Sorry your browser is not supported!

I want to load an image into a memblock, which I will query, I have a character on screen which will move about by key presses from the user, the routine will translate the characters position onto the memblock and return data from the image which will represent somewhere the character can or can’t move.

That’s the theory.

I need to translate the characters X,Y position into:-
dbReadMemblockByte ( int memblock, int position );

The screen’s co-ords 0,0 is in the top left hand corner and so I assume the memblocks position variable will also start in the top left hand side.

I am guessing that to get the X,Y co-ords into the position I just need to work out the size of the image in pixels (say 200 wide and 400 high) and then add or subtract 400 accordingly to the position variable every time the character move up or down and add / subtract 1 for left or right.

Is this about right? Or am I not understanding how these things work?
(Hope the above makes sense.)

The first 12 bytes of the memblock are the dimensions of the image held in 3 integer values - width, height and colour depth (in bits) - for a total of 12 bytes. Then the data starts at this point.

The position in the memblock can be calculated like so:
Position = ((Width * y) + x) * (depth / 8) + 12

If you know that you will always be dealing with 32 bit images, then the (depth / 8) can be replaced with 4, or 2 if always using 16 bit images.


Yes, almost. When moving left, remove (depth/8) from the position, when moving right, add it. When moving up, remove width*(depth/8) from the position, when moving down, add it.

Thanks for that Ian, how did you find all this out? There is virtually no documentation on this...

The format is the same as DBPro, and I've been using that for quite a few years now. The format is also fairly basic so it's easy to work out with a little experimentation.