It would be great if you could create a test file that demonstrates that the results are the same in geometry nodes and shader nodes.
I've added a file that can be used for validation.
Patch updated to main only for test and comparison to new gabor implementation.
Stop, this is not the correct to use negative size of the periodic module.
The replacement function behaves differently with negative inputs. It seems when it comes to negative numbers there…
Surely, the branching is there to handle negative numbers. Different outputs when running the new function with negative numbers compared to the existing function.
@CharlieJolly Sure, but what does
(float3(i > 0, j > 0, k > 0) - local_position) * cell_offset
compute?
It's computing the relative position nearest the edge of the cell grid based on the…
@CharlieJolly Can you explain how the condition works?
Add early exit for skipping cells that do not contribute to the noise.
The code looks wrong or incomplete.
@OmarEmaraDev…
@OmarEmaraDev Based on the previous patch (#125443) I have some further optimisations taken from my original implementation #110802.
- Add early exit for skipping cells that do not contribute…