Journal and Useful Links

This space is going to be a quick reference point for myself to fall back on some key concepts (could be anything from Renderer fundamentals right from subdivision, tessellation , color space, shading and lighting etc) that I wanted to understand and adopt moving forward in my career. I have been maintaining notes for the same from different learning resources over the internet  and today I feel that I should have one over the internet for easy accessibility with useful links to the wonderful resources shared by awesome Artists over the globe that I can fall back whenever I need it. And maybe it can be useful to other Artists matching my wavelength and aptitude in understanding CG terminology and concepts.Over the years I have worked over different rendering pipelines (3DeLight,Mayaman PRman, Renderman for Maya, Arnold, Renderman for Katana, Gaffer-Arnold) which have evolved for better. Transitioning into different choices of renderers have always taken its time and I did spend time in understanding what happens under the hood of each workflow which has always helped in my problem solving skills and just to stand apart from my peers not as TD but a liaison between Artists and TDs.During my career span till date, I have been fortunate enough to have TD friends around to explain and make myself understand in layman terms, since I am not that a technical person (which I differ to disagree at times since we as CG artists deal with creating ART using technology. Technical white papers are beyond my understanding and aptitude at present). 

Hence, all that I might mention here are my interpretation from all the resources over the internet and discussions with TD friends that I have had over the years. You may find some random general tests posted here. There might not be definite structure to this blog, but its just a means of logging my tests and learnings that I might end up doing with respect to shading, lighting, OSL and python.

I have to thank many many artists you have inspired me to create and learn, who have been generous enough to share their tips and tricks over the years. You will find direct links to their works and tutorials in this blog. A replacement of bookmarks for me and an easy access point to stuff that I found useful at a certain point of time while working or while finding answers to questions and finding solutions for problems.

Leigh Van Der Byl
Justin Holt
Paul H. Paulino
Sachin Shrestha
Xuan Prada
Chris Brejon
Arvid Schneider


Differences between Arnold and Renderman Subdivision and Tessellation:
Renderman subdivides and tessellates at runtime while Arnold subdivides and tessellates upfront. Why Arnold does that can be found in the below link in Section 2 and Section 7.4
For displacement details to show properly a geometry will have to be subdivided and tessellated. How things can be optimised in both the renderers will have to be further looked into and understood.

Sony Pictures Imageworks Arnold


Renderman Notes:

The following excerpt is from the Pixar renderman page.


In the Reyes mode of Renderman, Shading rate would define the degree of tessellation now known as micropolygonlength in the RIS mode.
This means that in order to get a similar tessellation level, using a shading rate value of X would translate in using a micropolygonlength value of sqrt(X):
shading rate = 0.25, micropolygonlength = 0.5
shading rate = 1, micropolygonlength = 1
shading rate = 4, micropolygonlength = 2
This is a useful conversion if you need to match previous tessellation.



Subdivision verus Tessellation:



Subdivision Surfaces in Katana:



Subdivision Surfaces in Arnold:

https://docs.arnoldrenderer.com/display/A5AFMUG/Subdivision


Arnold displacement Notes:

https://docs.arnoldrenderer.com/display/A5ARP/Displacement Arnold subdivision Tests:

Following are the per-object Arnold subdivision settings -
Type - catclark
Iterations - 2 (defines the maximum number of subdivision rounds on a mesh)
Adaptive Metric - auto
Adaptive Error - 0  (Uniform subdivision - No adaptive subdivision takes place)
Adaptive space - raster (Object mode will be  useful when rendering instances)
UV smoothing - pin_borders

Maximum subdivisions can be controlled globally as well as on per-object level as defined above.
The actual number of subdivisions for each object will be the lower of the two values.


No comments: