20
Layers
Layers are a way of splitting up and organizing the objects in your scene.
In the 3D viewport.
You can select the currently visible layers in the footer of the 3D view or under properties panel > render layers tab > layer rollout > scene layer selector. (With any layer selector you can hold down Shift to select multiple layers.)
You can see ...
15
No, you're not doing anything wrong.
The Z-Pass is one of just a few, when not even the only pass that doesn't have a range from 0 to 1. It goes from almost Zero to infinity. Every pixels gets assigned a value that shows how far away it is compared to the camera.
If a pixel is 75 Blender Units away it gets assigned the value of 75.
If you output the Z-Pass ...
13
SSS works just like the 3 other "light type" sets do, so you add SSS direct and SSS indirect, multiply the result by SSS color, and sum that result with the other sets.
Shadow is not an image component per-se, since if you rendered your lamps with shadows enabled, they will already be embedded in your lighting passes. (the shadow pass already existed at the ...
9
A method to get believable light interaction with the environment, is to recreate the environment.
I have chosen this template image. The process will be the same for moving, tracked footage, but the tracking must be very accurate.
I chose this background image by Hans Weingartz - Self-photographed, CC BY-SA 2.0 de. Click to enlarge.
I have recreated the ...
8
Object ID passes
Assign a unique 'Pass Index' value to each object
Enable Object Index pass (Properties > Render Layers > Passes > Object Index)
Create a 'File Output' node in the compositor via Add > Output > File Output
Create 'ID Mask' nodes via Add > Converter > ID Mask per object
Create the desired output sockets to connect the 'ID Masks' and the '...
8
Multiple transparent planes with low Max Transparecy Bounces
The "Tree1" object is made by multiple overlapping planes. Their shading properties comes out from a sharp Mix of a Transparent shader and a Diffuse shader.
As there are multiple intesection and your Max Transparency Bounces is set to 8, light can't filter trough all the planes.
We can clearly ...
8
I assume your screenshot shows what you would see in the image editor if you have rendered cryptomatte passes, and view them directly. In that case, they look absolutely horrible:
However, that's not how they are supposed to be used. The mattes have to be extracted from the created channels first using the Cryptomatte node. You can find it under Add -> ...
7
You ask a question to which there is no simple answer based on your requirements:
no python - this limits any automation and implies a lot of work
with anti aliasing - this is something cycles cannot do yet on masks
The anti-aliasing issue of ID masks (first render, second without, last with):
The perfect solution would be to render the scene with each ...
answered Apr 18 '16 at 12:05
Jaroslav Jerryno Novotny
46.9k4 gold badges108 silver badges201 bronze badges
7
The Combined pass is useless in Cycles as far as I know, it doesn't change anything if it is enabled or not. In Blender Internal you can exclude from this pass other passes (like Shadow pass etc.) which might be useful, but not in Cycles.
If you are just after the normal, zdepth or mist passes, consider using some very fast rendering material on all objects ...
answered Mar 4 '18 at 15:03
Jaroslav Jerryno Novotny
46.9k4 gold badges108 silver badges201 bronze badges
7
When in doubt, use the Image pass.
Image
provides unmodified render result
output also comes denoised by the Internal denoiser if it's enabled
Noisy Image
provides a Despeckled image (useful when render has fireflies)
may cause bad shadow artifacts for low sample counts
Further Reading:
Denoise node in Blender Manual
How to use Intel denoiser in Blender
...
6
This is because the vector pass can't "see" through transparent (glass) objects; in this case all it sees is the lens object, which is remaining motionless relative to the camera.
To render with motion blur in this situation, you'll have to use Cycles' built in motion blur. Enable it in Properties > Render settings > Motion Blur:
And remove the vector blur ...
6
I'll try to explain. Here is an example.
We have an orange ball on the gray plane, lit from the side.
What contains the Diffuse Direct pass? A reflection of light coming from the lamp on the side. As you can see, the layer does not contain information about that the ball is orange.
To get the orange color you need to multiply Diffuse Direct and Diffuse ...
5
The File Output Node is exactly what you are looking for.
Unfortunately the documentation does not mention how to get it to output to different subfolders, but thanks to sambler you can have each file go into its own folder.
The settings for the File output node are in the properties region, press N to open it. Such things as file format, compression, ...
5
Actually, your question is a little bit vague, you have everything you need to begin to understand this topic on the Blender Reference Manual
Short answer, because this is a huge topic, and your question implies that the concept itself isn't quite clear for you (the reason I advised you to first read the manual ;) ), but the concept is to :
activate the ...
5
No, Z pass, ObjectID and alike are not dependent on Cycles render samples. Their visual output will always be the same like rendered with 1 sample - so not nice.
It's better to obtain them in Cycles with a different method - Z pass with Mist pass or Ray-depth shaders, ID-masks with random-colored shadeless materials for example.
Or get them from Blender ...
answered Aug 20 '15 at 15:16
Jaroslav Jerryno Novotny
46.9k4 gold badges108 silver badges201 bronze badges
5
Blender's depth map should work fine. I have tested it and I do get the right distance.
How do you load the OpenEXR files? Not all OpenEXR readers are reliable. I have noticed that OpenCV can give inaccurate results (see https://github.com/opencv/opencv/issues/10114).
5
Thinking ahead, you want the Output Path in sync with all the File Output Nodes of the compositor.
I think one easy way to solve this is adding a 'version number' to all output paths (as usual) and keep all file names in sync. Here is a quick'n dirty Add-on to address that. Once you add v and some arbitrary number to the Output Path, a new 'slider' appears ...
5
Don't use ID Masks
Just don't. They have little use, this is a really poorly designed feature. You simply cannot make a good mask with a 2bit color depth image.
Cryptomatte is the solution
Look into using Cryptomatte. You need to enable it's layers in the View Layer properties tab:
You can then get any masks easily using Cryptomatte node in the compositor ...
5
Workbench
It is not that easy :) and in some cases just impossible. If you want to just make parts darker or brighter, there is a change you just dont notice, but it would be there.
This "imperfection" comes from anti-aliased pixels generated on object edge to make them smooth. They consists partly from foreground color and partly from background ...
4
I've released YafaRay-E (experimental) v2.0.0, which includes Render Passes, Render Views and, among other things, there is a samplerate pass available that shows the places where the adaptative AA is sampling more pixels. It's not exactly what you showed here, but I hope it's good enough.
You can download it from here:
http://www.yafaray.org/community/...
4
You can do this by using a Material Override and a special material to give the objects' positions.
First create a new material with an emission shader (you can add it to any object, it doesn't matter). Add an Input > Object Info node and plug the Location into the color of the emission shader.
Now you can create a fake user for the material and remove it ...
4
As 3dpointedit pointed out, there is a limitation of the current ID mask system:
Is Z-Render Pass accuracy dependent on sample count? Are others like object id? (Cycles)
As such, these layers are unusable for any post-processing workflow. You need to obtain these layers differently, or build the pipeline in another way.
The correct way of compositing onto ...
answered Oct 3 '17 at 13:39
Jaroslav Jerryno Novotny
46.9k4 gold badges108 silver badges201 bronze badges
4
It isn't straightforwards as Cegaton notes.
You could abuse Blender's ridiculous OCIO integration to achieve this however. Save the file as an EXR, and add the following stanza to the OpenColorIO configuration in datafiles/colormanagement:
- !<ColorSpace>
name: One-Way XYZ
family: Input Special
equalitygroup: ""
bitdepth: 32f
...
4
According to the manual,
the depth map is inverse linear (1/distance) from the camera clip start.
so
it depends on the "Clip start" camera setting
it shouldn't be read straightforwardly as a distance map, but the distances should be easy to work out
From what is written, it would appear that
$$Z_{\rm map} = \frac{A}{(d-{\rm clip_{start}})}\quad\...
4
Try setting Display Device to 'None'(because you do not want it to be an image viewd on any display device, but only pure unchanged values) in the scene's color management settings:
Seems to work:
Obviously the output needs to be a 32bit full float EXR image in this case as well if you want to save it.
4
Yes, you need to use the alpha channel as a mask since transparency is not taken into consideration in the depth pass.
This is how I have done it:
I used the node called "Set alpha".
I Plugged the depth into te "image" and the alpha pass into the "Factor"
the normalize is optional but helps to see the depth better.
Only top voted, non community-wiki answers of a minimum length are eligible
