[Kde-games-devel] Kapman sprite system (also, connecting-tiles systems)

[Matthew Woehlke]

Thomas Gallinari wrote:
> As I said in a previous mail, we do not have a lot of time to spend on
> Kapman for now, just be patient if we don't answer instantaneously
> please !

Apologies if I came across as impatient, I was just replying to "we gave 
[feedback]"... which, until I hear from you and/or Pierre-Benoit...

I'm being patient, honest :-). (Ok, maybe not, but that's Gaël's fault ;-).)

> IMO Mauricio and Eugene are two important artists of KDEGames, and they
> both deal with IDs, as other artist seem to do. This is for me a
> sufficient reason to keep the existing system that is used in Kapman.

Well... *I* am not going to write any tile code that requires all 47*X 
(where X is at least one, possibly 3 or 4) tiles to have id's, groups, 
and invisible rects to make the size right. I'm going to write it to 
requires *two* id'd elements, zero groups, and zero invisible junk.

If someone else wants to "fix" that, and more to the point, do all the 
extra work making 47*X invisible rectangles, then 47*X groups, then 
assigning 47*X id's... for /every theme/... more power to them. 
Personally I think it's an utter waste of time.

> Another reason for that, is that a tile based maze will be a huge
> change in the game, and moreover you seem to be ready to develop an
> animation library (that is a very good point!). I don't think that it
> is necessary to make the amount of work bigger than it needs to be.

The tile code is going to do position-based stuff anyway, so it's really 
very little work to adapt that to the sprites (especially if I'm poking 
at an animation system by then).

Whatever you're able to do as far as poking animation into working more 
nicely would be great :-). But that's for later, I need my frame 
extractor first.

My current priorities are:
1. frame extractor (library class)
2. hacks to use sprites.svg without (or with broken) animation
3. XML maze format cleanup
   3b. any changes needed to support this
4. make sure pathfinding, etc still works with 3b
5. tiles!

My preference would be to make (2) work alongside the current format for 
the moment. (And, to be honest, a lot of (2) stems from my desire to 
test walls with matching sprites.) Hopefully by the time (5) is done and 
it's time to look more seriously at animation, there will be involvement 
of more people than myself :-).

I might switch (3)-(4) and (5), though /not/ doing (3)-(4) makes 
realizing the benefit from (5) require more effort. (Plus IIRC it's 
pretty fast work, and doesn't have much in the way of immediate side 
effects. Long-term there will be some additional theming possibilities 
from (3b).)

-- 
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
-- 
"Who wants to sing?" -- Orcs (Warcraft II)