Messages

8821

Work in Progress/Tests / romullus wips

« on: 2014-01-13, 20:50:56 »

I saw the other guy on forum posted beautiful renders with vcm: http://forum.corona-renderer.com/index.php/topic,2450.0.html and wanted to try it myself, but run into problem of excessive fireflies. Fortunately with the help of other coronauts, here in forum, i managed to find a solution. It seems that vcm don't like bluriness in reflection and refraction slots. Especially in refraction. Put some map in that slot, other than white and you'll get a horde of fireflies.
Anyway here's my attempt:

8822

Gallery / Re: My test of Corona render

« on: 2014-01-13, 15:40:42 »

Fixture on ceiling in my scene has material with light emission on it, but it's very weak. Light source that simulates flashlight near camera, is corona light. It's small (20mmx5mm) and extremely powerful - 20K units of intensity. I tried to change its shape to sphere, but fireflies grew up in orders of magnitude :]
My system units is in mm.

I decided to make very simple scene to experiment more with caustics and after many tries and errors, i found one way to completely get rid of these nasty bugs. I found that setting glossiness value to 1, in reflection and refraction, gives fireflies free image. Now i'm trying to fix all materials in my main scene. Will see if it works in more complex scene.

8823

Gallery / Re: My test of Corona render

« on: 2014-01-13, 01:05:09 »

Keymaster and Dolphinoff, thanks for answers. Unfortunately, changing random sampler didn't helped a bit.
I tried to hide all objects with glass and metal materials on them and all fireflies has gone. Then i unhide big flat mirror - no fireflies, but as soon as i started to unhide small curvy objects with high specularity materials, those nasty creatures started to crawl back :[ My typical chrome material has  0 diffuse, 0.5-0.6 reflectivity, 0.96-0.98 glossiness, 18-25 IOR and glass - 0 diffuse, 0.8 reflectivity, 1.54 IOR and solid shadows. Not sure if i could change these settings much without altering believability too much.
Somehow i think that biggest secret of fireflies free picture from Dolphinoff, is lack objects with complex shape and high glossiness in his scene.

8824

[Max] General Discussion / Re: lighting in closed interior

« on: 2014-01-12, 12:03:35 »

Sometimes i get impression, that people gets spoiled easily by the speed of corona. Soon they'll start to complain about corona not rendering in realtime :] As myself, i came from mental ray here, so it never ceases to amaze me how fast this renderer is.

8825

Gallery / Re: My test of Corona render

« on: 2014-01-12, 11:50:33 »

Yes, unfortunally, light is small and very strong. I even used a little directionality on that light. I'm not sure if it can cause any problems. Glass is solid on all objects, as far as i recall.
Btw, does changing such parameters as PTS, MSI or LSM, have any influence on image quality in VCM mode? I did some test region renders while changing those and didn't notice any significant changes.

8826

Gallery / Re: My test of Corona render

« on: 2014-01-12, 10:55:18 »

Wonderful results. Could you show your render settings? I tried to do something similar, but all i get is ton of fireflies, even after 5K passes.

8827

[Max] General Discussion / Re: Corona Alpha4 Benchmark scene

« on: 2014-01-10, 14:26:33 »

Corona Renderer Alpha 4 benchmark scene
 Living room 100 passes
Intel(R) Core(TM) i7-4771 CPU @ 3.50GHz
Time: 0:4:38, Rays/s: 4,559,964

8828

[Max] General Discussion / Re: Select by material

« on: 2014-01-10, 14:25:32 »

3ds max 2014 sp3

8829

[Max] General Discussion / Re: Select by material

« on: 2014-01-10, 13:53:20 »

It's working perfectly fine for me in A5, but i don't think it's Corona related thing.

8830

[Max] Corona Goodies - User Contributions / Camouflage material

« on: 2014-01-10, 13:47:47 »

I'd like to share my procedural camouflage material with you. All maps in that material are 3d procedurals. This means you don't have to assign uv mapping to your objects.
Material created in 3ds max 2014, so i don't know if it will open in older versions.

8831

[Max] Resolved Bugs / Re: Disappearing proxies and other oddities.

« on: 2014-01-10, 13:34:08 »

Only 77K rays per second? Something must be wrong with how you setup your scene. You may want read this: http://corona-renderer.com/wiki/performance_debugging

8832

Gallery / Re: Night Scene

« on: 2014-01-06, 16:01:35 »

Looks good, but background is wrong - sun is high above horizon.

8833

[Max] Resolved Bugs / Re: Body Objects - rendered as low poly

« on: 2014-01-06, 10:30:31 »

Wow, it's getting harder and harder to wait for A6, then :]

8834

[Max] Resolved Bugs / Re: Body Objects - rendered as low poly

« on: 2014-01-05, 23:42:17 »

I don't know if it was changed in daily builds, but in alpha, body objects renders significantly different than in mental ray or scanline. In corona, one have to crank rendering aproximation settings much higher to match MR output. And MR somehow adjust triangulation depending on render size (the higher render output size one chooses, the more faces, body object will have on render), whereas corona always produce constant number of primitives depending on rendering aproximation settings in body object.
I understand that such things is very low on priority list, but hopefully in distant future it will be fixed...

8835

[Max] Resolved Feature Requests / Re: targeted corona lights

« on: 2014-01-05, 13:11:23 »

Glad to hear it :]