View Full Version : Negative values for materials and color...

08-10-2002, 09:57 PM
Since the RGBA components of materials and colors are not clipped to the positive range, I am wondering what kind of useful things you guys have found for negative values here ... Just a thought ... Anyone experimented with this before?

08-12-2002, 06:34 AM
bump ... http://www.opengl.org/discussion_boards/ubb/frown.gif

08-12-2002, 06:49 AM
Originally posted by Iceman:
bump ... http://www.opengl.org/discussion_boards/ubb/frown.gif
...mapping http://www.opengl.org/discussion_boards/ubb/rolleyes.gif ??


If I remember correclty when I started coding OpenGL some time ago, when I set the lighting colors to negative values it seemed as the lighting was done "inverse" that means the specular highlights were (1- specular color) and the diffuse light was black. But its to long ago to tell more details. Perhaps you can write a simple test app ?


08-12-2002, 07:11 PM
I did that earlier. I observed that dropping the numbers below 0 had a noticable effect down to about -0.3 to -0.4. After that I noticed no difference. I did this for amb/dif/spec/emi material properties and noticed that if i dropped 2 of the channels way down and set the other (non-alpha) components to a positive value it seemed to add to the material glow.

I was wondering if anyone else had experimented with this and found some cool uses for it ... hence the post.

If you (or anyone else) is interested, I can post the app on our website at work. It is GLUT based, and has a semi-crude interface. You select from Shininess/Amb/Dif/Spec/Emi by using the up and down arrow keys, you select from the subcomponents (R/G/B/A in the case of the reflectances) by using the left and right arrows, and add to or subtract from the selected component/subcomponent with the +/- keys. I displays a grid and several quadric objects. The sphere on the origin has the materials that are being interactively altered.

08-12-2002, 11:43 PM
Can't think of any cool effects, as the final vertex colours (post-lighting calculations) are clamped to (0,1) range before triangle interpolation...so where/why the 'bump maps' suggestion?!

08-13-2002, 02:09 AM
Originally posted by knackered:
...so where/why the 'bump maps' ...


Well it should be a joke because Iceman posted "bump".I know that there is no correlation to bumpmapping.


08-13-2002, 02:50 AM
3d-artists that don't have the computational power to do global illumination sometimes fake indirect lighting by using negative lights that remove light instead of adding it. It can be a useful tool to add atmosphere to a scene or to fake soft shadows cheaply.

08-13-2002, 05:32 PM
Nope, no relationship to bump mapping, this thread had sat for a day or two without a response so I 'bumped' it up to the top as a last ditch effort to see if anyone that may be interested had not seen it ...

Anyhow ... if anyone is interested in tinkering with the test app ... I posted the executable here ... (614Kb)


Here are the controls ...

s - Stops the motion of the lights
F1 - toggles full screen
F12 - Screen shot to JPEG

You can change the five material properties of the center sphere ... Shininess/Ambient/Diffuse/Specular/Emmissive. You select the property to be altered with the up and down arrows.

You can select the R, G, B, or A channels for the reflectances with the left and right arrows.

All properties are displayed on screen in text form also so you can see the negative numbers and their effects ... Just a toy ... enjoy ...

[This message has been edited by Iceman (edited 08-13-2002).]