0
$\begingroup$

I have made 8x8 array that should work as chess table and seperated faces to invidual objects. Then i have chess piece knight as object that i want to select active with mouse click and then find array plates that are possible to move on. So i know, that knight can have max. 16 checker areas with its's "L" movement where it can go on. What i would like to achieve is python script to first get knight's current location, and then find out if there is another object at for example knight location + (1, 2, 0)

EDIT:

For now i have come this far (i started learning python today)

import bge
import GameLogic as g
import mathutils
cont = bge.logic.getCurrentController()
obj = cont.owner
scn = g.getCurrentScene()
own = cont.owner

pos = obj.worldPosition
move1 = obj.worldPosition + mathutils.Vector((2,4,0))

so with this code i get one possible vector knightpiece can go, but haven't yet find way to check if there is other object at the same vector point

Improve this question
$\endgroup$

1 Answer 1

Reset to default
3
$\begingroup$

I would not use the 3D world at all. This is just to make it look nice. I suggest follow the MVC (model-view controller) pattern where the model is in Python.

Model

E.g. the board as 2d-array (list of lists) of fields. Each field knows what figure is on it (or if it is empty).

Any operation will be performed on that board (moving figures, removing figures etc.). The chess rules will be applied on that operations.

View

The view is what the user sees. So this is the 3D world. With all the meshes and materials. Whenever the model changes ... the view (3D world) will be updated. You can add eye-candy animations wherever you want. It has no impact on the model as it is not important to the chess game itself.

Important is that the model notifies the view when something changed. So the 3D objects know they need to do something.

Example A: the King moved from A1 to B2 (model change). You notify the view that there was a change.

The King sees: "I'm relocated to B2" -> so he jumps, crouches, flies .... to the location associated with B2.

Example B the King moved from A1 to B2 and tower (at B2) gets removed.

The King sees: "I'm relocated to B2 hitting the tower" -> so he fights .... to the location associated with B2. The Tower sees: "I'm need to disappear" -> hie fights with the King and disappears with an explosion.

I hope you get the idea. (If not imagine your display is a 8x8 block of ascii characters). The model should still be able to do chess.

Controller

How does the model know what operation to perform? Here you need the controller (not to confuse with the BGE logic brick controller). The controller notifies the model about any input from the view (or somewhere else).

Example A

When the user clicks on the king the controller tells the model the user selected the king. The model calculates what fields are available for moving the king (the model knows the rules!). Then it notifies the view that there is an update (selection changed, clickable fields changed). The view shows a nice representation (e.g highlight the King, let the available fields glow).

When the user clicks on a available field (B2), the controller tells the model that this field was selected. The model moves the King to B2 ... see example A of section view.

Example B

When the user clicks on the king the controller tells the model the user selected the king. The model calculates what fields are available for moving the king (the model knows the rules!). Then it notifies the view that there is an update (selection changed, clickable fields changed). The view shows a nice representation (e.g highlight the King, let the available fields glow). As B2 is occupied by the tower, the tower gets an highlight too.

When the user clicks on the tower/field (B2), the controller tells the model that this field was selected. The model validates if that was a valid move removes the tower and moves the King to B2 ... see example B of section view.

Conclusion:

The MVC pattern, let you isolate the core game logic (chess rules, board data, chess operations) from pure visual aspects and user input.

The model should be independent -> the model should not know about the view or the controller. It should just know there is a view and how to tell that it should update (e.g. communication via messages).

Keep in mind the model could use any view and controller such as text output and calculated turns (from a chess engine/AI) in any combination. There can be even several of them at the same time.

The model defines an interface for communication to the view and from/to the controller.

I do not think that there is a sufficient way to implement a chess model without Python in the BGE.

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thanks for the answer as it clarifys lot of things i have been thinking too compicated ways. Now my problem is here that i don't know how do i check if place is occupied allready. I can check location of B2 easily, but how i check if there is something in spesific location with python. $\endgroup$
    – Ari Liikamaa
    Jun 24, 2016 at 6:41
  • $\begingroup$ when you use a 2d array as suggested above you can store the status of each single field by assigning an object. E.g. board[1][2] (which represents B3) = None (which is empty) or board[1][2] = blackKing. You can easily check if a field is occupied: isEmpty = board[1][2] is None, what figure is on that field: figure = board[1][2]. The access is very fast allowing you to do a lot of checks at one turn. And yes you need some more data such as list of figures for black and white ... $\endgroup$
    – Monster
    Jun 24, 2016 at 7:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .