OctaneRender™ Standalone 4.01.1

[abstrax]

Native dialog windows has missing window tile bar:
viewtopic.php?f=14&t=61378&p=353173#p353173

Yes, that's an issue with the UI toolkit we use. We couldn't fix it yet.

Is it Apple's fault or something else?

The issue is somewhere in the GUI toolkit we use in Octane.

[abstrax]

I noticed enabling the noise pass restarts the render now. Is that an intentional change or a bug?

I think it is a bug. Octane should not restart if adaptive sampling is enabled and you enable / disable the noise pass.

[calus]

The exact same very simple static geometry is loaded in Octane as .OBJ but not as .FBX as you can see in the attached ORBX.

I have the problem in many scene on random objects, I couldn't find what is specific to these geometry to not be supported trough FBX,
please let us know what is exactly the limitation or the BUG to adapt our workflow accordingly.

By the way, there is no problem to load these FBX in any other software supporting FBX ...

FBXgeometryBUG.orbx

[WIPEXPO]

Octane Standalone
Hi, i found some strange behave of emission material with post processing enabled (bloom/glare)

mix 2 materials with emission nodes.(material mix)

1st material - all emission visibility off (diff/spec/transp/shadow )
2nd material - all emission visibility on

What i should get is ability to control only bloom glare amount (no illumination ) with my 1st material,
and 2nd material should act as a normal emiter (0.5 mix amount)

What i get is - 1st material is casting light on all surfaces.

is that correct ?

[vijay]

I just discovered FBX support in Octane is way better than I expected, very good indeed :
I was amazed even the linked polygon caching format is supported, and as this cache format is the same as in Maya I can just update the animation cache directly from inside Maya without exporting again the FBX in Standalone, pretty amazing...

However I found annoying issues, some geometry are just not loaded in Octane,
I isolated sample objects having this problem in the attached FBX.
I found if I change the UV mapping of the objects or subivide the mesh, this would eventually solves the issue,
but I really don't understand this bug/limitation in Octane and how to avoid it ? (the FBX have no issue to load in other sofware)

OctaneFBXbug.zip

Hi Calus,

Thanks for the bug report. This is fixed and will be in next release. There was an issue reading opacity for the FBX material as well, that is also now fixed. So please make sure transparency of FBX material is set properly from next version.

Thanks

[calus]

This is fixed and will be in next release.

Thanks Vijay,

In the mean time I got another annoying bug with FBX geometry in Octane :
Subdivision Surface Crease are wrong, the vertex or edge crease strengh is interpreted as if it was 10 times smaller.

For example a crease value of 100 in FBX subdiv give similar result as a crease value of 10 in Alembic subdiv :

FBXSubdivCreaseBug.orbx

[vijay]

This is fixed and will be in next release.

Thanks Vijay,

In the mean time I got another annoying bug with FBX geometry in Octane :
Subdivision Surface Crease are wrong, the vertex or edge crease strengh is interpreted as if it was 10 times smaller.

For example a crease value of 100 in FBX subdiv give similar result as a crease value of 10 in Alembic subdiv :

FBXSubdivCreaseBug.orbx

This is a Maya's FBX exporter bug, which scales down the crease strength by 10. Not sure why. For a workaround, I manually scaled it up by 10 in octane.

[calus]

This is a Maya's FBX exporter bug, which scales down the crease strength by 10.

I had already checked this: once imported back to Maya the FBX have correct crease values, so it doesn't seem to be an FBX exporter bug.
It just looks like FBX specification use a different mapping than Maya and Octane for crease values ( 1 to 100 instead of 1 to 10 ),
so in Octane the crease values from FBX must be interpreted accordingly (mapped back to 1 to 10 ), don't you think ?

[vijay]

This is a Maya's FBX exporter bug, which scales down the crease strength by 10.

I had already checked this: once imported back to Maya the FBX have correct crease values, so it doesn't seem to be an FBX exporter bug.
It just looks like FBX specification use a different mapping than Maya and Octane for crease values ( 1 to 100 instead of 1 to 10 ),
so in Octane the crease values from FBX must be interpreted accordingly (mapped back to 1 to 10 ), don't you think ?

If you check the attachment, you will notice that the exported FBX crease values are scaled down by 10. Not sure why, but this is what octane is reading. So I scaled it up by 10 if the file is coming from Maya.

[calus]

If you check the attachment, you will notice that the exported FBX crease values are scaled down by 10. Not sure why, but this is what octane is reading. So I scaled it up by 10 if the file is coming from Maya.

Are you sure this is specific to Maya FBX ?
maybe this is because autodesk FBX plugins in general just doesn't use the same units for storing crease values, might be an historical reason.
This could be confirmed or not by checking also fbx subdiv exported from 3DSmax.