Author: leonid
Date: 17:40:43 11/28/00
Go up one level in this thread
On November 28, 2000 at 14:30:28, Robert Hyatt wrote:
>On November 28, 2000 at 14:17:41, Will Singleton wrote:
>
>>I ran it out to 9 ply for my program and Macchess. Move ordering is similiar to
>>the usual, but I'm using PVS. My searches do tend to blow up in the higher
>>plys, for some unknown reason. I'm going to try negascout just to see what
>>happens.
>>
>>From this position:
>>
>>[D]rnbqkbnr/ppp2ppp/8/3pp3/3PP3/8/PPP2PPP/RNBQKBNR w KQkq -
>>
>>Amateur
>>ply score time nodes n/s pv
>>
>> 3 90 0.08 3332 24960 Pxe5 Pxe4 Qxd8 Kxd8
>> 4 90 0.17 8256 29100 Pxe5 Pxe4 Qxd8 Kxd8
>> 5 125 0.33 20785 37740 Pxe5 Pxe4 Qxd8 Kxd8 Be3
>> 6 118 1.20 66639 49920 Pxe5 Be6 Bb5 c6 Pxd5 Bxd5 Bf1
>> 7 95 3.48 218850 57540 Pxe5 Pxe4 Qxd8 Kxd8 Bg5 f6 Pxf6 Nxf6
>> 8 128 9.00 555919 61740 Pxe5 Pxe4 Qxd8 Kxd8 Bc4 Nc6 Bf4
>> 9 110 42.00 2739007 65160 Pxe5 Be6 Qd3 Nc6 Qb5 Rb8 Nc3 Ne7 Nxd5
>>Nxd5 Pxd5
>>
>>MacChess 4
>>02/01|00:00:00| 296| +22| dxe5 dxe4 Qxd8+ Kxd8 Bc4
>>03/01|00:00:00| 1195| +22| dxe5 dxe4 Qxd8+ Kxd8 Bc4
>>04/01|00:00:00| 6651| +22| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Bb4+ Bd2
>>05/01|00:00:00| 31790| +25| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Nd7 Bxf7 Nxe5
>>06/01|00:00:00| 121312| +32| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Nd7 Bxf7 Nxe5 Bb3
>>07/01|00:00:01| 422581| +36| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Nd7 Bf4 Nh6 Bxh6 gxh6
>>Bxf7 Nxe5
>>08/01|00:00:03| 1263936| +31| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Nc6 Bxf7 Nxe5 Bd5 Nf6
>>Bg5
>>09/01|00:00:13| 4478639| +30| dxe5 dxe4 Qxd8+ Kxd8 Bc4 Nc6 Bxf7 Nxe5 Bd5 Nf6
>>Bg5 Bb4+ Nd2
>
>
>My numbers look like this: (one cpu of course)
>
>depth nodes
> 1 124
> 2 1053
> 3 6981
> 4 16929
> 5 50059
> 6 149436
> 7 451934
> 8 1649304
> 9 4307609
> 10 14367818
Maybe my numbers for this position could say something about my move ordering as
some kind of reference. I will say only about brute search since I am not that
sure about official name of my selective search.
AMD 400Mhz. Fixed depth search. Each ply have special procedure. Will indicate
how many times each ply, between 8, was called during the search of this
position. Will indicate how many time legal moves were created for each ply and
how many legal moves were used for each ply.
In my program search start with ply 7 and end at ply zero.
Number of ply |Number of time that|Number of leg.|All legal moves
|ply was called |moves used in |created for this ply
| |this ply |
-------------------------------------------------------------------------
7 |2 |60 | 60
--------------------------------------------------------------------------
6 |60 |916 | 1405
-------------------------------------------------------------------------
5 |914 |2970 | 32728
-------------------------------------------------------------------------
4 |2970 |18305 | 91175
-------------------------------------------------------------------------
3 |18305 |106692 | 229851
-------------------------------------------------------------------------
2 |97430 |773783 | 133076
-------------------------------------------------------------------------
1 |725510 |1820855 | 8129210
-------------------------------------------------------------------------
0 |1820855 |3518164 | 3533725
-------------------------------------------------------------------------
Total number of all plys called during the search of this posit.: 2666046
Total number of all legal moves used in all plys: 6241743
Total number of all legal moves created for all plys: 13348230
Time of search: 26 seconds.
Leonid.
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.