Jump to content
Search In
  • More options...
Find results that contain...
Find results in...
timmie

Stuff

Recommended Posts

Well, as many of you may know, I haven't had much time to work on ZDoomGL lately because of a nasty combination of work and wedding plans. I just thought I'd say that I'm now (happily) married, so my free time / work ratio should be going up, so I can start working on ZDoomGL on a more regular basis again.

Just a few updates on some things that have changed since I last posted an update:

- cvar for controlling internal texture format (gl_texture_format). This defaults to GL_RGB5_A1, with valid values of GL_RGBA2, GL_RGBA4, GL_RGBA8 (best quality), GL_RGB5_A1 (obviously), GL_ALPHA, GL_LUMINANCE, and GL_INTENSITY. The last three make for some cool effects. http://members.shaw.ca/timstump/images/zdoomgl_22.jpg is an example of GL_LUMINANCE :)
- ZDoom 3d bridges are mostly working
- better support for unpegged lower/upper textures (although still not perfect)
- started working on getting texture scaling to work

I'm sure there's more, but I can't for the life of me remember what all has changed =/

Share this post


Link to post

Looking really good. It would be cool if that GL_luminance effect was used for invincibility.

Anyway, glad to hear that things are going well on both the marriage front and the coding front!

Share this post


Link to post

Heh, it'd be cool to have that luminance texture format mixed with some kind of motion blur (like in Soldier of Fortune) for invincibility. Although the blur would get kinda annoying...

Share this post


Link to post

Random development related question:
Many of the other current Doom 3D source ports, such as JDoom, EDGE, and VaVoom, use "GL-Friendly" nodes to reduce the number of errors the 3D engine makes. Additionally, to compensate for the added bloat caused by its use of many Quake-like features, VaVoom uses something similar to Quake's vis (to reduce overdraw).

Are any such features likely to wind up in the next version of ZDoomGL, or is it intended that all these calculations will be performed in-game?

Share this post


Link to post
lemming said:

Random development related question:
Many of the other current Doom 3D source ports, such as JDoom, EDGE, and VaVoom, use "GL-Friendly" nodes to reduce the number of errors the 3D engine makes. Additionally, to compensate for the added bloat caused by its use of many Quake-like features, VaVoom uses something similar to Quake's vis (to reduce overdraw).

Are any such features likely to wind up in the next version of ZDoomGL, or is it intended that all these calculations will be performed in-game?


ZDoomGL requires GL-nodes now (and generates them for you if the wad doesn't have them). Believe you me, that made the floor/ceiling rendering about 1000x easier!

I'm still using the standard Doom visibility checking, but I plan on changing that to a more robust method when I get things "feature complete".

Share this post


Link to post
SlayeR said:

Heh, it'd be cool to have that luminance texture format mixed with some kind of motion blur (like in Soldier of Fortune) for invincibility. Although the blur would get kinda annoying...


The only problem with that is to change the texture format you have to purge the texture cache and regenenerate all the textures. On normal maps this isn't that big a deal, but for complex maps (like xtheateriii) it causes some stuttering when running around the map because it has to regenerate all the textures as they come into view.

Share this post


Link to post

Glad to hear that you're using GL-nodes. Seems to me that every port without that feature winds up putting holes in the floors and ceilings--where they don't belong!

The possibility that you might use another set of vis functions is cool, too. I'm definitely in favor of anything that speeds up drawing!

Can't wait to see how this whole thing turns out!

Share this post


Link to post
Guest
This topic is now closed to further replies.
×