Here is some shooting from my fire-enhanced chain shotgun:
Here's a giant, electric-chaingun-shooting zombie with 1800HP:

And here is some test talking with rather foul language from a zombie.
I dunno if anyone really likes it or not.




Dvorak wrote:My pet project is a graphics engine.![]()
Or let's better say: it still is an algorithmic study, that continues the work I started together with kyra (who is a llamasoftie, too) back in 1992. What we had in mind was to render huge worlds.
Extremely huge worlds. Worlds that could be walked and discovered in real-time.
Imagine standing on the top of a mountain in the Himalayas looking over piles of billions of Saint Peter's Domes stacked up. Well, something like that.
Reading French and Belgian underground comics of the 70s surely helps in understanding what I'm talking about here...![]()
However, the available tech to that time layed some constraints on our efforts (We started with a 486 with a three megabyte graphics card, an Orchid P something, yes: three megs. 1 for DOS and 2 for Windows. I always assumed a strange engineering desaster behind this.) Nonetheless we designed a bunch of very different engines and fully implemented at least a dozen of them, IIRC. All of them circling around the main theme voxels, octrees and raytracing. Oh, and none of them worked: either they were much too slow, looked shite, or both. Most of the times both. We gave up.
Fast forward several years. Doing business IT now for a living and bored to death by it, I started again with a new engine some time ago.
Working title is "Aquamarine" (continuing an old tradition of using gemstone names for our engines).
Basically, the new engine is designed to consist of three components:
- a probabilistic octree raytracer
- a render cache
- a LOD system
The first two components are done and work nicely. Right now I'm working on the third.
If I get the LOD working as expected I'll have an engine that'll be capable of rendering incredibly huge worlds in realtime. If not I'll have nothing.
But yeah, why should machine No. 16 (or whatever it is) work after all those failures? I'm used to sorrow by now...![]()
Two screenies of the actual machine (without any LOD system):
Next steps:
- working LOD system
- replacing the textured cubes I use as stubs by something looking a bit more impressive.
Ach, still a long way to go, but that's not the point.
Finally I'm working again on something that's dear to my heart.
Feels good.