Hi
I have scratched my head over some jittery pixels in my z depth channel. See attached image.
I have come to realize that it occurs when I use a displacement map.
Have anyone else had this issue?
displacement map causes black pixels in passes?
Moderators: ChrisHekman, aoktar
- rleuchovius
- Posts: 51
- Joined: Sun May 03, 2015 1:47 pm
https://www.dropbox.com/s/qz5a24dz08y8qpq/disp.zip?dl=0
Please look at this simple scene. A plane with a displacement map on. If you look at the Z depth pass there is black pixels jumping all over the image.
Please look at this simple scene. A plane with a displacement map on. If you look at the Z depth pass there is black pixels jumping all over the image.
- rleuchovius
- Posts: 51
- Joined: Sun May 03, 2015 1:47 pm
- Subroutine49690
- Posts: 78
- Joined: Tue Apr 02, 2013 4:58 pm
hi rleuchovius,
I found out
if you set your kernel to info-pass and use only 1-20 samples, you can see the same effect - use more samples, you can see the dots disappear.
I tried higher max-samples for the info-pass but z seems to ignore it.
I found out
if you set your kernel to info-pass and use only 1-20 samples, you can see the same effect - use more samples, you can see the dots disappear.
I tried higher max-samples for the info-pass but z seems to ignore it.
- rleuchovius
- Posts: 51
- Joined: Sun May 03, 2015 1:47 pm
This seems to be a bug. I would say Z always renders only sample dense 1
If you set your kernel to info-pass and use the Z with only 1 sample ....you get the same effect
I would classify this as a bug. Aoktar I would be grateful if you could take a look at this and confirm / disconfirm
As a test I render the z as beauty-pass with more samples... you can see the dots disappear.
I really quite heavy on pass renderings so getting this to work is important to me.
thanks
If you set your kernel to info-pass and use the Z with only 1 sample ....you get the same effect
I would classify this as a bug. Aoktar I would be grateful if you could take a look at this and confirm / disconfirm
As a test I render the z as beauty-pass with more samples... you can see the dots disappear.
I really quite heavy on pass renderings so getting this to work is important to me.
thanks
Hi rleuchovius,
I cannot reproduce your issue here with your scene.
All the info passes are rendering smoothly in both LV and PV even at 16 samples
Which c4dOctane version are you using exactly and what about your hardware?
ciao beppe
I cannot reproduce your issue here with your scene.
All the info passes are rendering smoothly in both LV and PV even at 16 samples

Which c4dOctane version are you using exactly and what about your hardware?
ciao beppe
Hi Beppebepeg4d wrote:Hi rleuchovius,
I cannot reproduce your issue here with your scene.
All the info passes are rendering smoothly in both LV and PV even at 16 samples
Which c4dOctane version are you using exactly and what about your hardware?
ciao beppe
I could reproduce the problem in 3.03.3-R1 Win. Let me know if you need help reproducing it.
Please click for full resolution. @rleuchovius
I'm not sure if you are aware of it, but this is a known issue and a current limitation of displacement. It's caused by to high amount value. You will never avoid artifacts with a flat surface. It's best to use displacement for medium and high detail (trees and rocks) and well subdivided geometry for low detail (big hills). It will also render faster with lower amount. Again, it's important, make sure geometry is well subdivided and closely resembles the shape. Basically, the artifacts happen when original geometry is too "different" from the final result. I hope that makes sense.

I found out about this last year in the middle of a project unfortunately (In v2). I fixed it in post. Or, in some shots, I instanced a ground geometry bellow to fill the "holes". Once I learned how it's supposed to be used everything was fine.
Please don't make duplicated topics on the forum, it's hard to both reply and troubleshoot issues for both users and developers.
Also this is a wrong place to report this because it's a OctaneRender issue. Nothing to do with plugins.
Thanks
Milan
Colorist / VFX artist / Motion Designer
macOS - Windows 7 - Cinema 4D R19.068 - GTX1070TI - GTX780
macOS - Windows 7 - Cinema 4D R19.068 - GTX1070TI - GTX780
- rleuchovius
- Posts: 51
- Joined: Sun May 03, 2015 1:47 pm
Hi thanks for the answers. But the artefact you are referring to is something else. There is no holes in the geometry but jittery black pixels in the passes. In my production scene they appear all over the image not only over the displaced ground. The production scene had high res geometry that was displaced and not a flat surface as in the example I submitted.
I am a c4d and Octane user and does not have the expertise to judge why this happens I can only conclude that it does.
I am a c4d and Octane user and does not have the expertise to judge why this happens I can only conclude that it does.
I can put a sign under all words of milanm. Current displacement method is different. It works as voxels which are generated on realtime. Not with pre-triangulated faces. Imagine as stacked small voxels across face normals. So I think this is a handicap of current method on high amount displacement. And plugins are not rendering even one pixel. All rendering operations done by Octane. We're just preparing data.rleuchovius wrote:Hi thanks for the answers. But the artefact you are referring to is something else. There is no holes in the geometry but jittery black pixels in the passes. In my production scene they appear all over the image not only over the displaced ground. The production scene had high res geometry that was displaced and not a flat surface as in the example I submitted.
I am a c4d and Octane user and does not have the expertise to judge why this happens I can only conclude that it does.
Btw are saying about black pixels on non-displayed surfaces? It's very surprising.
Octane For Cinema 4D developer / 3d generalist
3930k / 16gb / 780ti + 1070/1080 / psu 1600w / numerous hw
3930k / 16gb / 780ti + 1070/1080 / psu 1600w / numerous hw
- rleuchovius
- Posts: 51
- Joined: Sun May 03, 2015 1:47 pm
I understand. Thanks for the clarification.
Yes it is surprising that the black jittery pixels occur not only on the displaced ground but all over the scene equally. This is why I want to see why this is, as it does not seem to be the displacement itself but a bug when using it.
Yes it is surprising that the black jittery pixels occur not only on the displaced ground but all over the scene equally. This is why I want to see why this is, as it does not seem to be the displacement itself but a bug when using it.