Thanks for implementing a better fix.
Occasionally I can get the illegal address error to appear. so I'm going to mark this report as confirmed.
System Information Operating system: Windows-10-10.0.22621-SP0 64 Bits Graphics…
Extra note because I just remembered it.
It's possible that even with Blender and Cycles being perfect, a user could still create a NaN normal with something like a custom OSL script (that can…
Here's some extra information.
I built a debug version of this pull request and opened it with break points in getPixelAtCursor
(GHOST_SystemCocoa.mm) and eyedropper_color_sample_fl
…
Testing on a M1 Pro mac on macOS 13.4.1, this doesn't work. As in, the colour picker still works inside Blender, but not outside Blender. I have attached a video showing this.
Things I…
I am expecting
light_tree_cluster_select_emitter()
to pick an emitter if there is one, unless all emitters in the cluster have zeroimportance[0]
. If any input is invalid it suggests that…
@LukasStockner I've re-read over various notes in #99447.
It seems to want to support a case of IOR = 0 with the specular layer. This pull request/commits counteracts that goal. So feel free to…
@baoyu manually dividing up the scene into different parts and setting the correct IOR for each medium change is annoying and it is difficult to get right in some scenes. But it is more physically…
@baoyu for as long as I've been using Blender (roughly 6 years), to render a nested nested dielectric accurately you typically needed to set one of the IORs to below 1.0. This pull request doesn't…
Before landing to the main branch I am very curious about that Illegal CUDA Address you've mentioned. How is it NaN causing an invalid address issue? It might be very good idea to mention…
If this commit gets accepted, it needs to be backported to Blender 3.3 and 3.6.
Note: EEVEE and OSL do not need updating to use safe_normalize. OSL always uses safe_normalize
when normalize
…
I decided to take a quick look into debugging this issue. This is what I could find so far.
- This issue is not necessarily something wrong with the
Vector Displacement
node. - The issue…
This commit should be backported to Blender 3.6 if it is accepted.