Jump to content

  • Log In with Google      Sign In   
  • Create Account

#Actualpatishi

Posted 08 July 2013 - 06:31 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move and a killer move that are actually the same(!).   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i am doing something wrong 


#10patishi

Posted 08 July 2013 - 06:31 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move and a killer move that are actually the same(!).   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i do something wrong 


#9patishi

Posted 08 July 2013 - 06:30 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move and a killer move that are actually the same(!).   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i do something wring 


#8patishi

Posted 08 July 2013 - 06:30 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move and a killer move that are the same move actually .   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i do something wring 


#7patishi

Posted 08 July 2013 - 06:29 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move AND killer move that are the same.   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i do something wring 


#6patishi

Posted 08 July 2013 - 06:28 AM

Just another question, i can't stop thinking that i am doing something wrong.   so i need to check some things:
 

A.  whenever the alpha beta function returns a score (beta cut off or not), i am saving the current position in the Hash table with the evaluation and the current move that  was tried.     this is the so called "hash move" that i will use later in case this position ever occur again.    but I read somewhere (a chess blog) that i don't always have a best move to store...that is a little confusing, because i DO  always have a best move!     the move under investigation will be the best move!   so i don't get it...what am i missing here?

 

 

B.  the times when i store a position in the TT after a beta cut off, the hash move that i store is the same as the killer move for that same depth (if i understood correctly,a killer move is a move that caused a beta cut off).  so many times i have both hash move and killer move that are the same.   is this normal?     i always try the hash move first, and after that i check if the killer is the same move,and if it is ,i don't play it and just move on to the next.    so many times, i don't  benefit at all from using the hash move cause i have the killer moves anyway.         up until now i haven't noticed a significant improvement from using the hash move sad.png      that got me thinking that maybe i do something wring 


PARTNERS