adonthell-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Adonthell-devel] Composite map objects


From: Kai Sterker
Subject: Re: [Adonthell-devel] Composite map objects
Date: Wed, 24 Mar 2010 22:20:30 +0100

On Wed, Mar 24, 2010 at 7:29 AM, Tyler Nielsen <address@hidden> wrote:

> OK.  I have more to report on this issue.
[...]
> I see two possible solutions.  One, always grow the tile positively (remove
> negative values from the xml), and adjust the starting location.  Two,
> adjust the chunk to store the correct Entire position.  I have code that
> does number two, and I think it's what we want to do so I'm going to push
> it.  If it turns out we would rather do number one, it's easy to back it
> out.  One thing though is this causes some other issues.  For example, now
> the character 'falls' to -5 instead of 0.  This causes some more problems
> around the stairs.  But sense I plan on looking at that next it shouldn't be
> a big deal.

I know this whole area is somewhat delicate. Two more issues I have seen:

* Jumping is broken on those -5 tiles, but seems still okay on other
tiles (probably related to the 'falling to -5').
* Rendering seems to be broken somewhat as well. Can be seen in the
path test map (also bet it's off by 5). See:

.libs/worldtest -g ../../adonthell/test/ path


>From your description of the underlying issue, I tend to agree that
fixing the chunk is the way to go. The problem is to locate all the
other parts of the engine that rely on this quirk and fix them as
well. That would be perfect, but might require a lot of work.

Forbidding negative values in the model might be a solution too.
Initially it was required to correctly align the sprite and the shape,
but now there is a separate offset to help with that. So if solution 2
turns out too complicated, solution 1 might be a feasible workaround.

So I'll leave it to you. If you have the time and patience to dig into
this mess, please fix it properly :-).

Kai




reply via email to

[Prev in Thread] Current Thread [Next in Thread]