Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: FRC Fritz 7.0.0.8 vs Fritz 8.0.0.8 (Fritz7.0.0.8 is much stronger ) !

Author: Stephen Ham

Date: 10:39:47 06/25/03

Go up one level in this thread


On June 25, 2003 at 11:52:15, Eduard Nemeth wrote:

>In this 15 Positions: I agree with You.
>
>I have beat easy Fritz 8 but against F7 no (see in archive my games).
>
>But: this 15 positions are not real FULL-FRC in my eyes!
>
>Eduard

Dear Guys,

This is indeed surprising. However, here's my uneducated opinions.

I've noticed that at very fast time controls, such as this (15') that "weaker"
engines sometimes prevail. But the stronger engines clearly dominate at longer
time controls. In fact, in my tests, the difference grows with increased time.
Most of my test games are conducted with 75'-90' per engine/game. Then I've
never seen a "weaker" engine win any tournament from stronger ones.

Why is this? Somebody made the comment above that F7 outsearched F8. Again, my
speculation is that F8 is "smarter" (I think I've read claims to this effect
from CB). So under these rapid time controls, it won't seach as deeply (all
things being equal) since its greater heuristics causes it to search in greater
detail, which slows it down. Being slower, it can't search as deeply (all things
being equal). Rapid games are usually won by tactics, so faster engines will
prevail over "smarter" ones, all things being equal.

Here's a caveat: I'm probably the least informed poster here regarding chess
engines. Nonetheless, my experiences above seem to support my conclusions.
However, I can only speculate regarding the "why" of this issue.

All the best,

Stephen



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.