[MUD-Dev] Quake II has gone GPL

[Bruce Mitchener]

Daniel.Harman at barclayscapital.com wrote:
> From: ling at slimy.com [mailto:ling at slimy.com]

>> Of course, it could be argued that the engine can be reworked, or
>> there's already a Quake 2 mod to add skin and bones (or there
>> probably is a project to do such a thing emerging right now).  My
>> point was that if this part is deemed to be needing a rework
>> along with whatever else, would the ROI on Quake 2 be worthwhile?

> Well I don't think you can expect to write the next Everquest/DAoC
> on it, but I think the point was to design a game around the
> engine. Worse than the boneing issue I would imagine, is the BSP
> based rendering engine. Whilst this is ideal for interior
> environments, its no use for exterior, so you'd have to splice on
> a whole different engine and put portals in to bridge (I don't
> know what portaling support quake 2 has, I suspect none?).

A big disadvantage of something like Quake is the level-based nature
of it.  For any of us that are used to the more dynamic
games/muds/worlds where building and such all happen at runtime,
it'd certainly be non-useful.

> I don't even think that rendering the dungeon walls is the hardest
> part really, character animation/path finding/collision
> detection/occlusion detection are all more difficult I feel, so
> given that you'd have to rework so much of this to build a typical
> MMORPG, I think you'd be pretty stuffed if thats what you were
> aiming for.

For character animation, check out Cal3D (http://cal3d.sf.net/).
That looks to be one of the most useful things to have fallen off of
the WorldForge tree.  For rigid body physics and such, there is ODE
(http://www.q12.org/ode/ode.html).  I haven't used either of those,
but they look capable from some quick glances at them.  Both of them
are LGPL, so they can be used in a commercial project without major
limitations.

> As someone mentioned previously, it might be ideal for a
> Neverwinter Nights style small group dungeon crawl type game. Of
> course half the focus NWN appears to be its tools, and I don't
> know how many budding GMs are going to be wanting to use complex
> map modeling tools.

The map modelling tools are a major thing that seems to be missing
from the open source stuff.  Really, much of the whole artist
toolchain is missing or lacking from a lot of the efforts.  Support
for 3DStudio Max plugins or Maya plugins appears to be
minimal. (Despite things like Flexporter for 3DStudio existing.)

I've looked into some of this on my own time out of curiosity and
because it seems like something interesting to learn about.  From my
perspective, some of the more useful features to have would be:

  * The basics (dynamic lighting, multi-texturing, animation)
  * Terrain support
  * Visibility/Culling
  * Documentation
  * Good artist toolchain support
  * Portability
  * Capable of having advanced features added like pixel/vertex shaders,
    imposters, cool shadowing stuff, etc.
  * Decent license (not GPL, etc.  LGPL, BSD, MIT/X11).

There seem to be a couple of engines out there that meet most of
those (but none of them have really good artist toolchain support).
There are even some decent terrain engines coming along that
integrate with other renderers for the rest of the world
objects. (Check out vterrain.org for info along those lines.)

Is anyone on the list working with a system like the above?  Have
you worked on improving artist toolchain support?  Is there any
interest from someone in working on that sort of thing? :) It seems
like that'd be an easier path forward potentially to getting
something that is vworld-capable faster (and with a better license
for potential commercial usage than the GNU GPL).

Just some random thoughts ...

- Bruce

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev