Welcome, Guest
Username: Password: Remember me
  • Page:
  • 1
  • 2

TOPIC: Large Repertoire Database vs Multiple Smaller Repertoire Databases

Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 4 months ago #26

Let me start things off on the new forum with a simple question:

Do you prefer to have one (maybe 2: 1 for White, 1 for Black) repertoire databases or do you find it more practical to multiple?

I have been doing the former: 1 big database for my Black repertoire where I have my Dutch against 1.d4 and 1.c4; my French against 1. e4, my responses to odd White first moves - generally Dutch in nature. It gets rather big.

But...it gets a bit unwieldy and if I decide to use, let say, my old Nimzo on occasion, I have to add it...and again, more 'totally different' defenses, this just makes management a bit more complex.

That is why I am thinking of changing to several smaller databases. I think it may make for easier 'on the fly' changes to a given opening approach - plus, I do not wipe out whole reams of lines I have worked on.

Thoughts? How are you handling database management?
Last Edit: 3 years 4 months ago by leavenfish.
The administrator has disabled public write access.

Re: Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 4 months ago #41

I'm rather new to CPT and was wondering about the exact same thing - one/two big databases or one per opening?! Maybe some experienced CPT user cares to comment.

leavenfish, you write that the bigger database got "unwieldy"? What did you mean by unwieldy? Like sluggish, slow?
The administrator has disabled public write access.

Re: Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 4 months ago #45

Hi
I've used CPT for a long time but probably not properly or as efficiently as possible! This is why I like the idea of this new forum to swap ideas.

When I first started with CPT,I used individual opening variation repertoires and imported large database (pgn) files. At that time CPT couldn't cope well with large databases and it wasn't a good way to use it.

More recently I have made both individual opening repertoires for black and white and a single overall and general opening repertoire (again for black and white). I hand tune these with selected moves and pgn games of importance in my repertoire and this limits the size of the files in the general repertoire. I find this a much better way to use CPT repertoires, but I am still learning how to use CPT fully and properly!

I hope this helps.Best wishes and good luck with your chess opening repertoire training and learning! :-)
The administrator has disabled public write access.

Re: Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 4 months ago #52

Thanks!

What I do like about the one big database solution is that the scheduled reviews just keep coming for my entire repertoire, rather than just for the opening who's database I happen to have open. I'd probably miss this if I'd go for one database per opening.

I will go with one big database for the time being and just hope performance won't get too bad once the thing grows bigger. That's really my main concern.
The administrator has disabled public write access.

Re: Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 3 months ago #74

My personal approach has been to use Grandmaster Repertoire books. I structured my database around these books.


Create folders: Make a folder called "Caro-Kann Defense". This will include all variations of the Caro-Kann.

Create Opening Variations: Make one opening for each main variation. Caro-Kann has Classical, Advance, Panov, Exchange, Pseudo-Panov, Sidelines. Then in each these opening input only the mainlines or bold text lines from the books. Also being sure to link positions to the chapters in the book as variations and sub-variations. I tend to name them like so "Caro-Kann - Classical Variation".


Create Tabiyas: These are created using sidelines or the non-bold text from the books. I tended to focus on tactical error made from the opponent's side.

EDIT:
Currently I have one database with 4000+ positions. I have imported 4 different books with all the mainline variations and comments effectively creating fully interactive digital versions on the books. My goal is to digitize all 17 books eventually. Not because I wish to learn the openings but for a reference. However I do wonder if I did this and trained until I had at least %90 recall ratio in opening systems. How high would my rating jump? The Caro-Kann of which I have %96 recall ratio on about 790 positions. My results in this opening have improved greatly I even beat 3 masters. So that says something about the type training CPT offers.
Last Edit: 3 years 3 months ago by SuperDave937. Reason: edit
The administrator has disabled public write access.

Re: Large Repertoire Database vs Multiple Smaller Repertoire Databases 3 years 3 months ago #75

Here are some of my thoughts on the subject.
The question of single or multiple databases, or a mix of both! I do not think there is a right or wrong answer to this question. You have to use what works best for you and even this may change over time.

I have experimented with both options and a mix of both. I think the following are considerations:

1. Are you introducing a completely new opening to your repertoire?
2. Your playing strength

If it is a completely new opening for you then it seems to make sense to keep it separate until you get comfortable with it.
If like me your playing strength is low (1,500 ish) then I found that importing an e-book was a total waste of time. This was for a couple of reasons; one of which as a result I no longer import e-books, other than for a specific purpose.
The first reason was that if you import an e-book, which is tempting because you want to play and learn the right lines, the chances are you will not understand it and when you come to the training you will quickly reach a position where you have no idea as to the continuation. Trust me I have done it!
The second point, which I think is critical is that it is not your repertoire.
So I create a separate file for the sole opening. I will populate this with the key lines I am interested in. Depending on complexity I might import and then quickly delete to a manageable size/lines or manually enter. The key thing here however is that for my side of the opening to be included I have to own it. This cannot be achieved easily with the import. Countless times I have played a move in the training session that is wrong only to realise that I have additional lines in “my repertoire” that actually are not or should not be part of “my repertoire”. So you get to the stage where you throw the file away and start again – this time building it step by step and making sure you understand it.
Sometimes I have created a duplicate file to serve a particular purpose. This file may have “three” openings. The first is the same as my original – that is “My repertoire”, the second is the “Book line” – this might actually be an e-book import! The third might be “Actual games”. Now I can play “my repertoire” and see how it compares to the “Book” line and to actual games played. I use this to decide whether to include new moves in “my repertoire”. Using a chess database I will try and see whether the number of actual games played justifies inclusion of a particular line. I think it is important to be selective in what you add into your repertoire so that it is efficient for you and not unwieldy – a personal judgement!
Here there is definitely one advantage to importing the e-book. I have been told many times that I need to understand the moves and not just learn them off by heart. Commendable advice offered by those who “understand” the moves. However I have found that “learn them off by heart” and then when you compare them to the moves you would normally make – the penny drops!
Now when I am happy with my opening file I can add it to a Single file containing all of my openings. This is where the advantage of the single file containing all openings becomes apparent. Mainly you can see transpositions and take advantage of the full scope of training and scheduled recall training.
The problem then becomes keeping the individual repertoire in sync with the combined repertoire. This is probably managed by being meticulous in your approach – which probably explains why my playing strength is what it is and not what I wish it was!
For a stronger player I guess the single combined file approach is perfectly adequate – if I ever get there I will let you know!
Lastly, I will add that this is a fairly new approach for me but based on past difficulties. I will undoubtedly tune my individual files and re-import them into the combined file. I have only just experimented with running the engine analysis over the files. Be prepared to experiment and do what feels right and effective for you.
Dave
The administrator has disabled public write access.
  • Page:
  • 1
  • 2
Time to create page: 0.986 seconds