CPT Community

Dear Stefan,

First, this is an awsome program. Having just discovered it, I feel that it's going to be an important tool for my chess training.

I read that you are preparing version number 4, and before it is released I would like to make some very minor suggestions (currently using version 3.3):

1. I think it would be useful to add move numbers, both in candidate moves and in the position tree. For example, instead of "e4", I think it would be useful to see "1.e4"; in fact, it's mostly useful after a few moves, especially in the position tree, for example "5...Nf3". This should be straighforward to add, and if others think it's not useful, perhaps an on/off tickbox in the options would please everyone.

2. I am a strong believer in the comments for both moves an positions. I believe it is more important to understand the position and the idea behind the move, instead of mechanically memorising some lines. This is why I would be most happy to have some options for setting font, and font size for all comments. (this suggestion I think it has been raised before, another user wanted also colors, bold and italics).

3. Perhaps it would also be useful to add somewhere the ECO codes (eventually with the name of the opening). I think it is useful, even though many users don't find any meaning in this, but I'll try to explain why. I read your advice on creating separate sub-repertoires for different openings, but a parallel method could be useful, to also having all the repertoirs in one single giant repertoire. I believe this is useful in order to see how different lines transpose into others (especially having some comments along the way).

4. (Perhaps the harder to implement). It is a beautiful feature that common position have the same comments, no matter what tree line is used to get there. But also, I think that once a common position is reached, the subsequent moves should inherit their comments in order to avoid duplications. One thing is having general comments over the position, and a totally different thing is to have comments on specific moves. In a similar fashion with the position comments, I think it would be fantastic to have the same comments over the same moves after a common position is reached.

Other than that, it is a spectacular software, congratulations.

Best wishes,

Adrian

Dear Adrian,

thanks for taking the time to give me your detailed feedback, appreciated!

1) The new position explorer will have move numbers. The candidate move list not. Of course it could be easily added, but then all moves what have the same move number. What's your idea behind showing the numbers for candidate moves?

2) A future version will let you freely edit comments and then you will be able to adjust any comments as you like. I'm not sure if this feature will make it into CPT 4 yet, but not much later.

3) CPT 4 will support "giant" repertoires. And tranpositions across openings will be detected. ECO codes is on my own wish list too. Probably not the first version of CPT 4 though.

4) I didn't get this one. Right now you can't have two different move comments / moves for the same position if the move notation is the same (e.g. Ne4). In other words you can have for position X move Ne4 with comment A and then another move Ne4 (for the same position) with comment B. What exactly do you mean with "inherit" in this context. Please let me know what I'm missing!

Regards,

Stefan

 

 

Hi Stefan,

Clarifying what I think Adrian meant. I hope I'm not stepping on toes...

(Also clarifying for Adrain, I hope)

 

StefanRenzewitz:

1) The new position explorer will have move numbers. The candidate move list not. Of course it could be easily added, but then all moves what have the same move number. What's your idea behind showing the numbers for candidate moves?

For example, in the Sicilian: 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nc6 5. Nc3 ... leads to a position where Black has many options.

I think Adrian would like the position to show the number 5.xx for the moves in the candidate move list and position tree. It might be doable for the tree but difficult for the candidate move list.

The position tree could insert a move number but that's not really the purpose of a tree and the number would be meaningless. The tree is just a linked list of positions showing candidate moves for available for each position. Since the tree can be built starting from any position, the "move numbers" wouldn't necessarily be relevant to any opening line.

The problem for the candidate list is CPT can't "know" what number to put there because that position may be reached from different lines through transposition and have different possible move numbers. The position may also have been "jumped" to and not have a previous move list at all.

Presumably the new Position Explorer is designed to facilitate studying "lines" instead of "positions".

In CPT 3.3 you can simply refer to the move list to see what previous moves you made and the move number you are on for any particular line.

 

StefanRenzewitz:

4) I didn't get this one.

He describes his requirement as CPT currently works.

It doesn't matter how you reach position X, the position comments and candidate move comments (d5! Striking at the center) are attached. If you are entering a different line from a book with different comments and reach the same position, you will see the comments in CPT that you left for candidate moves earlier. Since there is only one "common position" there is nowhere to inherit move comments from.

I think he might mean common moves for almost common positions.

For example:

Position X has a candidate list move comment "Black attacks the weak d4 pawn" for move [Xa:Rd8].

Position Y differs from X only in that a move like a6 has been played to deprive a piece of a square. But the position offers the "same move" [Ya:Rd8] in this position with the same strategy (attack the weakened d4 pawn).

Adrian would like move Ya to inherit the move comment from the move Xa so he wouldn't have to re-enter  "Black attacks the weak d4 pawn" for the move Ya in the other position.

I may have misinterpreted Adrian, but CPT already works the way he describes so I thought maybe he meant something like this.

 

The problem is:Where would you "inherit" from?

If white plays 17.c3 do you want to inherit "White goes for the Alapin" from 2.c3 in an earlier Sicilian position, or "A bad move" from a position reached in the Two Knights Defence, or "Winning the knight" from another position all together?

This kind of thinking arises when studying opening books because you are looking at different positions from somewhat similar move orders with the author drawing linkages and comparisons between lines. A move like Rd8 may look like the "same move" from the perspective of the variations you are studying when reading the book, but CPT doesn't view Rd8 as the "same move" in different stored positions.

Some CPT users still don't have the concept that CPT positions do not have any objective or historical relation to each other. The only thing any position knows it the available candidate moves and the positions they lead to. A CPT position does not know what move was entered to arrive there. It doesn't know future candidate moves from the positions it links to. It doesn't know how many other positions have appeared in the position tree.

When you use CPT, you are studying positions. What are the elements of the current position and possible candidate moves at my disposal? You are not studying "how did I get here?". As important as that may be....

Mahlon