Vector + Zdepth


I would like to know if i’m right with my nodes to composite my picture.

I don’t know how to combine vector and zdepth.

Thank you for your help :wink:

Hum, not sure of my test.

As a rule Of thumb: first defocus, then motion blur. Results however may in many cases be not good, especially when defocus is strong. Please mind that “vector blur” will use not defocused data to do its magic and there’s no way around this.

Ok, so what is the good way to have good results ?

Hey, she’ll love it anyway. :smiley:

But … “today’s the 13th, boy!” :eek:


I know, t’s tomorrow !

The best result I’ve had is to apply the motion blur first on the image as well as on the z pass and then apply the dof with the motion burred z pass.

However, I don’t know how to include the shadows into the speed pass.

Interresting !

Maybe with a shadow pass.

The shadow pass is just a separate pass of the same shadow. I’m not sure that Blender can handle to include it to a speed pass (except maybe if it’s possible the have the speed data in a shader node?)
The best solution is probably do the motion blur directly while the render and just play with the dof in the compositor.

Can you apply the vector blur to the shadow pass before multiplying it onto the vector blurred base image?

EDIT , silly me I forgot that the point is that the vector of an object isn’t translated to the surface carrying the shadow. Could you have a separate scene with real mo blur turned on for just the shadow pass? Everything else would be turned off though.