Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: a question to Tord about detecting threats in null move

Author: Vincent Diepeveen

Date: 18:17:43 10/04/03

Go up one level in this thread


On October 04, 2003 at 20:26:26, Sune Fischer wrote:

>On October 04, 2003 at 19:45:25, Vincent Diepeveen wrote:
>
>>So you go 1.h3 the opponent directly reacts and plays within a flash of a second
>>out of his book 1..e5
>>
>>You hope to confuse your opponent and play 2.a3 and avoid your opponent from
>>saving time by book. With the usual confidence of an Uri (because who else on
>>the world would be crazy enough to play h3? and a3?) you happily laugh to the
>>left and right. After a while you look again at the board and while swallowing
>>the last remnants of your meal that morning you realize that opponent has
>>already answerred instantly 2..d5 and that only YOU lost time.
>>
>>So a) you have a lost position
>>   b) you you have a time disadvantage
>>   c) you are very early out of book and the opponent has still
>>      another 10 easy to find moves to play where you struggle and struggle.
>>
>>There goes all your advantages of surprise...
>
>If you assume the opponent has a book that covers everything then there's not
>much to be done.
>However, if the plan works, he should be out of book early so the real fight
>between programs (not books!) can begin.
>
>>I remember that i surprised some years ago some engines with the very bad 1.a3
>>then after 1..d5 which they all played diep would go 2.d4 and then pray the
>>opponent was out of book.
>>
>>I even remember how i fooled Nimzo. It was just too plain aggressive psq
>>program. I had figured out that after 1.a3,d5 2.d4,d5 3.Bd2 it would be out of
>>book and give away a pawn at c5 directly. Then dxc5 and soon a3 c3 e3 and white
>>would be a pawn up.
>>
>>However that doesn't work anymore for quite a few number of years already. The
>>last time that something like this worked was like 1998 or so.
>>
>>That's 5 years ago.
>
>Last I checked it still works, just doing stuff like 1.d4 2.c3 3.e3 4.Bd3 will
>take many engines out of book.
>
>>This just shows how important it is to visit world champs regurarly.
>
>Yeah but world champs is not the only place to gather experience.

but guys like uri definitely need to get beaten pathetically for 11 times in a
row there before they start to believe some things the commercial guys are
writing here regurarly.

Like that 1.h3 is in book.

also your d4 c3 e3 bd3 thing is strategically already going to lose of course,
apart from being in the book anyway.

this is such trivial things tried already in Aegon tournaments up to 1997 that i
do not understand some really still in 2003 guess they can fool engines there.

>>Otherwise your 'tricks' are outdated quite some.
>
>My experience is that blindly following long book lines against programs with
>killer books, is certain death if your own book sucks. You _must_ deviate at

if you follow a mainline definitely that is the case.

>some point and let the engine take over, preferably before it's completely lost.

but you cannot deviate at move 1 simply with such idiotic moves like h3.

If out of the first 4 moves in your game you play 3 pawn moves to 3d rank then
you're gonna lose for sure.

>Just remember how Junior got smashed in the first game against Garry.
>Great example of a bad book line, Junior went for a long think out of book, but
>too late too little and the game was lost.

Not really. Gary as usual played an idiotic openings blundering a pawn away, but
junior played if i remember well e5?? or some positional blunder like that after
which it got smacked.

Gary just played some bluffing line without caring for objectivity clearly. That
junior blew it so quickly must have not amazed him after the horrors Deep blue
produced a few years ago.

I guess he was more amazed about the better endgame and far middlegame that the
engines have today.

He probably still thought them to be fritz5.32 level, because all those
Grandmasters i speak to, vaste majority uses chessbase + fritz5.32 that's the
level they prepare against.

Other engines do not run in chessbase you know :)

>-S.



This page took 0.07 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.