Same Transform3D calculations yielding different results on multiples instances of the same type

Are exit points properly oriented?

Orientation is irrelevant here since they work as placeholders.
Even if it was, it still wouldn’t explain why only one of the portals work as intended.

How is orientation irrelevant if you multiply with their whole transforms?

I’m intrigued but it’s very hard to debug like this. Can you post a MRP?

Both portals are facing towards -Z.
Same goes for the exits.

Both portals should be behaving the same way then, but they’re not


For the sake of the argument, I’ve rotated the exit of the orange portal (aka portal a in my previous comments but it wasn’t very clear so I’m now using the color coding)

Nothing has changed. At all. The orange portal is still behaving relatively to the blue’s exit.

We’re clearly on a case of two instances sharing the sames variables in memory as I’ve been trying to explain before. I’ve been hard at work trying to understand how this is even happening in the first place but I genuinely don’t get it.

I’ll make a duplicate script for the blue portal with variables named differently and I’ll see if the behavior changes or if my computer is simply cursed somehow

Oh, that $MeshInstance3D, the portal instances will share the material resource there. Try setting its resource_local_to_scene flag. I’m almost 100% certain it’s that. You’re displaying the same viewport on both meshes.

2 Likes