I'd like to do arbitrary vector math with the camera position, shading point, and possibly other points. As a test case (blend file), suppose I want to make a shader that will draw a simple gradient based on distance from the shading point to the camera in one of two ways.

The two methods of getting the desired distance are:

  1. "Camera Data.View Distance" as the control.

  2. Computing the length of a vector equal to the unit vector given by "Camera Data.View Vector" multiplied by the scalar "Camera Data.View Distance". This is the issue.

The node setup looks like this:

enter image description here

What can I put in the "Vec * Scalar?" frame so that the material will look the same no matter how the "Selector" Mix Shader Factor is set?


2 Answers 2


You should think more about the data types. If you want vector*scalar, use MixRGB set to Multiply (an RGB color has three components, like a vector), and not a simple scalar math node. See my answer to this question for an example of vector*scalar: Scale Texture with a Value-/Math-Node

Also your selector seems to be a Mix Shader (I didn't download your file, but it has shader inputs and outputs), and you should not put scalars into a node that expects shaders.

EDIT Just for completeness: There are also other ways of doing vector*scalar, like using a vector Mapping node and setting the same X, Y, Z Scale, or separating the vector components with a Separate XYZ and multiplying the three scalar coordinates individually, and merging them again with Combine XYZ.

  • 1
    $\begingroup$ MixRGB only does vectorscalar in the 0-1 range, otherwise you have to use vectorvector. It also doesn't accept negative constants. Mapping is only vector*vector constant. Separate XYZ seems to be the best solution; just stick it in a group and copy it around. $\endgroup$
    – Fax
    Commented Jul 1, 2018 at 20:42

If you don't mind using Open Shading Language, it's pretty easy to do make a script node that does vector component multiplication:

shader osl_vector_multiply(
    vector InVector1 = 1,
    vector InVector2 = 1,
    float InScalar = 1,
    output vector OutVector = 1)
    OutVector = InVector1 * InVector2 * InScalar;

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .