Any "standard" lighting for testing shader?


#1

Hi all.
Every time I play with the shaders. Two questions always in my head:
1.What kind of lighting conditions is suitable for testing materials/shaders? HDR? 3 point lighing? studio lighting? dome light? …
2.And what kind of shapes should the test scene has? A simple sphere seems not enough.

Plus, I see ppl put some “100%white” “50%white” in the scene. What’s that used for?

I’m a Maya user. Any suggestion is welcome. Thx :slight_smile:


#2

question, i would like to hear from seasoned pros too.
about lighting style, my personal take would be just a 3 point lighting with white lights. This way I know there are no coloured lights in the scenes.

i dont know about whites and greys, but in photography, if i am not mistaken, the light meteor tries to get an exposure of 17% grey, which means, if you aim your light meteor at a white surface and use the readings, the resulting image should be 17% grey. Applying this to 3d, you can use this as a guideline to ensure a proper exposure.


#3

Thx your reply.
How to apply it to 3d? I don’t have a light meter and don’t know what is 17% grey look like.


#4

The test lighting for a shader should be representative of the type of lighting that will be used in the production. For example, if you have HDRI environments in a lot of shots, you should have one when testing the shaders. If you use a lot of really bright rim light, you should test the shader with it. Same goes for whether or not it’ll have raytraced self-reflection, shadow maps or raytraced shadows, etc. If you know what a typical background plate will be where the shader will appear, then run the tests with that as your background.

Shader/texture tests are usually done on the actual model, after the model is approved. If the model isn’t built yet at least try to use a stand-in that’s as close as possible.

Talking about 50% gray and so on could be done in different contexts, I won’t post any guesses about what the person meant who said that.

-jeremy


#5

Edit - oops didn’t read the whole post. Using Maya I think you can achieve similar results using import, in the case of the camera and lighting rigs.

This is a method I use when using a Renderman renderer (such as Pixie), which is %98 of the time. I have a master RIB that contains all the shading controls, shader assignments and final frame ouput (file or framebuffer). I use ReadArchive commands in various places that alter what object, camera or lighting rig I use which are of course seperate RIB files containing the data needed. The lighting archive collection are just different lights in different locations and rotations, different colors and whatnot. The master shader test RIB is quite busy with commented out text but when using Cutter to do shader tests this allows for quick renders.

Not much of a description of what light rig I use I am afraid but it’s somewhat related, in case you were interested.


#6

jeremybirn. I thought I could use just one model to test all kinds of shaders before read your post.
neuromancer1978. I don’t use renderman. :slight_smile:


#7

How about build a number of light rigs first, then just build a tool so you can press a button to reference these rigs in.
I normally just have 3 teapots at the ready to test my shaders on !

B


#8

Teapot~! How could I forget!


#9

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.