OctaneRender™ V4 XB4

[wallace]

Base color defines the color of the base layer, the transmission pin only represents the degree of transmissiveness of the base layer (how much light is going through the surface), that's why the base color is called albedo. It makes sense when you change the degree of transmissiveness, a red diffuse surface will get transformed into a red glass without modifying the color.

The current diffuse term in universal material is taken off glossy's diffuse term, which roughness also doesn't modify. There are plans to add more diffuse brdf models which taken roughness into account back into universal material.

Diffuse transmission is covered with the transmission pin and the specular transmission with roughness = 1.0.

Regarding the other issues, I'm working on them.

[haze]

XB3 authorization has been extended by 30 days. We are looking at ways to make this visible in the UI, and available to plugins.

[calus]

Base color defines the color of the base layer, the transmission pin only represents the degree of transmissiveness of the base layer, that's why the base color is called albedo.
Diffuse transmission is covered with the transmission pin and the specular transmission with roughness = 1.0.

I understand the role of the Albedo color but I say transmission should be totally decoupled from the Albedo color else how can we get a different color for diffuse reflection and diffuse transmission like in the Diffuse material ?

Doesn't the way transmission is implemented now make the universal material not universal ? if it's not possible to get the same results as with the simple Diffuse material ?

[wallace]

I understand the role of the Albedo color but I say transmission should be totally decoupled from the Albedo color else how can we get a different color for diffuse reflection and diffuse transmission like in the Diffuse material ?

Doesn't the way transmission is implemented now make the universal material not universal ? if it's not possible to get the same results as with the simple Diffuse material ?

You are correct that the universal material currently still does not represent the diffuse material 100%.

The diffuse transmission BSDF is something we can still add to specify another kind of transmittance behavior into the material surface, and also changing the current transmission pin would allow us to specify the color that gets transmitted into the material surface. That should cover both use cases you have mentioned.

[calus]

I understand the role of the Albedo color but I say transmission should be totally decoupled from the Albedo color else how can we get a different color for diffuse reflection and diffuse transmission like in the Diffuse material ?

Doesn't the way transmission is implemented now make the universal material not universal ? if it's not possible to get the same results as with the simple Diffuse material ?

You are correct that the universal material currently still does not represent the diffuse material 100%.

The diffuse transmission BSDF is something we can still add to specify another kind of transmittance behavior into the material surface, and also changing the current transmission pin would allow us to specify the color that gets transmitted into the material surface. That should cover both use cases you have mentioned.

Also I noticed, transmission + roughness 1 ( to get diffuse transmission ) seems to work only with "octane" bsdf, other bsdf give very dark result.

[wallace]

Microfacet BSDFs suffer from energy loss with high roughness, this is because in general microfacet BSDFs don't model multiscattering, they only model singlescattering. We'll be implementing multiscatter terms to overcome this deficit, but that really becomes a totally different BSDF all together.

It's really an inherent issue with the BSDF itself.

As of now, if you wanna model rough glass, you would want to use the old Octane BSDF. Different BSDFs aren't really meant to look alike, that's why options exist.

[Notiusweb]

Is there any plan yet as to what XB5 will have in store for us?

My dream list, myself, once V4 is all said and done:
-dramatically faster raw render speed than V3
-separate denoiser settings for bright and dark pixels
-import/export gLTF scene stuff

[Goldorak]

Is there any plan yet as to what XB5 will have in store for us?

My dream list, myself, once V4 is all said and done:
-dramatically faster raw render speed than V3
-separate denoiser settings for bright and dark pixels
-import/export gLTF scene stuff

FBX import and export actually in V4 (just not exposed yet) and we have a gltf 2 module that plugs into the FBX system. Octane 2018 is focused on better procedurals (among many other improvements), and v2019 (which is already in progress) is about speed both in viewport and offline. More on that at Siggraph.

[funk]

I was doing some denoiser tests with a scene that was quite dark. I was setting the imager exposure to 8 to brighten up the image. The denoiser was producing splotchy results

When I set exposure to 1, and increased the brightness of all my lights to match the previous image, the denoiser was doing a much better job.

Would it be possible to have an option to tell the denoiser to take exposure into account (maybe like the "expected exposure" setting in adaptive sampling)?

Or maybe we need a setting that changes the exposure of the image before it gets to the imager (eg. burning in an exposure at the kernel, or a global light multiplier)?

[coilbook]

should I use denoise on completion or denoise every so many seconds for animation? is there much difference? Thanks