Author: Bo Persson
Date: 16:10:55 02/17/04
Go up one level in this thread
On February 17, 2004 at 17:47:10, Dann Corbit wrote:
>On February 17, 2004 at 17:22:59, Bo Persson wrote:
>
>>On February 16, 2004 at 19:08:38, Dann Corbit wrote:
>>
>>>
>>>The search is more important than the evaluation, in my opinion.
>>
>>I think that they are both important, but that a simple alpha-beta routine is
>>explained in many books on algorithms. It is also rather short and well
>>researched and documented.
>>
>>An evaluation is much harder to start from scratch, IMO. Partly because there is
>>much less solid material to read up on.
>
>Where have you been looking? This is quite good:
>http://home.vicnet.net.au/~chess/posi.html
Wow! A really thorough guy. Thanks for the link.
>
>Here is some more stuff I turned up in a jiffy:
>http://myweb.cableone.net/christienolan/coach/evaluating.htm
>http://en.wikipedia.org/wiki/Chess_strategy_and_tactics
>http://www.markalowery.net/Chess/Tactics_Strategy/Menu/tactics_strategy_index.html
I have seen some of this before.
>
>This is what I use:
>Play Winning Chess (Winning Chess) by Yasser Seirawan, Jeremy Silman
>Winning Chess Brilliancies by Yasser Seirawan (Author)
>Winning Chess Endings by Yasser Seirawan (Author)
>Winning Chess Openings by Yasser Seirawan (Author)
>Winning Chess Strategies (Winning Chess) by Yasser Seirawan, Jeremy Silman
>Winning Chess Tactics by Yasser Seirawan, Jeremy Silman (Author)
>
Seirawan I haven't read. Maybe I should.
>>>> The alpha/beta code is also less than a page long, out of
>>>>the 60k(?) lines of Crafty.
>>>
>>>~42000, unless you add in the EGTB stuff of Eugene, in which case it will be
>>>larger:
>>
>>Well, I didn't actually count it but "borrowed" the number from one of Bob's
>>posts.
>
>Did you notice that Bob's program is 6 times larger?
Yes, Bob does some C things (manual inline?) that make some of his code really
long.
Of course I also get your hint that the smaller program cannot be a complete
copy of the bigger one. :-)
>>
>>Bob has also numbered the bits in the "wrong" direction.
>>Can't he get anything right? :-)
>
>If you had numbered them in the same way as Bob, because of seeing how he did
>it, do you feel that would have been doing something improper?
If that was the only reason, maybe. On the other hand, the two most obvious ways
to number bits is left-to-right and right-to-left, so it is just
circumstantial(?) evidence at best. :-)
Now I know that Bob got his numbering from the Cray bit scanning instructions,
while I found it equally obvious to map my enum {a1, b1, c1, ...} to the equally
numbered bits on a PC processor. That just happened to be right to left.
Bo Persson
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.