Workshop: From the mailbag

My question = When I have several engines in deep analysis mode, does the first engine take notice of the others engines analysis and implement that when it make the next move?

Stefan Andeer

No, it doesn't. The engines work independently. That makes sense, really; you're using different engines to get their (possibly/probably) differing suggestions, so (in my opinion, anyway) having one engine influence the others would be counterproductive.

----------

Somebody, very nice, from ChessBase sent me a file (eco.cod) that is 1.4 mega bytes. I have, finally, Fritz7 and ChessBase Light. Both have an eco.cod of just 128K.

I copy the new eco.cod in both programs but I do not see any difference. I may be wrong ( I am just an hobby player of 57 years old trying to release daily world tension or stress by playing chess) but I do not know.

What is this file for? What can be done with it? Etc.,

Carlos Vargas Vidal

Oooooo, a stumper! I like getting these every now and again.

Carlos, I don't know for sure, but here's what I think. Since that file also appears in ChessBase 8 & 9, I believe it's the file that allows the software to identify the ECO code and variation for a particular game. For example, when you play against Fritz you'll see the name of the opening displayed briefly in the Information bar or Chatter pane. I'm pretty sure that this information is drawn from the eco.cod file. And the reason the file you were sent is so much bigger is that ChessBase 9 includes a lot more ECO subclassifications than its predecessors; the file you were sent is the expanded version from CB9 (and I'm guessing that this new expanded file will be included in the forthcoming Fritz9).

And thank you very much for the good wishes you included in your e-mail!

----------

I know that all of your older material is archived, but despite this, I think it would be nice if you could, whenever possible, include in your articles how to perform the article's topic with older versions of ChessBase and Fritz, if the way of doing it is significantly different from the current versions. I don't know how this would be best done, but perhaps something along the line of an embedded link to sketched instructions for using ChessBase 8/Fritz7 to accomplishing the same tasks.

Bob Stepp

I'll tell you true, Bob, I wish I could. It really comes down to the amount of time required to write an additional set of "retro-instructions" for previous software versions: in some cases it would take me triple the amount of time to bang out an article. If I was still a fulltime writer I'd be tempted, but since I literally squeeze my writing in whenever I can, I just can't do it at present.

Also, although I have literally every piece of software Chessbase has produced, not all of it is currently on my computer. And I'd be hard-pressed to even find some of those older program disks in this rat's nest of an office (I was recently asked a message board question about a particular training CD and, despite an hour's search and several book and disk avalanches in my office, I couldn't find it anywhere -- I have no clue as to where it got to). I guarantee you that if somebody asked me a Fritz5 procedural question, the best I could do is scratch my head and look stupid.

But I might try to include links to prior articles in the future if I can remember having written about something before (with more than 400 articles under my belt, that's getting harder to do, too).

----------

I have learned an interesting attacking scheme. Against a Black kingside fianchetto, White plays h2-h4-h5xg6. If Black responds with ... h7xg6, then the Black f7-pawn is now overworked, having to defend both e6 and g6. Then White exploits this overworked f7-pawn by putting a piece on e6, either with or without capturing, to deflect the f7-pawn from its task of defending g6. I saw this idea in the excellent IQP treatise by Baburin and then immediately realized that this is exactly what Alekhine did in the famous Reti-Alekhine 1925 game. Alekhine played 20 ... h5, 22 ... h4, 23 ... hxg3 24 hxg3, and then played the now famous 26 ... Re3!!, the e3-square no longer being defended the overworked White f2-pawn has to keep the Black c7-queen out of g3.

I was wondering if you could set a search mask in Chessbase 9 and have it automatically look for this sequence of moves: h2-h4-h5xg6 (...h7-h5-h4xg3) followed by putting a piece/capturing on e6(e3). Thanks.

Jeffrey S. Hall

Sure. First have a look at this article in which I describe the dialogue's technical details (and check out this one, too).

There's a couple of ways to go at this search. You could set up a list of linked maneuvers in which the pawn marches up the board, etc., but I suspect that the pawn moves won't necessarily be sequential (there will be other moves inserted in the midst of the pawn march).

So I'd recommend setting up a Position search first, in which a White h-pawn is on h5 ready to capture in g6. Make sure you have the Black Kingside pawns included on the search board, too.

Then you'd go to the Maneuver search part of the Search mask and set up the following lines:

wPh5g6xP
b???e6

and do the search. Since this is a Maneuver search, you'll likely need to do a separate search for the Black "mirror" positions. Set up the Position search board with a Black h4 pawn and the proper White Kingside pawn structure. Then in the Maneuver dialogue you'd enter:

bPh4g3xP
w???e3

And that ought to get you what you need.

----------

I have Shredder 9 and Junior 9 and I use both to analyse my games on either full or infinite analysis mode.

I hear that Junior is better for analysing sacrifices and Shredder better for closed positions. Is this correct?

It would be really helpful to have a description of the styles of all the different Chessbase programs and hints on how to use them in tandem; as kibitzers? full analysis mode? Etc

Laurence Roberts

That's an interesting question and one which I hear all the time. The answer is really a pretty relative one. All chess programs are primarily tactical monsters (as we've been discussing in the ongion "intelligent mistakes" dialogue); you're not going to see Steinitzian positional play from any chess engine currently on the market.

However, chess engines do differ from each other in a relative sense: some are a bit more positionally-oriented than others.

In regard to the engines you mentioned, Junior has a reputation for occasionally offering somewhat speculative sacrifices in its own games while Shredder enjoys a reputation for playing a more "solid" (more concrete, less speculative) game.

What troubles me about your question is the use of the word "better", because I'm not going to be able to give you an answer that'll hold true 100% of the time. In general, I think your idea is appropriate if only because I think Shredder is the best of our engines in handling closed positions. But even in open positions I've very occasionally seen (emphasis on "very" because I can count the number of times it's happened on one hand and have fingers left over) Junior suggest a sac that I wouldn't touch with a ten-foot pole.

My suggestion (which I've discussed in previous columns) is to have more than one engine analyze your games, then consider all of the suggestions they offer. For my own purposes, I tend to have three engines do overnight analysis on a game: Fritz, Junior, and Shredder. I use Fritz for the tactics, Shredder for positional suggestions, and Junior as the "other opinion". Fritz and Shredder are usually enough, but Junior sometimes comes up with something interesting which the other engines didn't.

If you'd like to see some additional suggestions on how to use engines for game analysis, check out the index of my articles for the year 2000, scroll down to the bottom of the page to find the January through April series called "A User's Guide to Fritz6", and read Parts 10, 11, and 13 (I don't know where the heck Part 12 got to -- I don't have direct site access or I'd replace it).

----------

That'll do for now. There's more in the mailbag (including additional comments on the "intelligent mistakes" discussion) and we'll look at them later. Until then, have fun!

You can e-mail me with your comments on ChessBase Workshop. All responses will be read, and sending an e-mail to this address grants us permission to use it in a future column. No tech support questions, please.

© 2005, Steven A. Lopez. All rights reserved.