Other (?) info stored in surface names.

More
18 years 5 months ago #14353 by Shane
Ah... okay. Then my tests are premature. If I can be of any help there (or by doing anything else), please let me know.

Well, I've had a bit of a rest, so I'm back to it. See if I can figure out where I went wrong. ;)

Please Log in or Create an account to join the conversation.

More
18 years 5 months ago #14405 by Second Chance
Since Shane doesn't have his old test ball anymore and is busy helping test Jasper's converter, I thought I'd make one to test my stuff with and take some pictures. I thought I would stick these here, since we're talking about surface information and maps.

Now that I've actually got Lightwave 5, I've learned a few things about the unusual way in which it handles the specular and luminosity surface maps for EoC.

For specularity, the maps work as expected. With the map completely overriding the specularity setting, and the gradations from 100% black to 100% white representing 255 level of specularity; from totally dull to highly specular and everywhere in between. (I say 255 levels of specularity because 100% black is non-specular.) Sadly, the old LW doesn't have a map channel (or even a numeric setting) for glossiness. It uses only a pull-down selector with Low, Medium, High and Maximum as the only options. But, it doesn't even matter, because those settings don't affect the surface's glossiness in flux at all when using a specular map. (But I do believe these settings work when not using a specular map, just the specular level. At least they do in newer LW exported to 5.6.) As you can see in the first big picture below; the specular mapped surface on the left had Glossiness set to Maximum (not high) and the one on the right had it set to Low. Not only is there no visible difference between the two, you can't even see the seam in the middle where the two meet.

For Luminosity in EoC, LW 5 does something I've never seen in all my years in 3D art. Usually, luminosity (also called self-illumination or glow) maps are handled in the same way every other map that isn't diffuse color, specular color or ambient color are handled; as a grayscale alpha-map, with areas of 100% black being transparent and representing 0% effect and areas of 100% white being solid and representing 100% effect. So for a typical glow map, white areas would cause the pixel color at that spot on the diffuse color map to be rendered at 100% saturation with no specular highlights or ambient shadows. In other words, simulated self-illumination. Every 3D package I've ever worked with or seen does things this way. Even Lightwave itself works this way, for normal in-package renders. But when rendered in the flux engine; the luminosity map, rather than affecting the pixel color on the diffuse map, acts more like an overlay or decal. Replacing the diffuse map colors (or mixing them) with it's own colors where the self-illumination occurs. For instance, look at the first large picture below. I used a black and white map in the luminosity channel of the surface on the middle right side. What should have happened (and what does happen in the Lightwave render), is that the grid should have shown up as bright pale gray (based on the color of the diffuse map), not pure white. But flux is actually overlaying the luminosity map on top of the diffuse map, and using it's own color instead.

Now look at the left side. The colored grid is not part of the diffuse texture. It is a colored grid on a black background used in the luminosity channel of the surface. The actual map is below the big pic.

As you can see; black is still considered totally transparent, but gradations into color, any color, cause that color to show up as a self-illuminated area on top of the diffuse surface.

The bottom left surface was supposed to show a transparancy map, but I messed it up somehow and didn't feel like screwing with it when I took these pictures. The bottom right is supposed to be for testing reflection mapping, but nobody knows how that works yet. And when I ran the ball through the converter with a map in the reflection channel, it wouldn't produce any LBM textures with the pso. So I left it blank for now.



The colored luminosity map.


Ok. So now that we've seen how luminosity and specularity maps work. Here's how they work with gradient fades to black. As you can see in both examples, 100% black is completely transparent and gradients to white (for specular) or color (for luminosity) are smooth and clean.

I hope this is useful to anybody.





mailto:second_chance@cox.net
The Ultimate Guide To Modding: I-War 2 - Edge Of Chaos
.

Please Log in or Create an account to join the conversation.

More
18 years 5 months ago #14414 by GrandpaTrout
SecondChance,

Given what you now understand about Luminosity maps, would you care to conjecture on how planets are setup?

The behavior seems very similar.

Please Log in or Create an account to join the conversation.

More
18 years 5 months ago #14418 by Second Chance
Hmm. That's a toughie. Being quasi-procedural, they seem to rely more heavily on the graphics code built into flux than on external texturing methods. I have a bunch of rules and factoids about how planets and their textures behave, but they don't add up to an answer about how they work internally.

I'd be glad to get in on working it out, but it's a pretty big subject by itself and probably deserves it's own new thread (start one and I'll be there). Then, we can all put what we know together there and maybe figure it out. I know someone, cambrogol maybe, did some great work with new planets and textures. There's probably much more information about them now than there was last year. And it's going to require someone with knowledge of coding and INI files to sit in. Then if we can decode a planet model with Jasper's converter (I haven't figured out how this works yet), that should be enough to figure it out.

An interesting item I noticed recently about planets is that they don't have a setup or avatar scene associated with the model. That means that the limit of what is done in Lightwave is done in Modeler. Meaning that a surface has been assigned to the polygons and named, surface attributes (like diffuse level, specularity and glosiness) are set (maybe) and smoothing angle is determined (again maybe, these two can also be done through code). That's it. No texture maps or even mapping methods are assigned. That tells me that everything hinges on the surface name and the code it activates.

Before I take up too much more room; is anyone else interested in contributing to a new thread to compile all the information known about planets?

mailto:second_chance@cox.net
The Ultimate Guide To Modding: I-War 2 - Edge Of Chaos
.

Please Log in or Create an account to join the conversation.

More
18 years 5 months ago #14419 by Shane
I'd be interested. ;)

Please Log in or Create an account to join the conversation.