I'm still a little wiped out after last week's rambling screed on Kasparov and Deep Blue, so this week's Electronic T-Notes will be a brief one.
When I first received the Fritz3 module for CBWin a couple of years ago, one of the first questions that crossed my mind was "How do Fritz's numerical evaluations compare to its Informant symbol evaluations?" In other words, when Fritz shows the symbol for "White is slightly better", how much worse (in numerical terms) does Black's game have to get before the symbol changes to show a clear edge for White?
I sat down with some games in ChessBase and let the Fritz module run merrily as I played through them. By the end of the evening, here's what I'd discovered:
An evaluation of 0.20 pawns or less is considered an equal position by Fritz.
An evaluation of 0.21 pawns to 0.69 pawns means that White/Black has a slight edge.
An evaluation of 0.70 pawns to 1.24 pawns means that White/Black has a clear edge.
An evaluation of 1.25 pawns or better means that White/Black is winning.
I believe that the Hiarcs modules may use different evaluations but I need to do some more research on it.
Hopefully, this will clarify the numerical evaluations for you and will
provide some perspective on what the numbers mean.
Please let me know if there are any tech problems, questions, tips, or general chess topics that you'd like to see covered in ETN. Submissions are also appreciated, and dirty jokes (while irrelevant) are prized, too, if I haven't heard them yet. Just e-mail me to let me know what you'd like to see in ETN. I can't guarantee a personal response to your e-mail but your tips will lead me in the right editorial direction (and your jokes may lead me in the wrong moral one).
Until next week, have fun!
As always, I'm very interested in reading your opinions of this (or any other) issue of T-Notes,
as well as my book Battle Royale. I invite you to post comments to our
ChessBase Users Group or
else you can e-mail me directly.