ELECTRONIC T-NOTES


CHESSBASE USA'S WEEKLY ON-LINE NEWSLETTER


FOR THE WEEK OF MAY 12, 2002


FROM THE MAILBAG -- PART TWO

by Steve Lopez

Contrary to what a few people have claimed, I do not know "everything" about ChessBase software. In fact, I'm often astounded by the ingenuity displayed by users of our software.

Case in point -- this e-mail from Russ Bastable positively boggled me for a number of reasons. First of all, I hadn't realized that you could replace the tactics problems in Pocket Fritz with problems of your own (either downloaded or created using our other software). But, more importantly, Russ has devised a means of using the tactics training feature that has nothing to do with tactics at all! To wit:

...let me offer up a suggestion for using the tactics training feature: Replace the file on the pocket pc with a file containg the moves, variations, and sub-variations of your favorite opening [(that you] get horribly lost in) -- and studying opening lines actually becomes fun. Just remember to replace the first move with the resulting FEN after White's first move when you are studying for black (so that [the] comp starts with pawn at e4 and says "Black to move"). Using ChessBase's "Setup position" makes this simple with the "Copy FEN" button!

Man, that is some first-class stuff, Russ! I'm impressed -- the thought of replacing the tactics database with a non-tactics one and using it for opening training would never have occurred to me in a million years! Thanks for sharing your idea.

Ingenuity again comes to the fore with this tip from Tom Maynard:

I believe I have discovered an even easier/faster way to build a repertoire database from scratch:

I create or open a Book, click on the "Book" tab, and then simply use the arrow keys to select a move, and then press right-arrow to transfer the move to the "Notation" tab! Entering variations is a snap: simply back up (left arrow) to the branching point, and start arrowing your way down the variation. The variation is automatically entered in the Notation frame without any additional effort! If you prefer, the "T" and "M" (enter/end variation) keys still work -- and they save some time if you have a heavily branched opening.

Once the line is complete to my satisfaction, I switch to the Notation frame to annotate the moves, and then just save the whole line to a Repertoire DB and start on the next line in a clean window.

(The same technique works equally well to enter OTB games from paper logs, and saves the trouble of clicking all over the board window to enter the individual moves.)

That, too, is some danged good advice, and I thank Tom for submitting it and graciously allowing me to share it with you. As he said, this technique saves you a whole lot of clicking when entering games. I transfer a lot of games from printed sources into my various databases, and I've also found that you can easily (and routinely) enter the first ten to fifteen moves of GM games by using the cursor keys in a large opening tree.

Here's a quick one from my old pal (and brother Kamikaze) Bob Webb, regarding the crosstable display in ChessBase:

Double-clicking on either player's score for a game will load that game --- saves scrolling through the list.

Works like a charm! And it's something that you'd only discover completely by accident -- how'd you do it, Bob?

And, finally, from Michael Patterson we received more grist for the "hash table mill":

First of all, let me just say, that I am not weak in mathematics. I have no trouble using your formula, but I use another method for optimising Fritz7.

It's real simple, basically I use Fritzmark to tell me which hash settings are best. You just set the hash tables at different settings and then run Fritzmark. Through trial and error, I have found that, on my system, I get the maximum Fritzmark at 32, 48, and 64 meg. of RAM. If I go below 32 meg. or above 64 meg. Fritzmark begins to drop. I have an 1100 MHz processor in my system, so 2*PFreq[MHz]=2200. If I divide 32,000kb by 2200 I get the fastest time control at which Fritz plays with maximum Fritzmark, which is roughly a 10 minute game. Conversely, if I divide 64,000kb by 2200 I get the slowest time control at which Fritz plays with maximum Fritzmark, which is roughly a 19 minute game. From all this I conclude that Fritz will play his strongest game at 10-20 minutes time control. Although, strictly speaking, we have an overlap here of both blitz ( game<15 ), and standard (game>15 ), these conclusions support your statement in T-NOTES that computers are best at blitz.

Thanks, Michael!

Until next week, have fun!

You can e-mail me with your comments, suggestions, and analysis for Electronic T-Notes.