Hello.
I’m doing a project where we composite 3d elements into live action footage.
by doing this we will need to render out different passes, one of the passes that we use is matte shadow
our rendering enginee will be Vray 1.5.
So my problem is that I need some advice for the matte shadow pass. I can’t find any tutorials on the net about how to deal with the matteshadow pass with vray.
Any advice? Thanks!
Can you be more specific about what you mean by the phrase, “deal with?”
Specifically, what problems are you having now? Or, what problems do you anticipate that you might have? The more details you give, the more helpful this community can be. Try to present “a solvable problem” to us.
My point is that is there any way to adjust the matte shadow pass in nuke?
Because I used VrayMTLwrapper and rendered it out with several passes INCLUDED matte shadow pass. My problem is that I cannot adjust the shadow pass WITHOUT affecting the alpha to the ELEMENT that cause the shadow.
If you still don’t get it. The simplest question; how can I render out passes with alpha without the shadown in Vray?
So its a vray issue you’re having, and not a comp issue? Shouldn’t the post be moved to the correct forum?
Also, have you tried to render the shadows seperate instead? Render one pass with the ball “cast shadows off” and one with it on, and turn the “visible to camera” off.
-theo
What Theo said,
make sure you render one pass with only the ball and one pass with only the shadows.
Yeah, That might help, But you get the shadow pass WITHOUT having the geometry
and vice versa?
Because i think it doesn’t help if you only hiding the geometry while trying to render out geometry’s shadow.
Could you explain it more detailed?
What do you mean by render a mask for object and invert it?
The problem is when I render all passes out and my alpha contains the shadow of the geometry that I don’t want to bring, because if I do that later in compositing program, for instance I want to smooth out the shadow it will affect the geometry too, vice versa.
Vray has several passes to create masks. You can use it to limit any operation in compositing.
What is the nature of your input file, right now? Is it “an image,” as in “RGB+A?” Or are the various rendering work-products stored distinctly (as they might be, say, in an OpenEXR file?).
If everything has been reduced to “an image,” then there’s really not much that (I know of that) you can do: vital information is irretrievably gone. Everything has been mashed together into pixels, but you no longer have any way to know why those pixel values are there.
What you need to have, in your original source-data going into your pipeline, is that “everything is separate.” For instance, you might have color, specular, alpha, normal i[/i] channels just for “the ball.” Then, separately(!), channels just for the density of shadows that are being cast by that ball. (Perhaps a separate channel for each light.)
A rendering sequence has a lot of separate channels of data flowing through it that must have been produced separately, and that have remained separate throughout. They are distinguished, not by their (RGBA) characteristics, but rather by their purpose. “This is shadow #X of object #Y,” and so on. The very last steps are “to combine them.”
If your original render files included that information with sufficient separation of data, then you are fine. If not, you will have to re-render.
Ya, I know that. I’m using OpenEXR too. But I don’t know how you can render alpha pass only for the geometry in the scene WITHOUT bringing the shadow in.
The only solution now me and my friend found is that we can render out Vray_Wireframe color
and change the geometry color to black and the rest you don’t want in alpha pass into WHITE
so that you can “cheat” alpha pass.
Dude,right click objects and de select “cast shadows”. That should do the trick. Unless you’re doing lots of indirects…
Agree.
You should find that each of the different work-products (shadows, etc.) can be separated out into distinct channels. At the very least it should be possible to omit them. But if your product is reasonably up to date, the capacity for separation surely is there… somewhere.
For example, in the Blender system (which I use), it’s called “Render Passes” and this page provides an extremely good visual illustration of exactly what I’m talking about. (Look at all of the different “wires” that are proceeding out of the left-most box.)
It is possible to be very selective about exactly what objects appear in that box’s output. (See this page if you happen to be interested in how Blender does it.) The key idea: you can select things, in effect, based on why they are there.
I haven’t used VRay (yet) so I can’t comment on it further. But this is basically what you are shooting for …
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.