Author: Jonas Bylund
Date: 01:27:17 09/24/03
A typical response to a new engine seems to be (lately) "EngineX is not as strong as i anticipated, what is wrong with it?" and "EngineX does not beat the ssdf #1 all the time so it sucks" then when EngineX has been out for a couple of weeks/months people seems to rave about it (unless it truly does not do well). Then book issues come into play, no pun intended, and we find out that it matters if the book is suited to the engine or not. Now that is an important aspect of compchess and i am not saying that engine playing strenght is not important, however that is not all... I tend to, lately, to use engines for specific analysis in specific positions and i am sometimes amazed at how well an engine of relatively low rating can handle some positions, where the top dogs either needs VERY long to understand what is going on, or will not get it, within reasonable time, that is. Now the question that i think is equally important is strength of analysis and what engines does well in what positions, this is of course harder to measure, but my point is that there is more to "strength" than a good book and great play. For example: If i where to play as a centaur at say, playchess, and i had a pawn storm on the king side against my opponent then i would still use Gambit Tiger 2.0, for endgames i would use Shredder X, for unclear complicated positions in the middle game i would use Junior X, for very equal positions i would use Fritz X, if i had a position that any of the above considered fairly equal, but i felt i had attacking/winning chances, i would use Gandalf X (in long games that is) etc. (a lot of "weaker" engines could be mentioned too, but this is just an example to illustrate how i view that engine strength is relative to specific positions) What i am saying is that in my experience, engine strength is often relative to the position at hand and finding out what kind of positions that best suits what engine is relative to your approach and understanding of the game + a healthy doze of trial and error. I have learned much more from watching engines play, than looking at the results... Jonas
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.