[Radiance-general] Radiance and Daysim

[Guglielmetti, Robert]

In a word: yup. This recent technical report from Roland Schregle on his photon map work also discusses at length the octree advantage, in this case from the perspective of photon map storage.

https://www.academia.edu/25528679/The_RADIANCE_Out-of-Core_Photon_Map_Technical_Report

On 6/8/16, 7:08 PM, "Thomas Bleicher" <tbleicher at gmail.com<mailto:tbleicher at gmail.com>> wrote:

Ziya

I am copying the Radiance-general mailing list on this conversation. Please subscribe to the list (http://radiance-online.org/community/mailing-lists/subscribe/radiance-general-mailing-list) to follow the discussion there. The list members are far more knowledgeable about this than I am.

I general I'd say that the octree format was chosen because it is an efficient way to organize geometry in computer memory if you have to frequently "query" the geometry for intersections, like during the rendering calculations.

Daysim uses the octree format because it is a specialized version of Radiance. Under the hood the data structures are identical.

On Wed, Jun 8, 2016 at 1:49 PM, Ziya Usta <ziyausta61 at gmail.com<mailto:ziyausta61 at gmail.com>> wrote:
Dear Mr. Bleicher, I  want to know why Radiance and Daysim use octree representation to run simulations. What are the benefits of octree for ray tracing? Any information about that would be very helpful.

All the best,

Ziya