Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: depthfirst versus depthlimited

Author: Robert Hyatt

Date: 09:39:20 05/21/04

Go up one level in this thread


On May 20, 2004 at 12:06:50, Jonas Bylund wrote:

>Bob have you considered that VD has found a way to make you help him even though
>you said you wouldn't again?
>
>Just post some complete nonsense and have you point out where he was wrong :)


I really doubt that, because he _knows_ the kind of search Cray Blitz used.  It
was written up at least in the "Computers, chess and cognition" book.  He
_knows_ that it uses normal alpha/beta with iterative deepening, fractional ply
extensions, etc.  He _knows_.

He's just so incredibly dishonest that he wants to "discred at all costs, even
it if means writing outright lies here."

That is exactly what he does, and he knows it.  Just watch.  He tries one way to
discredit.  It doesn't work.  Or in the case of my 1.7X and 3.1X speedup
numbers, several offer proof that it is fairly accurate.  So then he takes a
different angle of attack "remove that pointer and the non-parallel version will
be 10% faster.  I know, I asked several professional programmers and they all
agree."  Even though _I_ know what the cost was when I first put the pointer in.

But facts don't change his style.  They just change him to a different
direction.  A new set of lies.  A new set of exaggerations.  A new level of
dishonesty...




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.