Author: Ralph E. Carter
Date: 05:44:03 12/24/98
Go up one level in this thread
On December 23, 1998 at 09:05:57, Robert Hyatt wrote:
>On December 23, 1998 at 04:24:09, Ralph E. Carter wrote:
>
>>On December 23, 1998 at 04:20:43, Alois Ganter wrote:
>>
>>>On December 23, 1998 at 03:17:29, Ralph E. Carter wrote:
>>>
>>>>Do you know that engines under Fritz and Junior have their hashtables cleared
>>>>before every move? And that this has a disproportionate effect on those engines
>>>>that intelligently save information between moves, like hiarcs and crafty?
>>>>
>>>>If nobody cares about questions such as these, I am leaving.
>>>
>>>I mailed Chessbase the question last week and they told me to try this
>>>experiment with Crafty as prove that it is decision of the engine alone what to
>>>do with hash tables: Take the second move of a position and let Crafty calculate
>>>some time. Then clip the analysis with search info from the search window to
>>>Notepad. Then go to the first move and let calculate for some time. Then go
>>>again to the second move and let calculate. Then clip analysis and compare with
>>>the first clip. Now the Crafty engine takes 25% less kN to find the same lines.
>>>I tried with the new 16.2 version from Chessbase download and it is right. And
>>>they say 16.2 is much stronger because Bob Hyatt corrected a problem in hash
>>>management.
>>>
>>>The Doctor? engine makes the same effect.
>>>
>>>Alois
>>
>>Posted by James T. Walker on December 22, 1998 at 19:47:21:
>>
>>Bob I hope you see this and can explain. This is the Crafty 15.20 inside of
>>Nimzo 99 and inside Junior appears to work the same way. Notice the times it
>>takes for Crafty to move. When this is run under Winboard Crafty plays the
>>entire variation in only a few seconds. Notice Crafty takes 72 seconds to make
>>the Mate in 1 move. Crafty searches 10 ply for that move but notice that is
>>typical of all the moves. It's a mate in 7 I created several years ago (18) to
>>test some of the old dedicated chess computers.
>>
>>Event "?"]
>>[Site "?"]
>>[Date "????.??.??"]
>>[Round "?"]
>>[White "Crafty 15.20"]
>>[Black "?"]
>>[Result "*"]
>>[Annotator "Crafty 15.20"]
>>[SetUp "1"]
>>[FEN "1k6/1P3p2/K7/8/8/8/7P/8 w - - 0 1"]
>>[PlyCount "13"]
>>
>>{59392kB, nzbook.CTG. PentiumII
>>} 1. Kb6 {#7/12 46} 1... f5 {11} 2. h4 {#6/14 31} 2... f4 {1}
>>3. h5{#5/12 40} 3... f3 {1} 4. h6 {#4/11 39} 4... f2 {1}
>>5. h7 {#3/11 38} 5... f1=Q {27} 6. h8=Q+ {#2/10 11} 6... Qf8 {3} 7. Qxf8# {#1/10
>>72} *
>>
>>Jim Walker
>
>No way to explain that. One point is that after I do a search and play a move
>that leads in mate, the next search terminates as soon as a shorter mate is
>found (to be sure I am making progress). But once I find a mate in 1 I should
>play it instantly unless something else is going on. It is certainly possible
>that the "new" command clears enough things to break this as well...
Here are the results I get for this problem: (PII-300)
Mate in: Time:
#7/12 34
#6/14 15
#5/12 15
#4/11 15
#3/9 14
#2/9 13
#1/9 25
Note that even the mate in one takes 25 seconds.
Wanting to be CERTAIN, I ran it again with a different time control, with
similar results.
THE MATE IN ONE, WHICH WOULD BE PLAYED INSTANTLY UNDER WINBOARD, TAKES 25
SECONDS UNDER THE CHESSBASE INTERFACE.
When the same problem is given to Junior 5, same hardware, same time control,
after rebooting, its results are
Mate in: Time:
#7/18 2
#6/12 0
#5/12 0
#4/9 0
#3/9 0
#2/3 0
#1/3 0
UNDER THE CHESSBASE INTERFACE, JUNIOR 5 SOLVES THE MATE IN ONE INSTANTLY, AS IT
SHOULD.
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.