Author: Albert Silver
Date: 08:34:18 07/31/04
Go up one level in this thread
On July 31, 2004 at 11:02:06, Rob Basham wrote:
>However, from my previous testing experience of GM moves, sometimes
>"pure Karpov" does not mean that it is a "good test move"
>for our computers.....:-)
>
>For example, chess programs today have a "problem" in finding most of
>Paul Morphy's winning moves he made in the past....:-)
>
>Regards,
>Rob
Well, bear in mind that the main issue isn't a simple search or tactical one.
That's why the move is described as a positional move. As to Morphy, I hardly
think one can compare him with Karpov. Morphy was great compared to his peers,
but in a match against Karpov, even today, he would be crushed, and not just due
to openings. Karpov is one of the greatest positional players of all time with a
sense and ability that kept him head and shoulders above everyone else for 10
years, and even into the early 90s, was only put in check by Kasparov himself.
GMs not only lost to him, but were left wondering what had happened. Yussupov, a
great positional player in his own right, described his first game against
Karpov in 'Training for the Tournament Player':
"I have not often felt that I have been outplayed, but in this case that is what
I did feel. However, the most annoying thing was that I had very little
appreciation of how this happened or why."
Karpov is undoubtedly the greatest prophylactical player ever, and in a sense
the move given is right up that alley. It simply prevents White from active
plans while Black builds up the position. Engines can either play it or not. If
they don't, it's not clear that even an hour of analysis will lead them to the
move, yet ask an IM or GM what they think, and I'd be surprised if they
disagreed. Anyhow, just my 2 cents, which is why I shared it. We see enough
tactical blows, but these are the more insidious ones.
Albert
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.