Author Topic: ** Corona Animation ** -- relationship setting quality/time  (Read 165495 times)

2013-10-08, 17:39:53
Reply #150

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile
Hey rafpug,

Have you tried using Bucket renderer for animation?
After some testing I found out that using the Bucket method helps with noise by adding adaptation, while still maintaining the advantages of an unbiased solution as opposed to using HD. Plus its relatively fast..

Under "Renderer: Bucket" settings try something like:

subdiv AA: 2-6 (Overall Quality)
adaptive steps: 2-6 (Adaptation)
apaptive thresh.: Higher values increase adaptation as opposed to lower..

You might need to experiment with the above values a bit to get optimal results but generally higher values give better results but takes longer to finish the pass..

If you test the above method let us know of the results any info will be much appreciated since I'm still experimenting as well..
« Last Edit: 2013-10-09, 17:31:10 by Eian »

2013-10-08, 21:56:43
Reply #151

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Hello Eian

Your animation is a sequence of images (with the next step, use a video editing) or by calculating the output video directly from 3dstudio?

Which of these two methods are you using?
The video size you are using?

Regarding the method, one can use both the progressive that the bucket, but it must reason in a different manner with the two methods paying attention to the parameters!

The variables are many
size - frame - passes=time limit  or time limit=passes - AA - GI sec ...etc etc !

Greats
Raf

2013-10-08, 22:08:00
Reply #152

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
I forgot to tell

that if it comes to animation
the frame or the passes=time limit  must be less than 2min., otherwise, the length of the video becomes interminable!

Greats
Raf

2013-10-09, 14:18:46
Reply #153

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile
Oh ok sorry, it seems like I didn't get the point of this thread quite well..
I thought it was about optimal settings that can get you best results in less time..

As for animation I think you should first try and find the optimal settings for different shots in each of your scenes.
Animation is nothing more than sequences of images, a renderer as in this case Corona has no effect on the way a video is encoded it just gives you the images.
There are many programs you can use to composite these images and make the final video.

As for the settings I believe they're pretty straightforward if you don't use the "Debug UI".

If you want to get an unbiased solution which yields the best results you could try leaving everything to default (PT+PT) and then changing only parameters that might be needed for your scene.

For example if you have a complex refractive object you would have to raise "max ray depth" parameter in order to
change the maximum number of times a ray can be refracted and get the correct result.
As for "path tracing samples" 16/32 (few times 64) is best, in order to provide an even distribution of sampling..
And so on..

Now I mentioned "Bucket Method" because it gives you adaptation which I believe is what you would look for when you want to have fast results with even noise distribution across your images. It's closer to the way Arnold Renderer works which is used in films lately, hence used for animation.

Still excuse me if I fail to understand the context of your question.

Cheers..
« Last Edit: 2013-10-09, 16:47:43 by Eian »

2013-10-09, 19:11:56
Reply #154

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Hello Eian
thank you for answering

....no, quiet....indeed, I want to better understand and share your information.

a premise:
I do not have much depth rendering engines like Maxwell Indigo Vray rendering, but this Corona Render I'm going to know in detail, because at first glance and 'very fast.

You did well to write.
I'm here to investigate the use of this rendering engine and try to share with others the advice.

Now I'm out and I'm writing from your mobile phone.

Tonight I try to put in order my final tests with the example of a video and share it with your advice in order to find the optimal solution.

Thanks
Raf

2013-10-10, 00:43:07
Reply #155

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile

For example if you have a complex refractive object you would have to raise "max ray depth" parameter in order to
change the maximum number of times a ray can be refracted and get the correct result.


Hello Eian

then

1)  object complex refractive ---> value "Max ray depht" > 25
this value has a great influence on the rendering time?

2013-10-10, 00:44:43
Reply #156

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile
Hey rafpug,

I'm glad if any of my information could prove useful.

I'm still experimenting with Corona as well, and having some in depth experience with a couple of other rendering engines
I must say that Corona seems indeed very promising..
Having used Arnold Render lately, which I find it to be quite similar to Corona (especially when Bucket mode is used)
I noticed some aspects that Corona seems superior, like complex transparent materials..

I hope that it keeps getting improved and ported to Maya soon so I can test it in a greater extend..

Waiting for your final tests..

Cheers..

2013-10-10, 00:51:45
Reply #157

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile

As for "path tracing samples" 16/32 (few times 64) is best, in order to provide an even distribution of sampling..
And so on..


.. (few times 64) is best |  is better in such cases?

2013-10-10, 00:54:23
Reply #158

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile
Quote
Hello Eian

then

1)  object complex refractive ---> value "Max ray depht" > 25
this value has a great influence on the rendering time?

Hey rafpug,

It depends on the scene and the complexity of the refractive object..
Though having tested it with a fairly complex object I can say that it doesn't have a great impact on rendering time..

2013-10-10, 01:03:13
Reply #159

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Hey rafpug,

I'm glad if any of my information could prove useful.


Yes Eian
because I take my last test PT+HD with BUCKET and apply it to these parameters that what you're suggesting

The method PT+HD works better than PT+PT only if you make changes on the GI secondary ...

2013-10-10, 01:04:55
Reply #160

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile

As for "path tracing samples" 16/32 (few times 64) is best, in order to provide an even distribution of sampling..
And so on..


.. (few times 64) is best |  is better in such cases?

Although I didn't thoroughly tested it a value of 32 should be enough for final renderings.
As stated in the documentation a value of 64 might pay off in scenes with deeper areas with small light entrances.
In most cases though, a value of 16 works fine, providing even distribution of sampling..

So a value between 16-32 would be best depending on the needs of your scene..

2013-10-10, 01:10:05
Reply #161

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
The method PT+HD is much faster, and if you change the GI  secondary  is eliminated completely the effect of noise


In a previous test result is better than the Bucket, both as a speed that as yield.

But now, I'm curious to apply these two parameters suggested by you, with my parameters tested

2013-10-10, 01:14:16
Reply #162

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
You use the Alpha version in June?

2013-10-10, 01:21:24
Reply #163

Eian

  • Active Users
  • **
  • Posts: 42
    • View Profile
Yeh I've only used the
Jun 12 2013 (ALPHA v5)

I might test some newer ones when I find some time as well though.

Well as for the settings, PT+HD can prove adequate in many occasions but know that it's not a fully unbiased solution which in general means less realism..
Not saying of course that you can't get great realistic results with the above settings..

2013-10-10, 01:26:00
Reply #164

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
These are the values ​​that I adopt the method with PT+HD

Renderer :PROGRESSIVE
PT + HD

PTsamples 26
Max. samples intensity 20

Lights
Samples Multipler  4,666

Progressive Rendering
Time limit  00:01:20


GI: HD CACHE  1,067 - 271 - 7 - 2 - 2367     Secondary  2 - 15 - 3 - 0.9 - 0    Time: 0:01:27 | Passes: 10

example:
http://forum.corona-renderer.com/index.php/topic,1486.msg11852.html#msg11852


This weekend I'll do the tests on objects with complex refractive index, using your suggestion!