OctaneRender™ 2020.1 RC2 [superseded by 2020.1 RC3]

[wallace]

Hi there.

@aether_nox has encountered a Bug and gave me the scene for inspection:
This is not directly related to the latest version of Octane as it also occurs in other versions.

If you have a Octane Spot Light and shine it through a refractive object all works as expected. Caustics form nicely.
Unless you change the material BSDF to anything else then Octane. Then the caustics do not resolve nicely.

Please find attached the Orbx

Octane_Spot_Light_Caustics_Bug_01.png

I replied to you in another place, I'll put my reply here:

It's basically caustic blur being more effective on Octane BSDF compared to Microfacet BSDF model. i.e. a lower caustic blur value blurs the Octane BSDF more relative to the corresponding Microfacet BSDF model.
This is something I can tune to make the caustic blur work more similar as each other.

[jimho]

Regard to the network rendering again,
I found it will still fail when the slave is mixing RTX and GTX cards, if GPUs from a slave are all RTX or all GTX, they will work just fine,
Guess it is because RTX is by default on and there is some of the GPUs do not functionally support RTX make the failure,

Probably it will help if there is an option in the daemon installing process that to turn off RTX for these kind of mixing configuration slaves.

[mojave]

Regard to the network rendering again,
I found it will still fail when the slave is mixing RTX and GTX cards, if GPUs from a slave are all RTX or all GTX, they will work just fine,
Guess it is because RTX is by default on and there is some of the GPUs do not functionally support RTX make the failure,

Probably it will help if there is an option in the daemon installing process that to turn off RTX for these kind of mixing configuration slaves.

Hi,

Thank you for your help. We aware of this issue, we will work on it so hopefully can be fixed in the next build.

[pxlntwrk]

Hi,

Perhaps this is a question that has already been asked
Why is there no "Glossy layer" in the "Material layers"?



thanks

edit: I get it !

[Notiusweb]

Ooomf--

Bringing in a groom into Octane is like a Vectron, it really chugs the render engine hardcore.

What is it that makes this happen, the Opacity affects but of course alters the strand width.
Is it the actual number of hairs, the density?
Is there a balance between lowering these things and then getting Octane to render it faster?

[coilbook]

Can we get besides samples/s also how much RTX accelerations contributes to it like a percentage value or another sampls/s with rtx

[mojave]

Can we get besides samples/s also how much RTX accelerations contributes to it like a percentage value or another sampls/s with rtx

Not really sorry, this is not the way it works as you would need something to compare against for which you would have to render your scene without RTX first.

[Notiusweb]

So in doing more testing with Groom hairs it seems the density is affecting the render speed.
I am going to guess it has to do with a greater # hairs = a greater # of calculations

But also with Opacity, an opacity of 1 renders fast and then towards 0 renders slower. This is not just with Grooms but also with any opacity masked texture.

Which brings me this question - Why is this the case with Opacity?

Is it rays of light have to navigate through that opacity texture, to produce the correct shadow etc....
But then why would a higher opacity towards 1 be easier than a lower opacity towards 0.
Is it because the absolute geography the light muss be calculates is greater towards 0 than towards 1?

In a nutshell - Opacity...WTF?...

[nejck]

Afaik every pathtracer has "issues" with opacity maps In other renderers disabling texture filtering per bitmap helps.

[frankmci]

So in doing more testing with Groom hairs it seems the density is affecting the render speed.
I am going to guess it has to do with a greater # hairs = a greater # of calculations

But also with Opacity, an opacity of 1 renders fast and then towards 0 renders slower. This is not just with Grooms but also with any opacity masked texture.

Which brings me this question - Why is this the case with Opacity?

Is it rays of light have to navigate through that opacity texture, to produce the correct shadow etc....
But then why would a higher opacity towards 1 be easier than a lower opacity towards 0.
Is it because the absolute geography the light muss be calculates is greater towards 0 than towards 1?

In a nutshell - Opacity...WTF?...

As I understand it, the higher opacity of a surface, the more quickly any given ray that passes through it will reach the Path Termination Power, and stop being calculated. Each time a ray passes through a surface that is less that perfectly transparent, it loses some power until it hits the Termination cut-off. So for example, a stack of 1000 90% opaque surfaces will quickly kill a ray in the first few intersections, but a stack of 1000 0.5% opacity surfaces will require many more intersections and calculations before cut-off. You should be able to tweak the behavior with your depth samples and the Path Term Power setting, but it's just the nature of ray-tracing.