[Sam Treadway]

[Pierre Tessier] ” …probably because of AO. We can see its effect when the scene is rendered. Not good.”
With respect to AO this shouldn’t happen. It may be something else. Check the AO settings in the render settings and see if maybe AO cache is enabled. If so, then disable and try again.

That would be the quickest fix:
Render a pass with the object and a pass without it and dissolve between them in a compositing app or NLE.
You only need to render the frames where the object will disappear twice so if the render range is 0 to 100 and the transition is 30 to 60 then render one pass with the object from 0 to 60 and the pass without the object from 30 to 100….or just render the entire range with and without the object for more flexibility.

If there are a lot of technical issues to consider then you may want to setup AOV’s etc. and add a position pass effect to the render settings.

[Pierre Tessier] “Also, when an object already has the alpha channel activated, you can’t use that technique. So, I was wondering if there were another way to make an object slowly disappear.”
You can use a layer shader and animate a solid color blending from 0 to 100% on top.

[Sam Treadway]

Use a layer shader and load in two textures/shaders as layers and animate the blend by setting keys for the entire layer shader while inside that view.

[Sam Treadway]

Set the “Intermediate Points” value of the spline you drew to “Uniform” and adjust the number and see what happens.

[Sam Treadway]

The sub-d object in C4D has a viewport subdivision level and a separate render subdivision level. If you are worried about performance then create the object first without any children and set the viewport level to 0 or 1. 0 has no effect on the model in the viewport of course. At level 1 it should still be manageable and you should have enough information about how the edges are effecting the subdivided version. When it comes time to export, then turn off the dub-d object and export the model without any subdivision. In the other package (if possible) do the same as before with subdivision levels to keep it non-destructive and allow for toggling on/off.

I’d avoid ngons altogether because it is just bad practice and won’t work in a larger collaborative environment.

I’d still find a way to pay for the good courses instead of just going after the free crums.

Is there a particular reason you wish to export the model for use in Keyshot? A feature that is only available in that app?

[Sam Treadway]

Do yourself a huge favor and skip over YouTube. There are some good tutorials to be found there and other places but it’s mixed in with a lot of waste.

When you are keeping everything within a single application then a workflow that is most efficient for that particular application is great but it is always best to be cognizant of edge flow and taking the extra time to model it right if you plan on exporting to another package.

What I see outside of those beveled circle indents are massive ngons. Any subdivision applied to the model afterward will cause pinching and overlap because of the concave shape.

Try one of these “Making it Look Great” courses:
https://motionworks.net/?s=making+it+look+great

MILG 10 or 11 will give you good guidance for hard surface modeling with proper edge flow and demonstrate techniques to get around tricky situations like the one you have in your model.

[Sam Treadway]

There are many ways to achieve this. Here are two of the simplest:

1. You can select all the materials, set a key on the color attribute (even though the color shown is the last selected it will key all materials as is). Then go to the time where you want them to all be orange and while all materials are still selected, change the color and set a new key.

2. Instead of keying the base color of each material you can assign a color shader to the texture within the color channel. With all materials selected click the triangle button next the the texture field and select color. Make it orange. Now with all materials selected you can simply key the mix strength of the texture from 0 to 100%.

3. User the node-based materials and a mix node.

4. From your description using xPresso there is a part of the setup missing. You have a RGB color value to start with and then you want to end with another but the starting point is different for each material. The first step would be to subtract the starting values over time while adding the new values over time.

[Sam Treadway]

You must do a 2D track of the footage first as that data is needed to solve a 3d camera.
The 3D solve will stop when there are no useful tracking points left to work with.

The 3D solve should work but it does need 2D tracking data first. The 3D solve works by measuring the change of individual 2d tracking points frame by frame to determine surface information (parallax and spread) and then is able to generate camera vectors by averaging and negating the 2d track results. If there are not enough 2d tracking points per frame, the 3D solve will fail when it drops below a set threshold and assume the job is done.

Depending on the footage you are solving for it’s usually best to flood a bunch of tracking points from the start, auto track, then scrub through and delete the points you don’t want to evaluate in the 3d solve.

If you are talking about object tracking. That’s a different setup and you will need to correctly solve the camera motion track first.

[Sam Treadway]

See that button on the 2d tracking tab that reads: “Create Auto Tracks”.
Don’t press it! You can but you’ll have to create them throughout the clip.
Instead just go straight to the “< Auto Track >” button below it.

[Sam Treadway]

There is the “Make Preview” command which lets you choose from full render or hardware openGL in R21. In Maya this is referred to as a “playblast”.
In r21 the command is in the Animate menu. In earlier versions it may be located in the Render menu.

Hardware OpenGL is the quickest but will show just as it does in the viewport so you may want to turn off splines and other distractions and turn on Enhanced OpenGL features before making the preview render.

Another option is to duplicate your current render settings (bottom left of render settings dialog) and make that duplicate a child of the original and name it preview. Change the render engine on the child to hardware openGL and under the openGL category turn on/off the enhanced features to balance speed and readability. Once you find settings that you like then right-click on the setting name and choose “Save Preset…”

[Sam Treadway]

I’m assuming you’re using the standard renderer. If not then make a copy of your render settings and switch it to standard renderer.
Turn on material override in the render settings.
Create the most generic material…in fact, you can just turn all the channels off.
Add that new material to the material override of the render settings.
Under the multipass group, remove or turn off everything but “Post Effects”
Make sure the position pass effect is added to the render settings.
Now under Save, turn off single pass save and turn on the multi-pass save option and set your separate path.
Under Antialiasing, set it to “none”. Why? Because the position pass doesn’t use it but it will slow down the render.
Under “Options”, turn off everything but displacement if you need that and of course leave post effects on.

The position pass should now render out incredibly fast.