Author: Robert Hyatt
Date: 07:10:27 05/20/04
Go up one level in this thread
On May 20, 2004 at 08:41:39, Vincent Diepeveen wrote:
>On May 19, 2004 at 23:07:20, James Swafford wrote:
>
>>On May 19, 2004 at 22:28:50, Vincent Diepeveen wrote:
>>
>>>main() {
>>> search(10); // search 10 ply depthfirst
>>>
>>> for i = 1 to 10 // depth limited 1..10 ply search
>>> search(i);
>>>}
>>>
>>>search(int depth) {
>>> if depth == 0
>>> then return eval();
>>> else for all moves
>>> search(depth-1);
>>>}
>>>
>>>Robert Morgan Hyatt doesn't seem to understand this in his thesis.
>>
>>
>>What you describe is called "Depth First Search With Iterative Deepening"
>>by George Luger in his text "Artificial Intelligence: Structures
>>and Strategies for Complex Problem Solving", pg. 106.
>>
>>Note he still calls it "Depth First", but adds the qualifier.
>>
>>You wouldn't argue semantics with Luger, would you?
>
>He has it wrong then.
>
>Even the deep blue team calls it depth limited in their articles.
>
>In fact it is even more breath first search than depth first search in fact.
that is bullshit.
>
>You first search all siblings before moving onto the deep.
>
More bullshit.
Breadth first on a 10 ply search generates _all_ moves at ply=1 and searches
each by applying the static evaluation. Then, for each ply-1 move you generate
all ply-2 moves and apply the static eval to them. Then for each ply-2 move you
generate all ply-3 moves.
That is a _huge_ difference from depth-first.
You _really_ look like an idiot here...
>But it definitely is depth limited.
>
>>--
>>James
This page took 0.01 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.