A natural situation

[yonosoy]

Hi.
A very simple situation. Still nude, only geometry, without a physical spectrum definition for sun and atmosphere.
Real scale. Meshes from Houdini thanks to a procedural node from cmiVFX (fractal terrains). Process, materials, technical questions and much more will be posted with time.

[yonosoy]

With the sun in another azimuth. The reference of time and samples second corresponds to an Amd desktop and a i3 notebook connected via LAN (8 threads).

[CTZn]

Cool ! Do you know that Indigo has an Earth-like atmosphere simulation ? It is tricky to set up but it provides a very realistic result.

See:
http://www.indigorenderer.com/planet-sc ... simulation
http://www.indigorenderer.com/forum/vie ... 35#p101435

The camera can fly through that atmosphere, it's pretty exciting to have the camera around the ionosphere ! This feature was used to bake the sun & sky lighting scheme that exporters now use by default.

There was a test scene posted in the forum but I can't find it right now...

[yonosoy]

Hi.
Yes, but thinking that this was a complement to the sun-sky system. So silly...

Some weeks testing with atmosphere medium (and the sky behind..) with a couple of problems that seems to be better now.
Actually the scene is carried out throug Cinema 4D. The radius of Earth was estimated at the begining in 6376 km. In C4D material´s preview atmosphere remains black wiht a Y value of -6376 km. At a value of -6377, -6378 the previsualization is right. If the scene is below of the sea level (in reference to the atmosphere consideration) the light seems don´t do all the travell. The sky lights and can be visible but the scene is basically or totally black. The global earthmesh was losing around 1-3 of effective kilometers because of the traduction in polygons.

The other tricky question is that little fire-flies becomes chromatic when a real tabulated media is present and post pro AD is on. Over a supersample factor of 4 the fire-flies (or chroma) are not visible in the photograph, but for AD could be a festival.

This is a shoot from 1.2 km of height (5 km below than the firsts) with Indigo´s atmospheric medium and a real scale sun (at the moment a primitive sphere wit a 6000º K blackbody material (I know that it´s possible to use an extra atmospheric sun in Sun-Sky system that for sure has a complet spectrum)).

[OnoSendai]

Hi Yonosoy,
The Indigo atmosphere medium starts at exactly 6378135 m from the centre of the Earth.

[CTZn]

I have attached the globe_test.igs as it is on my hard drive currently, but I think that I have tried merging several tests scenes into that file, not necessarily with much success in this case. However the data within should be original, so you can get usefull informations from it. Notably, the globe radius is 6378135 meters. <- *spoiled*

The postpro AD is a 2D filter wich doesn't discriminate pixels, that's the downside of its readyness.

PS: me gusta el appelido que has eligido, que lindo

[OnoSendai]

Here's a better / more recent scene:

[yonosoy]

.

[lycium]

*Awesome* work in this thread, hi and welcome yonosoy!

[yonosoy]

.

[yonosoy]

.

[yonosoy]

.

[yonosoy]

.

[yonosoy]

And the last for a days. So brut in this moments, but it's a try with real tabulated absortion and SSS in real scale. Work's only in console in this little machines and run well !. A simple bump pos(OS) (awaiting for this). Now, to learn ISL ...

P.D: Manages this automatically the pages? (seem too long)

[yonosoy]

Testing with Metropolis Path Tracing and glass acceleration on.
Silver's sphere diameter 1 kilometer. (89º zenith). 20 h.

The first hours of Path Tracing (this time glass acc. off) scares a bit, but it's expected a convergence near the hour 60 (like other precedente images (4 old threads of 2.40)). To be updated when finished. (72º zenith).

Note: in the try to photo develop the exr raw format since a time (sorry for this, sometimes loose coherence).

At the same condition time over an i3 (4 no old but slow threads of 2.10) and Metropolis Path Tracing (glass acc. off again). Maybe a convergence time of a half expected. Since a versions, glass acc. is allowing (pretty bad) the travell of ligth in bidireccional mode.