipper chess engine in online chess tourney.
Ahah you guy are very funny !
Well I never answer to yours silly / trolling topics since 2-3 months , very good if you do the same for my posts , please ignore me and concentrate your efforts to learn to skipper to play chess and listen à lot if can , the experts advices 😉
I personally follow the philosophy of "first get it working, then get it working well". Maybe memset is safe, maybe it isn't. I just initialise it manually. It's not time critical code, so paranoia is fine.
ZirconiumX wrote:Hi Daniel! (Remember me from the HGM tourney?)
I personally follow the philosophy of "first get it working, then get it working well". Maybe memset is safe, maybe it isn't. I just initialise it manually. It's not time critical code, so paranoia is fine.
Of course I remember , Zirconium
Ok I will try to initialise by hand to see what happen...
Because I think I have some bugs in my zobrist scheme , I want to try all the possibilities
mmcknight wrote:You don't even need to initialize. The default value will never match your hash index.
Matt
Accessing uninitialised data is just asking for trouble. And the likelihood of anything is never zero. Granted, the chance of the random data perfectly matching is 1 to 2^64, but not all engines store the entire hash in an entry.
mmcknight wrote:You don't even need to initialize. The default value will never match your hash index.
Matt
Accessing uninitialised data is just asking for trouble. And the likelihood of anything is never zero. Granted, the chance of the random data perfectly matching is 1 to 2^64, but not all engines store the entire hash in an entry.
Indeed Mathew but I tested today with memset , it seems safe , all values was 0 .I must find TT bugs somewhere