I have noticed that the Camera collision follows different rules on imported objects than other physics objects. Firstly, when I import a model from Tinkercad my camera collides with it even if the physics property is disabled (as you can see in the first circle maze I enter, the ball has phased through the wall but my camera cannot). If I enable physics on my imported model, it only uses the bounding box for collision with other physics enabled objects in my scene (hence the hovering ball) and yet my camera’s collision has an understanding of the individual walls available and restricts my movement.
The easiest demonstration of this is the ramp which I show off first. I have a ball I’d like to roll down the ramp but the ball only respects the bounding box around the ramp. Meanwhile my camera is able to move up and down the ramp just fine. Is there a way to make physics objects respect the same physics colliders as the camera for imported objects?
For reference, the reason I would like to have better implementation of physics on imported objects is so that I can create a lesson plan with students where they model their own solution to a physics problem that can be simulated in CoSpaces, for example designing a tool that can sort falling objects, or designing the best performing ramp for a car jumping a gap.
1 Like
Hi @Gregory_Osborne can you provide us with a remixable link to this CoSpace so we can take a look for you? Thanks!
Hi @Gregory_Osborne 
In CoSpaces there are two different collision systems: the camera collision system and the physics engine. When you enable physics in the item inspector, it makes the item interact with the physics engine, but it doesn’t affect camera collisions.
You can’t make physics objects respect the same colliders as the camera for imported objects but you can create your own collider for any imported 3D model.
Here’s how you can achieve this:
- Create a collider using CoSpaces building blocks (e.g., boxes, cylinders, capsules, etc.).
- Enable physics for each building block.
- Group the building blocks together.
- Make the collider group invisible by setting its opacity to 0.
- Attach your 3D model (with physics disabled) to the collider.
This method gives you an invisible physics collider with your visible 3D model attached.
Here’s an example of how to do it: CoSpaces Edu :: Remix of CircleMaze
Hope this helps!
Is there no way for me to make the item use the camera collision system?
One of the reasons I want this is because CoSpaces is not designed for creating an exporting 3D models, but Tinkercad is. So it would not make sense for students to go through all the trouble of making a 3D model (that could theoretically also be 3D printed) in Tinkercad and then import it into CoSpaces if they’d have to go through all the trouble of recreating the physics colliders manually in CoSpaces.
My hope is to combine the two web-browser accessible tools together to create a lesson plan where students design models to solve a physics based problem simulated in CoSpaces, then iterate on those models to acheive the best possible results. Iterating is much harder if you have to make changes in two different places every time.
And the worst part is that I can see my camera doing the exact thing I want from the rest of the physics objects in the scene. It just doesn’t apply to my physics objects for some reason.