Transforming objects vs. Moving the camera

General TouchDesigner Discussion
1

I don’t completely understand the steps/processes a video card goes through when rendering stuff in TD

I read on here somewhere about how you should move geometry objects instead of sops because it took less processing power. What I’m wondering is how this relates to using the camera to create movement instead of the actual objects (or moving them as little as possible). Is it better to have an object that you are creating or transforming in real time remain stationary while moving the camera away from it?

I’m also wondering exactly why it makes a difference again as far as geometry objects vs. sop transformation. I’m messing around with some different tunnel like effects based on a grid system and feel like there are a lot of different ways I could go about creating the perception of movement.

2

When doing transforms in SOPs like with the Transform SOP, every point needs to be transformed on the CPU so all the correct point information is available for the next SOP to operate on. This can be a very expensive operation.

When doing transforms using COMP parameters then those transformation are done on the GPU, and are essentially free.
There is no difference between moving geometry using the transforms in a Geometry COMP or moving it using the transforms in a Camera COMP. Both are essentially free.

3

Thanks for the reply. Makes a little more sense now