I'm not so sure...
I've been asking myself this same question recently, and am fairly new to correspondence chess.
I guess it depends on what your goals are.
For me, I'd like to use CC to improve my OTB. "Well, play OTB, or FICS then!" Sure, sure, in an ideal world maybe. I work 6 days a week and have a life outside of work and chess as well, so finding time to play slow OTB is NOT easy. I manage about 1 or 2 slow games a week.
Why not quicker chess games? I believe quick games are detrimental to one's chess development. (I'm an unofficial student of NM Dan Heisman, google him.)
So to recap:
1. I'm trying to improve slow, tournament like OTB chess.
2. I don't have a lot of time for slow OTB practice.
In steps Red Hot Pawn! Which allows me to play at a s-l-o-w pace. I can even manage to get some moves in during lunch at work.
Downside is I don't get practice of Time Pressure, like you'd have with OTB. So it's a compromise, but at my current level, I don't believe that's to bad.
So, given all the above. Why oh why would I choose to use a database? I wouldn't be able to in a live game, so I won't. "Ooooh, so your CC rating won't go up very much." to that I say... big deal. The skills that I develop will by MY skills, that I can apply to a OTB game.
So am I way off base here? Is my plan to use "CC without a database" to improve my OTB sound?
Originally posted by XanthosNZwhat do you do when your opponent diverges from your own db? go hunt in the big database or similar, pick a couple of favourite strong players, and check a lot of games before adding them to your db?
A self-constructed one. If you know what is in your database you can be more confident in the conclusions you draw from it.
Originally posted by The Starry KnightThere are two ways to use a database in correspondence chess - follow the theory blindly as a move-chooser and then as soon as you're out of book panic, or, explore openings using databases to get used to the typical ideas and positions. The latter will help you OTB, and who knows, maybe you'll come up with the odd TN?
Why oh why would I choose to use a database? I wouldn't be able to in a live game, so I won't. "Ooooh, so your CC rating won't go up very much." to that I say... big deal. The skills that I develop will by MY skills, that I can apply to a OTB game.
So am I way off base here? Is my plan to use "CC without a database" to improve my OTB sound?
Originally posted by The Starry Knightnothing wrong with that, but you'd improve more using databases. there's nothing good about imprinting the wrong patterns into your brain playing your own ad-hoc openings. it's true you'll learn things analysing your own openings, but you can & should do that with databases also.
So am I way off base here? Is my plan to use "CC without a database" to improve my OTB sound?
the brain is a pattern recognition machine, a classifier. don't feed it false data!
Use all the databases you want. At SOME point your opponent is probably going to play a move that isn't in your database and you're on your own. If you don't understand the position, all using a database is going to do is allow you to start playing bad moves later rather than earlier. You are better off to use a book that explains the ideas behind the moves you are playing. That way, when your opponent varies, you won't be looking at the position like you have no idea what's going on.
Originally posted by Sicilian SmaugIt depends on what your aims are. In my case, I specialise in 1.c4 as white and 1. ... e6 as black - and only in a few variations of each. My db started with games by strong players who use these particular lines, and over time I've added other games (most importantly my losses and draws), notations, post-match analysis results etc.
How does you create your database, is it hard?
My aim is to build a strong, but targeted db on the openings that interest me. I can't imagine doing this for a wider array of chess openings. Much too hard for me - but I'm sure many others are trying this.
Various Responses to Various Questions
I guess my self-compiled comment wasn't the most clear. I wasn't suggesting that people use databases constructed of only their own games (although that does have a place in analysis [if you play a certain opening and lose every time that tells you something important]). I was suggesting that people don't just buy a database and ignore what is happening behind the curtain but instead know what is in their database to help them know how far they can trust their conclusions.
I have a pruned database (that still needs more work) of half a million recent high quality Master games that I've sourced from various places that I use most often. A (free) good source not already mentioned is ThisWeekInChess (back issues are available too).
I also have a slush database that encompasses all of the above games plus all the games I've pruned from it plus a whole lot of either older (going back to Greco - NN hilarity) or lower quality (there's some weird scholastic chess in there somewhere) games. I use this if I just want to see if something is a novelty or not or in a few cases to use in my games.
As for pruning your database to only contain a single opening, I don't quite understand the reasoning for that. I can filter my databases in Chessbase so it will only show certain openings (or games with certain positions or a myriad of other options) but the rest of the database is still useful in other circumstances (transpositions, analysing games that aren't mine, interest). If I'm using my database to look at my game positions what does it matter how many French Defence games I have in there (other than a slight search time increase)?
For anyone asking what program to get to access databases I highly recommend Chessbase. You can compile pretty much any stat you want in there, it will handle huge databases without problems and it's not a memory hog. The Fritz interface (Shredder/Hiarcs etc. come with the same thing) is a little more awkward to use as a database tool. It lacks some of the functions of Chessbase and is a bit of a hog (being primarily an engine interface). Chessmaster as a database tool is pretty rubbish. I've heard horrible things about the starter database it comes with (let's face it, 500k average games is like 5 minutes work for someone with access to the Internet and I suspect that's how much work went into it) and also about the interface itself.