I think the API in viewportfx's blender/gpu is close to being stable. I was going to refactor cdderivedmesh, csg_subsurf, and editderivedmesh soon. I'm not touching the game engine for a long time ;-)<div><br></div>
<div>I actually think doing your work on top of my branch would be best since I've already extended ghost to initialize using egl, but I'll most likely need to help you get things running on you Dev system <span></span></div>
<div><br>On Friday, March 14, 2014, Arnaud Loonstra <<a href="mailto:arnaud@sphaero.org">arnaud@sphaero.org</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On 03/14/2014 08:44 AM, Jason Wilkins wrote:<br>
> Viewport fx branch contains an up to date OpenGL 3 core profile and<br>
> OpenGL ES port. I left out the android stuff, it only contains the gl<br>
> stuff.<br>
><br>
> I estimate it would take me less than a week to port the game engine<br>
> over to use the new GL support BUT I have not done it since the main<br>
> work is still incomplete and it would just make my branch a bigger<br>
> target for merge conflicts (it's already huge, as you can imagine)<br>
><br>
> I separate effort to port the game engine independent of viewport fx<br>
> would be possible. I certainly wouldn't want to hold you back. I'd<br>
> have to defer to somebody who had a more informed opinion about how much<br>
> drawing code blender and the game engine should share.<br>
><br>
> (Honestly the code in the game engine makes my eyes hurt)<br>
<br>
I've never dared to say that but I completely agree. That's what made me<br>
move to gamekit and other frameworks. Also that's what makes it<br>
difficult for me to tell how much work it would be. (P.s. are there any<br>
plans for a rewrite? Or adoption of GK?)<br>
<br>
If there is already an effort on the way I would rather join that one as<br>
opposed to starting a separate one.<br>
<br>
I've had more ideas to add to the game engine. I.e. I've always missed<br>
the integration of networking. Doing it through python the right way is<br>
not easy. Having an OSC brick would be nice but even better would be<br>
having support for file descriptor polling in the game engine. It could<br>
be added to the proposal. Another idea would be enhancing the video<br>
backend. But before adding more ideas, my first aim was getting the GE<br>
on embedded hardware.<br>
<br>
Rg,<br>
<br>
Arnaud<br>
--<br>
w: <a href="http://www.sphaero.org" target="_blank">http://www.sphaero.org</a><br>
t: <a href="http://twitter.com/sphaero" target="_blank">http://twitter.com/sphaero</a><br>
g: <a href="http://github.com/sphaero" target="_blank">http://github.com/sphaero</a><br>
i: freenode: sphaero_z25<br>
_______________________________________________<br>
Bf-gamedev mailing list<br>
<a href="javascript:;" onclick="_e(event, 'cvml', 'Bf-gamedev@blender.org')">Bf-gamedev@blender.org</a><br>
<a href="http://lists.blender.org/mailman/listinfo/bf-gamedev" target="_blank">http://lists.blender.org/mailman/listinfo/bf-gamedev</a><br>
</blockquote></div>