Maya Forum
Welcome to Autodesk’s Maya Forums. Share your knowledge, ask questions, and explore popular Maya topics.
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Maya Vector Render Technology

Message 1 of 1
86 Views, 0 Replies

Maya Vector Render Technology

Right away, I need to say that this question is not directly tied to Maya itself, but I think that, here is the best place to make this question.

When Maya implemented the vector render in 2006/2007, they licensed a technology called RAViX  (RApid Visibility eXtension) from Electric Rain, it was not made in-house. It was a long time ago, nowadays that company is out of business.

To be fair, I have tried to contact them, and I have failed, so I am trying to reach to the software manufacturers that they have licensed the RAViX to, now you know why I am here.

I am trying to know if autodesk can still contact the original authors of that software, or if there is any chance that they can license that to me. I am trying to understand how the renderer works and looking at the code is easier than looking at the exported results and trying to guess what is going on.


To be honest I do not expect this to lead me anywhere, so I am going to report all my findings about RAViX in hope that someone reading this may have a clue that I missed.

If you have worked with the renderer, you noticed that it only works with triangles, and I can explain you why that is the case. It is just for convenience, with quads for example, you can take a plane and move two opposite corner points down, and you can precisely position it in a way where you can only see one triangle. If quads were used in these kinds of scenarios, which are very common, when portions of a quad would not be visible they would need to be converted down to triangles, otherwise in the final render it could create a mess with the “rasterization priority of each quad” (some quad portion may be visible and other may have been obstructed by other quad and cannot be rendered).

Then it is up to calculate the points position with the normal orthographic perspective matrix, and you have the points that need to be in the final render. You could join all the triangles with the same solid color, optimizing the vector output, but I did not analyze if it does this.

The specular light and shadows are calculated for each triangle and stored; in some cases (more realistic renders) it seems to me that it also uses the more advanced “algorithm” that it uses for reflections.

It first calculates for each of the three points of the triangle, the right color. Then it will create a gradient buffer and will scale, skew, and translate it to the triangle, until it matches the points, it may also use the triangle normal to help it to find all the required values, but I did not yet manage to understand how it does manage to accomplish it.

I also want to note that this comes from my understanding of the exported SWF file from the renderer and then I passed through the JPEX FFDEC to get an SVG, which is the one that I analyzed, so it may be because of a flash limitation that I’m not aware, but the gradient buffer goes from -819.2 to 819.2, but I do not know why, it is just my supposition, as for anything else of my prior explanations.

Tags (1)
Labels (1)

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report