Microdisplacement; adaptive dicing introduces interpolation errors in the rendered view #49048

New Issue

Adam Friesen · 2016-08-08T20:51:55+02:00

Adam Friesen commented

2016-08-08 20:51:55 +02:00

System Information
Win10 64 bit; 24 gigs RAM; third-gen i7 CPU; ATI 79xx GPU

Blender Version
Broken: hash a0c1876
Worked: never

Short description of error
When you enable the rendertime dicing in the subsurf modifier (for displacement), the rendered view will show interpolation errors between the various patches (the F12 render meanwhile seems to be relatively unaffected). This also does not happen when displacement is being used without dicing.

Exact steps for others to reproduce the error
Open the .blend file and switch to the rendered view

image

blend
MicroDispBug1.blend

**System Information** Win10 64 bit; 24 gigs RAM; third-gen i7 CPU; ATI 79xx GPU **Blender Version** Broken: hash a0c1876 Worked: never **Short description of error** When you enable the rendertime dicing in the subsurf modifier (for displacement), the rendered view will show interpolation errors between the various patches (the F12 render meanwhile seems to be relatively unaffected). This also does not happen when displacement is being used without dicing. **Exact steps for others to reproduce the error** Open the .blend file and switch to the rendered view **image** ![Microdisp_bug1.jpg](https://archive.blender.org/developer/F334362/Microdisp_bug1.jpg) **blend** [MicroDispBug1.blend](https://archive.blender.org/developer/F334364/MicroDispBug1.blend)

Adam Friesen commented

2016-08-08 20:51:55 +02:00

Changed status to: 'Open'

Adam Friesen commented

2016-08-08 20:51:55 +02:00

Added subscriber: @Ace_Dragon

Mai Lavelle self-assigned this 2016-08-08 22:30:17 +02:00

Mai Lavelle commented

2016-08-08 22:52:50 +02:00

This is a limitation of the subdivision code, which currently doesn't preserve adjacency info for patches and thus cant generate proper normals at patch boundaries. This is only a problem when the "true" displacement method is used, as "both" will calculate normals from bump mapping and doesn't need adjacency info. This should also not be noticeable if the dice rate is high enough (1 micropoly per pixel) as then the artifacts become too small to see.

This will be fixed eventually, but I haven't decided how best to handle it yet.

This is a limitation of the subdivision code, which currently doesn't preserve adjacency info for patches and thus cant generate proper normals at patch boundaries. This is only a problem when the "true" displacement method is used, as "both" will calculate normals from bump mapping and doesn't need adjacency info. This should also not be noticeable if the dice rate is high enough (1 micropoly per pixel) as then the artifacts become too small to see. This will be fixed eventually, but I haven't decided how best to handle it yet.

Brecht Van Lommel commented

2016-08-24 20:36:56 +02:00

Added subscriber: @brecht

Brecht Van Lommel commented

2016-08-24 20:36:56 +02:00

Marking this as a To Do item since this will still be an experimental feature in 2.78.

I guess this is somewhat similar to #49049 where averaging positions rather than normals could help.

Marking this as a To Do item since this will still be an experimental feature in 2.78. I guess this is somewhat similar to #49049 where averaging positions rather than normals could help.

Brecht Van Lommel changed title from ~~Exp. Microdisplacement; adaptive dicing introduces interpolation errors in the rendered view~~ to Microdisplacement; adaptive dicing introduces interpolation errors in the rendered view

2016-08-24 20:37:12 +02:00

Mai Lavelle was unassigned by Brecht Van Lommel

2018-01-25 08:34:46 +01:00

Brecht Van Lommel commented

2018-01-25 08:34:46 +01:00

Added subscriber: @MaiLavelle