Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: My guess for rest of Smirin match: 0.5 0.5 0.5 0.5

Author: Daniel Clausen

Date: 05:01:20 04/22/02

Go up one level in this thread


On April 22, 2002 at 07:09:12, Otello Gnaramori wrote:

>On April 22, 2002 at 06:15:43, Daniel Clausen wrote:
>
>>
>>Um.. what exactly did you see as an "excuse"? I'm not defending either side...
>>You seem to imply that I'm one of the "GMs are better than comps" guys. I'm not.
>>But I'm also not one of the "Comps are better than GM guys" btw...
>
>I'm glad to hear that...

:-)


>>And my impression is that Smirin could reach
>>an advantageous position, but didn't play precise enough to turn the position
>>into a win.
>
>The precise play is a common human problem as you probably know, as well among
>GMs.

Agreed.

>The time cannot simply be a valid excuse, AFAIK isn't a bullet timing.

Let me rephrase what I mean. Maybe it's clearer then.

As long as we're not speaking about 'perfect chess', there are several ways to
play strong chess. Both GMs (I use GMs as a replacement for strong human
players) and computers are capable of doing this. The _way_ they do this is
completely different.

While the GMs typically are great strategic planners their tactical skills are
not that good. (Please note: compared to computers!!!) On the other side we have
the comps which are more or less the opposite. (godlike in tactical positions
and sometimes close to idiots concerning strategic planning)

Now let's see how the strength/weaknesses of both are affected when given more
time.

(a) Computers given more time

(a1) Comps - Tactical skills
Will be improved a bit, but don't play a very big role in comp-GM matches
because the comp is tactically way stronger anyhow already. [In comp-comp games
this plays a bigger role though because both comps are tactically strong]

(a2) Comps - Positional skills
While it's true the computers can find positional moves given more time, often
this is not enough, because they would need WAY more time in order to 'convert'
the positional reason into a tactical one. So positionally comps don't get much
profit from more time.

(b) GMs given more time

(b1) GMs - Tactical skills
More time definitely helps GMs to significantly improve their tactical skill.
(see bullet games on ICC :p)

(b2) GMs - Positional skills
Will be improved a bit, but not that much.

So to summarize:

-More time increases their (comp&GM) strength a lil bit, but not too much
-More time more helps the GM to become better tactically than it helps the comp
to become better positionally.

This is why I think "more time" more helps the GM than the comp. That and you
earn a cookie if you read this far. :)

Sargon



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.