Even more letters...

by ChessBase
7/30/2006 – The letters to ChessBase Workshop keep pouring in. In the latest edition, columnist Steve Lopez fields queries and offers observations on games which end in time trouble, Fritz training modes, deleting games from databases, Linux software, database caching, and how Chaturanga fits nicely into his fairly dissolute lifestyle. Workshop...

ChessBase 17 - Mega package - Edition 2024 ChessBase 17 - Mega package - Edition 2024

It is the program of choice for anyone who loves the game and wants to know more about it. Start your personal success story with ChessBase and enjoy the game even more.

More...

One more round of e-mails before we move on; as before, my replies are italiczied.


Just to ask you a question,

How to find games that’s lost in time esp.in corr chess database?.I’m using chessbase 9.

Regards,

Masros
Malaysia

It took me a moment to properly parse this one. "Lost in time" has numerous connotations; after all, aren't they all lost in time to some extent? That's the beauty of a program like ChessBase: helping you find that stuff. But then I realized that my correspondent was looking for games in which one player's time had run out; the tipoff was the mention of the Correspondence Database, in which such games would be of special interest.

The whole thing hinges on whether or not an annotator has made such a note in a databased game. It's certainly possible (likely, in fact) that numerous games exist in which no such nore is made. But when the note is made, there are three main ways in which they appear in ChessBase databases:

  1. The last move has the simple (English) text comment "Time" after the move;
  2. The last move has the simple (German) text comment "Zeit" after the move;
  3. The last move has the Informant symbol for "Time" attached to the final move.

Consequently, finding such games might require three separate searches (this depends on how much "weeding" you want to do later; see below). To start a search, right-click on the database's icon in the Database window in CB9 and then choose "Search" from the popup menu to bring up the Search mask. It's a good idea here to click the bottom "Reset" button (the one between the "Help" and "Cancel" buttons) to clear any prior search criteria that might still be hanging around in the search mask.

Click the "Annotations" tab to bring up that search dialogue. For your first search, type the word "Time" (without the quotation marks) in the "Text 1" box. If you want to cut down the number of searches, you could also type "Zeit" (again without the quotation marks) in the "Text 2" box. You can also add a "Symbols" search by left-clicking in the "Symbols" box (to place the cursor in it) and hitting SHIFT-CTRL-Z simultaneously on your keyboard. The Informant symbol for "Time/Time Trouble" (which looks a lot like a gunsight's crosshairs) will appear in the box. Then click "OK" to start the search.

This will pull up all of the games in which any of these three search criterion apply. My search in Correspondence Database 2004 found 461 games. However, this is the part where the "weeding" comes into play. The first game on the list (Seibold-Keres, 1932) wasn't lost on time; instead, the word "time" appears in the comments to Black's eighth move. However, Game 444 on the list (Bodnar-Mikusik, 2001) has the "Zeitnot" symbol attached to the final move. So you'll need to check the final move of each game to see if it actually was lost on time; if not, the text or symbol annotation appears earlier in the game. That's why I mentioned that you may or may not want to do this as three separate searches; doing separate searches means that you'll have fewer "hits" per search and it'll make the weeding process a bit easier.

Thanks for an interesting question! -- SL


I knew that my previews of Bangiev's Squares Strategy CDs would provoke a lot of commentary and the responses have proven me right. I've seen quite a few comments on chess message boards regarding my previews, and the e-mails continue to come in even now. It's about a 50/50 split between people who love them and people who hate them, and this, too, was not unexpected. Here's one from the "hate them" camp:

I purchased Bangiev's three CD's on squares strategy about a month ago. The first CD is incomprehensible without looking at the other two CD's first. Had I read the many unfavourable reviews earlier ( on various chess sites) I would never have wasted my money on these CD's. I was looking for a quick fix and there is no quick fix in chess. Undoubtedly Mr Bangiev has put in a lot of effort in his theory but for an average club player like me the CD's are just not worth it. The money is better spent elsewhere...Bangiev's CD's are languishing in my cupboard while I spend time on good books such as 'How to chose a Chess Move' by Soltis etc and working on my limited repertoire. Perhaps I'm not perservering enough to study the Squares Strategy CD's until the thought process becomes second nature to me. Meanwhile I play a lot of Internet Chess ( Blitz ) and generally make do with pattern recognition and whatever tactics I'm capable of. Guess I should try out Bangiev's method in my correspondence games.

Capt. Amar Sekhar

The line "I was looking for a quick fix and there is no quick fix in chess" positively leaped off the screen at me. I told you that, right there in my preview of the first disk in the series!!! My entire emphasis in that preview was that these disks require a lot of work on the reader's part. So it's not as though you weren't warned.

I will say, however, that your plan to try the "B-Method" in your correspondence games is excellent. If one is going to try a method which requires them to restructure their whole means of thinking, it's best to do it in games in which time is not of the essence. I'm currently doing the same thing myself in my poker play. I'm reading a book which introduces me to a completely different method of thinking than the one to which I'm accustomed and I'm attempting to apply those principles in my games against computer software rather than in my online games (in which you have x seconds to act when the betting comes around to you).

So, to some degree, I feel your pain (although I'm still disappointed that you didn't heed my prior warnings). All I can suggest is that you give Bangiev a shot in your correspondence games and see how it works out for you. Thanks for writing. -- SL


And, in response to one of my other comments about IM Bangiev and his work ("The best you can do is the best you can do") came this:

"Good enough is good enough" is my equivalent of your motto. All mine own I think.

Wyn Evans

Wyn also added some personal context (which I omitted; I didn't know if you really wanted that published, Wyn), which I do understand completely.

As Jason Boland (one of my faves) once sang, "Who needs to fix it if it ain't broke?"

Life is good. Thanks for writing, Wyn; your e-mail made my day! -- SL


Just curious, is there an article on developing a repertoire? I have CB9 and have seen references on making a repertoire DB and was wondering if there's a good guide on how to use that feature in CB9?

Mike Grant

There sure is, Mike -- well, sort of. The article was written for CB7 and although the "click here" instructions might be a bit different in CB9, the basic ideas are still the same. You can find the article here. Thanks for writing! -- SL


Regarding the new training modes in Fritz9:

Is there a way to specify which move is taken by the program?
Taking first move of every game in Big or Mega database is surely not best way.

Felix Kayser

No, sorry. There's an internal algorithm built into the program which lets the software figure out what it considers to be the best position for setting up a training drill, and it's not user-configurable. -- SL


Just after reading your column on the three training-features i gave it a try. Maybe i am wrong, but i think i found a little bug.
While doing the defense-training a pawn was defended by a rook, but between the pieces was a Queen of different colour.
Fritz wanted me to click the pawn to continue thats wrong, isnt it?

Uwe Jazeschen

Yes, if it was indeed defended by the Rook. But not if the Queen was between the pawn and Rook (since the Queen would be interposed, the Rook isn't directly defending the pawn). This is all just speculation, since I didn't see the position; send it to me and I'll have a look. But in Fritz' defense I'll say that I've used these new training modes a lot (and they have definitely helped my board vision), and I've yet to see an error by the program. -- SL


I added some training games to the opening database in Fritz9 but initially no notation or moves were shown when I accessed the opening training dialogue box.Consequently I wanted to remove/delete these from the database and did so.However these still remain listed in the opening training selection dialogue box & this is quite unhelpful (& quite annoying).
Can you please explain this to me and tell me how to remove the unwanted games from the dialogue box?

Richard Dickinson

I need a little more information, but I'll take a stab at this. My guess is that you did the first step for deleting games (marking them for deletion) but not the second (physically removing them). The first step you've doubtless found: right-click on the game, select "Edit" from the popup menu, and then select "Delete". The game now appears in the list with a line through it. The second step will physically delete the game from the database: go to the Tools menu, select "Database" and then select "Remove deleted games" from the submenu. This will irretrievably remove from the database all games marked for deletion, which is precisely why it's been made a two step process. -- SL


I also like the position tutor but probably for a different reason than you. I'm a night owl and when all my friends have long gone to bed and I feel like I want company I fire up Fritz, turn on position tutor, coach, written chatter and play away. Almost feels like I've got company :-)

I noticed your examples have some of the grammatical errors corrected. Frederic sent me a complimentary early version of F9 (German instruction book, but interface was English). One of the big errors in F9 tutor was "Black/White threats to win material". Finally drove me up the wall so after playing around in various files I figured out how to correct those errors. Also decided to fix some spelling errors (e.g. "Assassin" in the playing levels), and for fun changed the coach's name to Sensei. I've never done anything like that before. Never underestimate the motivating power of grammatical annoyances in us anal retentive types...yes, I know i've mangled grammar somewhat fierce here myself.

Anyway, I have nothing of import to add, nor do I have any questions (I think you've answered just about every question I can think of in your various T-notes and articles over the last few years). Just felt like sending off a friendly response to your article. Keep 'em coming, Steve.

Ken Cowcill

Thanks, Ken, for both the e-mail and the chuckles! I've spent more late nights with Fritz (and a beer) than I care to shake a stick at. (There's a mangle for you: ending a sentence with a preposition). And I'm glad that you didn't break anything while tweaking those files; you're a braver man than I am.

Fred's a great guy. Someday I'll tell everybody about the stuff he's sent me -- the t-shirt story's pretty funny. -- SL


Would you write something about tablebases in CB9?

How to get them working?
How to utilize them?

TapaniS

Yup, will do. The "getting them working" part was something I did a few columns back. I touched on how to utilize them, but I reckon I could expand on that a bit in the future. If I forget, you remind me, OK? Thanks for writing! -- SL


I am a user of fritz 8 and also Hiarcs 9 and a daily visitor of your web site.

By the way I want to know that whether your newly released fritz 9 has a Linux verson or not because who caan bear the price of windows as well as fritz and on the other hand the great operating system Linux is absolutely free. Please inform me

gopal das

No, sorry. There's an economic reason for this: it takes a pile of time and effort (i.e. money) to develop software for a particular OS and if it's not going to sell enough copies to cover the development costs, it's not going to get made. And I'm not just talking through my hat here; we had years of requests for a Mac version of ChessBase and, after one was released, the sales were (shall we say) somewhat less than mediocre and loads of money got lost on that particular venture.

I'm in no way busting on Linux here, but it's nowhere near as popular as Windows. If the day comes when the split between them gets closer to 50/50, I'm sure that a Linux version will be seriously considered. -- SL


Here is an interesting idea for the playchess server: a four player shaturanga chess. Team games are generally much more fun than individually played ones.

Gabriel

This is probably another "development issue" (see above), but I'll say that I would personally love that, especially if it was "classic" Chaturanga (with the original pieces, the dice, and everything). I'm a Chaturanga freak; I play it a lot (two, three, and four handed, sometimes underhanded) with my little band of "outlaw" musicians and, yes, drinking and wagering (often non-monetary; think of it as a grown-up version of "Truth or Dare") are frequently involved. I blame it on our total lack of character. If classic Chaturanga was offered on Playchess, you'd have a hard time getting me off of that server. -- SL


I own several ChessBase products and have also been a subscriber to ChessBase Magazine for several years. I enjoy your regular column on chessbase.com!

Something I've always wanted to know: is there any way to 'freeze' Fritz's analysis of a position temporarily but have it resume again later where it was stopped? For example, to let it think long enough to get to 16-ply in a position, tell it to save its state, and then later (eg, the next day) have it resume analyzing at 16-ply, rather than have to start over at the first ply?

Also a comment re: ChessBase 9 that I hope you will share with the developers-- I like the always-running 'reference' tab that automatically searches one's reference db while one plays over a game, but in my opinion it would be greatly improved if ChessBase 9 would cache recent db search results-- it's annoying that if you back up a ply or two, ChessBase starts its search all over from scratch rather than simply fetch a cached result. (Or is there an existing setting I may use to control this?) Especially when using the Megabase as the reference database, position searches can take a long time: even with a centrino Pentium M it sometimes takes 30 seconds or longer.

Another speed improvement would be an option to load the whole reference database into RAM at the outset, if the PC has enough (or does such an option/setting already exist?). Reading data in RAM is orders of magnitude faster than reading from disk.

Thank you for writing such a useful and insightful column!

cheers,
--Will Rhee

Thank you, Will!

There's no way to interrupt an Infinite analysis session in Fritz overnight and have it resume the next day. If I had to guess at a technical reason (early on a Sunday morning and while on my first cup of coffee after a very long Saturday night), I'd say that it would require some big honking files to be stored on the hard drive in order to cache things like the ongoing hash tables, and that a user setting would be required for deleting these files when they were no longer useful.

ChessBase 9 does have a database cache setting. You can change the parameters in the Tools menu. Click "Options", then the "Limits" tab, and you'll find a setting for "Cache size reference DB", along with a helpful display of your available RAM. Keep in mind that this applies only to the database you've selected as the Reference DB and that any RAM allocated will detract from the amount you can allocate to hash tables and the endgame tablebases while running an engine analysis on a position. But I think this'll do some or all of what you're asking, depending on your available RAM. -- SL


Finally, some observations on the "intelligent mistakes" and "distributed computing" columns. Your comments and observations have been (and are still) appreciated, but I'm no longer publishing replies to them for several reasons. As I've observed before, such a discussion is theoretically endless and there's other ground that needs to be covered in this column. Another reason, though, is that many of the replies are either reiterating what's already been said, or have been reduced to personal attacks on me or other commentators -- this latter has, oddly enough, come mainly from people claiming to have advanced academic degrees. I don't care what you think of my IQ (or lack thereof) but I'm not terribly interested in a mathematician flaming me for being a historian; save your cross-discipline flamewars for the campus lounge. But I'll give you this: that kind of thing has greatly increased my enjoyment of the TV show Numb3rs -- I'm laughing a lot more.

Until next week, 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.


© 2006, Steven A. Lopez. All rights reserved.


Reports about chess: tournaments, championships, portraits, interviews, World Championships, product launches and more.

Discuss

Rules for reader comments

 
 

Not registered yet? Register