Community
Arnold General Rendering Forum
abbrechen
Suchergebnisse werden angezeigt für 
Anzeigen  nur  | Stattdessen suchen nach 
Meintest du: 

aiStandardSurface GGX/Beckmann

6 ANTWORTEN 6
Antworten
Nachricht 1 von 7
Anonymous
819 Aufrufe, 6 Antworten

aiStandardSurface GGX/Beckmann

Could you please add the GGX/Beckmann option to aiStandardSurface? It's slightly ridiculous that I have to use the legacy version for this.

Tags (1)
Beschriftungen (1)
6 ANTWORTEN 6
Nachricht 2 von 7
DeclanRussell
als Antwort auf: Anonymous

Currently the standard surface uses GGX for its microfacet model. Are you finding that there something the beckmann microfacet model can do that GGX cannot?

Nachricht 3 von 7
Anonymous
als Antwort auf: Anonymous

Are you finding that there something the beckmann microfacet model can do that GGX cannot?

Are you finding that there is something apples can do that oranges cannot? You get the point, they are simply different. In doubt have a look at -> https://www.merriam-webster.com/dictionary/different

Isn't the very purpose of a shader to create different looks? Otherwise we could just assign a white Lambert to everything.

Nachricht 4 von 7
DeclanRussell
als Antwort auf: Anonymous

I disagree with your statement that these distributions cannot be compared given that both distributions are designed to model how light interacts with the microfacets of a surface in real life. Don't take my word for it though, check out the original GGX publication in which they do just that and directly compare GGX and Beckmann in relation to how well each model fits with what happens in reality (https://www.cs.cornell.edu/~srm/publications/EGSR07-btdf.pdf).


Perhaps I was more interested which kinds of surfaces do you find that the Beckmann distribution models better than GGX?

Nachricht 5 von 7
Anonymous
als Antwort auf: Anonymous

I disagree with your statement that these distributions cannot be compared

Me too because I didn't say that. I said they're different and that doesn't rule out a comparison of course.

check out the original GGX publication

I don't care about the lab results. If I switch between Beckmann and GGX I see that GGX is somewhat duller and has a wider spread.

I was more interested which kinds of surfaces do you find that the Beckmann distribution models better than GGX?

Nothing specific. I don't work scientifically but adjust the shaders as I see fit for each object in each situation. And sometimes Beckmann matches the desired look better and sometimes GGX or something else.


If I have more options I can adjust things better. Not really something that needs an explanation.

Nachricht 6 von 7
DeclanRussell
als Antwort auf: Anonymous

Thank you for the explanation. Its good for us to understand the motivations behind features before adding them because we may have to support them for a very long time. I've created ticket #9534 to address this.

Nachricht 7 von 7
Anonymous
als Antwort auf: Anonymous

Its good for us to understand the motivations behind features before adding them

Well it's not exactly a feature request since aiStandard had it already and Beckmann is still there via AI_MICROFACET_BECKMANN.

we may have to support them for a very long time

Sure, I can see the financial factor but you already ditched most of the BRDFs from Arnold 4. So having at least one alternative left is far from extravagant.

I've created ticket #9534 to address this

Thanks!

Sie finden nicht, was Sie suchen? Fragen Sie die Community oder teilen Sie Ihr Wissen mit anderen.

In Foren veröffentlichen  

Autodesk Design & Make Report