Hi everyone.
I was running some simple tests with displacement maps in H9 - just a grid with a texture and a displacement map.
With the Micropolygon renderer everything worked fine. Then I tried to render with the PBR and suddenly I got some strange results/artefacts. I don't think it's the same issue like those black dots that you can get when the displacement bound is too low. Cause I got them first and then fixed them by using a higher value. And with Micropolygon everything looks fine too. I got also got a strange output when using Micropolygon PBR. :?
Did anybody else run into this issue? Do I overlook a setting or something?
cheers,
Daniel
Displacement problems when using PBR
6613 7 3- schyzomaniac
- Member
- 69 posts
- Joined: July 2005
- Offline
- Allegro
- Member
- 696 posts
- Joined: March 2006
- Offline
- andrewc
- Member
- 1002 posts
- Joined: July 2005
- Offline
- schyzomaniac
- Member
- 69 posts
- Joined: July 2005
- Offline
Okay, I investigated a little bit more and found the trigger for this behavior.
As soon as I turn on “Polygons as Subdivision” under the Geometry tab inside the Render settings of an object, I get those strange results. When I turn this off, the render looks fine with the PBR.
Unfortunately I can't upgrade to 9.1 easily since my downloads from the sesi website are really slow. So I decided to attach my scene file. Anybody using 9.1, please feel free to test it and report back here if you got the same problem as I do. I'm running H9.0.794 right now.
cheers,
Daniel
As soon as I turn on “Polygons as Subdivision” under the Geometry tab inside the Render settings of an object, I get those strange results. When I turn this off, the render looks fine with the PBR.
Unfortunately I can't upgrade to 9.1 easily since my downloads from the sesi website are really slow. So I decided to attach my scene file. Anybody using 9.1, please feel free to test it and report back here if you got the same problem as I do. I'm running H9.0.794 right now.
cheers,
Daniel
- probbins
- Member
- 1145 posts
- Joined: July 2005
- Offline
I can confirm the same results in 9.1.124.3
displ_pbr.hipnc
I believe it is a problem with the vex displacement vop.
The attached image may be of help. The only thing I've changed from the supplied hipnc file is to remove the reference to the image in the displacement node. Well no, I converted the vops to a material and used a material sop, but that didn't change the original problem.
The “cracking” is still there but the more unusual thing is that the top half of the image doesn't seem to be receiving much light. If you boost the brightness in mplay you'll see it is getting some illumination.
I notice in your file you are still using bare shops and a Shader sop.
The methodology for H9 is to convert vops to material nodes and use Material sops.
displ_pbr.hipnc
I believe it is a problem with the vex displacement vop.
The attached image may be of help. The only thing I've changed from the supplied hipnc file is to remove the reference to the image in the displacement node. Well no, I converted the vops to a material and used a material sop, but that didn't change the original problem.
The “cracking” is still there but the more unusual thing is that the top half of the image doesn't seem to be receiving much light. If you boost the brightness in mplay you'll see it is getting some illumination.
I notice in your file you are still using bare shops and a Shader sop.
The methodology for H9 is to convert vops to material nodes and use Material sops.
“gravity is not a force, it is a boundary layer”
“everything is coincident”
“Love; the state of suspended anticipation.”
“everything is coincident”
“Love; the state of suspended anticipation.”
- schyzomaniac
- Member
- 69 posts
- Joined: July 2005
- Offline
Maybe this thread should be moved to the Houdini 9.1 forum then to report this as a bug. A rather critical one it seems to me.
probbins, thanks for testing the file. And yeah, I know I use the old method of applying materials. I'm pretty new to Houdini, even though I registered six years ago to this website. So I'm hoping to learn all those new methods with the new 3dbuzz online class.
cheers,
Daniel
probbins, thanks for testing the file. And yeah, I know I use the old method of applying materials. I'm pretty new to Houdini, even though I registered six years ago to this website. So I'm hoping to learn all those new methods with the new 3dbuzz online class.
cheers,
Daniel
- ToddDurant
- Member
- 24 posts
- Joined:
- Offline
- andrewc
- Member
- 1002 posts
- Joined: July 2005
- Offline
-
- Quick Links