Testing Surfaces in EoC

A few people are now starting to model ships for EoC, so I thought I might post some results on just what the engine can handle. Some things we've (I've) been told are true, some are not.

The Testball:

I created a simple sphere 85 m in diameter, divided it into 13 surfaces, and then applied a different surface attribute to each section. Here's the breakdown of surfaces:


SURFACE EFFECT

001 chrome
002 tinted glass
003 bump map
004 diffusion map
005 specularity map
006 LW luminosity
007 EoC luminosity channel
008 Secondary textures (fractical over standard)
009 Percentage of planar map dissolve
010 Repeating texture
011 Smoothing 50%
012 Smoothing 100%
Default Standard grey surface, no attributes.

And here's a pic of what the ball looks like when rendered with Lightwave (which we know does the surfaces right).

http://images.snapfish.com/3396492%3B23232%7Ffp58%3Dot%3E2328%3D%3C2%3C%3D8%3A8%3DXROQDF%3E232348%3A44865%3Aot1lsi

Then the item was translated into EoC. To get clear lighting, I took it to the nearest star (Santa Romera II I believe). This is what it looked like:

http://images.snapfish.com/3396492%3B23232%7Ffp54%3Dot%3E2328%3D%3C2%3C%3D8%3A8%3DXROQDF%3E232348%3A44%3A%3A%3C7ot1lsi


Results:

001 Chrome Surfacing: FAILED. The engine cannot create a reflection map for the surface. It only appears glossy. The reflection map for this surface did not appear at all.

002 Tinted Glass (though this glass is nearly clear): PASSED. My settings made it a little too dark, but, like the previous tests with the transparent stations, the surface works.

003 Bump Map: FAILED. The engine will not display the bump map at all. (When I earlier believed bump maps did work in EoC I was in error. Damn engine fooled me again.)

004 Diffusion Map: FAILED. The engine only displays the texture as a standard texture, not a diffusion map. If it had shown it correctly, there would have been no green in the 'scrubbie pad' texture I used.

005 Specularity Map: FAILED. Same as bump maps. The engine displays the map as if it were a standard texture map. There is no variation in the specularity applied to the surface... the entire surface receives a global spec value.

006 Lightwave Luminosity (set to 100%). FAILED. The surface does not maintain any luminosity attributes. (I really didn't think it would... the SDK says it won't, but I needed another test to fill out the testball and so decided to try it out.)

007 Glow Channel. PASSED. Standard Flux Glow Channel... Like 006, this was placed in just to fill out the testball segements.

008 Secondary Texture Layer. FAILED. Used standard texture map for a ship's hull, and then overlaid a secondary texture map (fractal noise) at 50% opacity. Neither texture appeared in game.

009 Map Opacity. FAILED. Used standard texture map for a ship's hull and set opacity to 50%. Opacity value was ignored by Flux, which indicates opacity values do not work on projected textures (like UV maps), but only on procedural textures.

010 Repeating Texture. FAILED. Repeating textures will not repeat. Only one instance of the texture was created by Flux.

011 Smoothing 50%. PASSED. Texture did smooth, but did not carry across to adjoining polys without smoothing like in the LW render. You can see it only smoothed about half the surface. Though it did join nicely with the next texture.

012 Smoothing 100%. PASSED. Though it is difficult to say if there is any variation between texture 011 and 012. In all the screenshots the surfaces look identical.

Default Surface: (The circle at the poles of the testball.). PASSED (of course).

So, basically, it seems we're stuck with the basic texture map, procedurals which are transparent, and avatar channels.


A myth: I was informed that textures over 512x512 would not display in the game. This is incorrect. The top and bottom texture maps for the frigate are both 1024x1024, and work just fine.

Hope some of this helps somebody. More results as I discover them (over time).

<font size="1"><font face="Book Antiqua"><font color="black">"Never before in the history of the world had such a mass of human beings moved and suffered together. This was no disciplined march; it was a stampede-- without order and without a goal, six million people unarmed and unprovisioned, driving headlong. It was the beginning of the rout of Civilisation... of the massacre of Mankind."
--H. G. Wells The War Of The Worlds</font id="black"></font id="Book Antiqua"> </font id="size1">

Cool! nothing like real data. I noticed something odd with transparent textures, but I am not sure I can explain it. If you visit a mine, it has a shiny reflective dome. You can see the systems sun reflect off the dome. I thought it might reflect any light source - so I moved very close to examine it for running light reflections - none. I tried firing at it to look for weapon light reflections - none.

Perhaps some textures reflect the system suns - but no other light source? Any thoughts?

-Gtrout

As I understand it, there are no other "real" light sources other than the suns (which are simply intense point lights). So the reflection of the suns in the dome is a specularity bloom from a LW point light. The same as having a high-gloss surface in Layout and a light source creating a bloom upon it. As you rotate the object, the bright spot moves.

But this is not real reflection... it's just a property of specularity/glossiness. For real reflections, LW needs to be told what to reflect (it's not very smart). Thus, the ability to create a reflection map in LW.

In the LW render of the testball above, the chrome surface reflects the lights in the scene. But I did not give it a reflection map, so it only gets bright spots from nearby lights.

Nebula and weapons fire somehow create glows which can affect everything within a certain radius, but I've not dug into that deeply enough to find out how they do it. I can speculate: if I look closely enough at the antimatter bolt avatar, I bet I'd find a light embedded (under the texture) which is causing the flash.

In which case, it should show up in the dome... []

I've run myself in circles.

I have no idea why it does not show.





<font size="1"><font face="Book Antiqua"><font color="black">"Never before in the history of the world had such a mass of human beings moved and suffered together. This was no disciplined march; it was a stampede-- without order and without a goal, six million people unarmed and unprovisioned, driving headlong. It was the beginning of the rout of Civilisation... of the massacre of Mankind."
--H. G. Wells The War Of The Worlds</font id="black"></font id="Book Antiqua"> </font id="size1">

Hmmm... another inconsistancy.

I had problems with the Terrapin tonight. The hangar screen did fine, but, when loading (just before the launch sequence starts to play... at the end of the progress bar the program would freeze. Locked down solid.

I tracked the problem to the collision hull (I had been using the puffin collision hull for all testing before tonight. The Terrapin had two nonplanar polygons, a double point poly, and two single point polys (which crept in on me somehow... sloppy! .

I removed the double and single point polys, but the game still hung at the end of the loading sequence.

I tripled the nonplaner polys to make them normal polys. No more problem. The Terrapin launches fine.

The moddeling tips page give this advice about collision hulls:

Collision hull objects may be built from nonplanar polygons and should be very low poly.

Perhaps this is not the case. [;] It certainly deep-sixed my evening.

<font size="1"><font face="Book Antiqua"><font color="black">"Never before in the history of the world had such a mass of human beings moved and suffered together. This was no disciplined march; it was a stampede-- without order and without a goal, six million people unarmed and unprovisioned, driving headlong. It was the beginning of the rout of Civilisation... of the massacre of Mankind."
--H. G. Wells The War Of The Worlds</font id="black"></font id="Book Antiqua"> </font id="size1">