Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Corrected...........

Author: William Penn

Date: 03:01:21 11/07/04

Go up one level in this thread


On November 06, 2004 at 15:11:05, George Tsavdaris wrote:

> My starting purpose was to find the relation between the strength of computer
>Chess engines in comparison with human "good players" (GM's-IM's), and time.
> I wondered what the relation would be and what the diagram of it would be. So i
>could have a guess of, where the strength of engines tends to go at the future,
>if the obvious increasement it stays the same or is decreasing.
>
> Since i don't have the appropriate data for doing such a thing, as it don't
>even exist(very small number of "serious" games between computers and GM's), i
>have done something that looks similar but it's way different and of a much less
>value.
> I've gathered the ratings of the engines that was at first places on the SSDF
>for all these years and made a graph. Unfortunately we can't really conclude
>something important but anyway. The only thing perhaps we see from the graph is
>that the engine strength when they play between them, taking into consideration
>and the hardware evolution, is steadily increases and in fact we don't see any
>decrease into this inceasement, so we can expect further improvement.
>Unfortunately the graph has not the appropriate form, to show us if there is a
>limit at engine's strength.
>Since at 27/10/2000 SSDF have removed 100 points from their list i have done the
>same to the lists before 27/10/2000 in order to have a correct graph.
>
>Here is the data and the graph:
>
>Date	       Rating	   Program
>--------------------------------------
>17/1/1996       2347    M.Chess Pro 5
>4/2/1996        2342    M.Chess Pro 5
>2/3/1996        2340    M.Chess Pro 5
>21/4/1996       2330    M.Chess Pro 5
>28/5/1996       2330    M.Chess Pro 5
>3/8/1996        2321    M.Chess Pro 5
>11/9/1996       2320    Genius 3
>23/10/1996      2382    Rebel 8
>1/12/1996       2373    Rebel 8
>8/2/1997        2362    Rebel 8
>9/3/1997        2362    Rebel 8
>24/5/1997       2487    Hiarcs 6
>2/8/1997        2455    Hiarcs 6
>4/9/1997        2458    Hiarcs 6
>24/10/1997      2439    Hiarcs 6
>30/11/1997      2444    Hiarcs 6
>31/1/1998       2433    Rebel 9
>22/2/1998       2489    Fritz 5
>9/6/1998        2473    Fritz 5
>10/8/1998       2474    Fritz 5
>14/9/1998       2474    Fritz 5
>7/11/1998       2470    Fritz 5
>24/12/1998      2466    Fritz 5
>28/1/1999       2476    Hiarcs 7
>28/3/1999       2476    Chessmaster 6000
>1/6/1999        2477    Fritz 5.32
>26/9/1999       2581    Fritz 5.32
>26/11/1999      2596    Chess Tiger 12.0
>30/1/2000       2606    Junior 6
>27/10/2000      2630    Fritz 6
>28/12/2000      2629    Fritz 6
>5/3/2001        2650    Deep Fritz
>11/6/2001       2653    Deep Fritz
>2/10/2001       2726    Deep Fritz
>27/12/2001      2715    Chess Tiger 14
>15/2/2002       2715    Shredder 6
>27/3/2002       2748    Fritz 7
>22/5/2002       2730    Fritz 7
>30/7/2002       2741    Fritz 7
>28/10/2002      2763    Deep Fritz 7
>5/12/2002       2755    Deep Fritz 7
>16/1/2003       2754    Deep Fritz 7
>13/2/2003       2768    Shredder 7
>26/5/2003       2768    Shredder 7
>6/7/2003        2810    Shredder 7.04 UCI
>17/10/2003      2812    Shredder 7.04 UCI
>25/2/2004       2808    Shredder 7.04 UCI
>7/5/2004        2812    Shredder 8
>
>And the graph is at http://www.angelfire.com/ultra/georgecm9000/SSDF.jpg
>
>If you click at the link it will not work, but if you copy and paste it at the
>address bar it will! Damn computers.....

Yes, I had the same problem with that link, but it works if pasted into the
address bar of my Internet Explorer 6 (SP2). I don't know why.

That's interesting. Offhand it suggests that the top chess engines will achieve
something in the 2850-2900 range next year - if SSDF continues to exist. There
seems to be some doubt whether SSDF will continue. And without SSDF, we may
never know for sure!?
WP



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.