Complete.Org: Mailing Lists: Archives: freeciv-dev: January 2003:
[Freeciv-Dev] Re: Graphic resource editor, and an architecture proposal
Home

[Freeciv-Dev] Re: Graphic resource editor, and an architecture proposal

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: Daniel L Speyer <dspeyer@xxxxxxxxxxx>
Cc: freeciv-dev@xxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Graphic resource editor, and an architecture proposal
From: "Eric S. Raymond" <esr@xxxxxxxxxxx>
Date: Sat, 18 Jan 2003 14:16:40 -0500
Reply-to: esr@xxxxxxxxxxx

Daniel L Speyer <dspeyer@xxxxxxxxxxx>:
> At least with the old rendering algorythm, combining graphics that weren't
> explicitely designed for eachother tends to be problematic.  A lot of
> alignment and arrangement is done with small transparent regions, and some
> of the tweaks are pretty ugly (conceptually speaking).  The new renderer
> I'm working on should improve this, but I still wouldn't trust the client
> at runtime to decide what looks good together.

Noted.  That's not a problem we need to think about solving until Phase 2
of the reorganization, though.  Phase 1 is just designed to unglue things
that are inappropriately glued together right now.
 
> It is nessesary to apply multiple labels to one graphic.  We do it only
> occasionally in the curent codebase, but all the time for message window
> icons.  It's a great way to allow tilesets to add features without
> code-modifications.

Easily handled in the new organization.  Hard links are your friend.

> Also, having multiple sprites in one graphic can be very helpful.  Some
> sprites (especially in the new algorythm) are best drawn four-at-a-time,
> with the spec-file dividing them up.  Also, it's a lot easier to draw many
> sprites on one file than to put everything in a different window.  I speak
> from experience.

This is the one bit I don't understand.  Are you telling me that the drawing
algorithm is actually aware of the coordinate pairs in the specfile?

> One final note: the artist meta-info is often wrong.  I haven't
> particularly tried to keep it accurate.  We might consider jettisoning it
> -- after all, nowhere else do individual features get accredited that
> way.   Or we might not.  In any case, we should remember that it's not
> reliable now.

I see.  That reduces the value of preserving it accurately.  But my 
reorganization does that anyway.

> P.S. was it hinted at the beginning of this thread that more chapters of
> TAOUP might be coming soon?

They will be, yes.  There are 18 chapters now.
-- 
                <a href="http://www.tuxedo.org/~esr/";>Eric S. Raymond</a>


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