Frosted Glass Challenge (Any Render Engine)

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

REPLY TO THREAD
 
Thread Tools Search this Thread Display Modes
  07 July 2010
Originally Posted by tomNL: If you mean my Maxwell Material setup, Ref0 is a dark grey. Making it black still works though...


Yes. Okay. I've examined your material and have it rendering in fryrender. It looks much better than the other guy's attempt after only 32 passes. I'm running it on an i7 with 234908 other apps going, so it will take some time.
__________________
- Fran
Surreal Structures
Pirates don't skip.
 
  07 July 2010
25h rendertime ? something is going in the wrong direction here where is the speed thoughts

here is a mental ray render 34sec on a quad core 2.8ghz



Edit: cc the image
__________________
blog
IMDB

Last edited by swag : 07 July 2010 at 10:06 PM.
 
  07 July 2010
Mental Ray image added, thanks.
__________________
Next Limit
 
  07 July 2010
Octane is not really capable of frosted glass yet.
 
  07 July 2010
Doesn't it have a roughness setting?
__________________
Next Limit
 
  07 July 2010
OK, now it's clear, thanks.
Quote: Hi,

You will have to wait until we have a frosted glass material option.
The bump map solution is very hackish and is not efficient at all.

It won't be long before we have roughness on the specular material, it's on our high-priority list.

Radiance / OctaneRender Team
__________________
Next Limit
 
  07 July 2010
Originally Posted by swag: 25h rendertime ? something is going in the wrong direction here where is the speed thoughts

here is a mental ray render 34sec on a quad core 2.8ghz



It doesn't look much like the inspiration render from Jensen. Which is the challenge. Not speed.
__________________
- Fran
Surreal Structures
Pirates don't skip.
 
  07 July 2010
"The bump map solution is very hackish [...]"
... yeah, that's why I didn't invest more time into trying to render that dragon with Yaf(a)ray... This would've been the only way to get frosted glass.

Anyway, meanwhile the next LuxRender image is finished.

20 hrs ...
 
  07 July 2010
Thanks, Marco. Lux Render image is updated.

swag; please read the first post carefully. Rendertime is not important in this challenge at all. You can render it in 1 second or whole day. The challenge is about making a photorealistic frosted glass just like in the initial example. Honestly, yours is way behind this look we're after. Nevertheless, I will keep it on the table until a better Mental Ray example comes out.
__________________
Next Limit

Last edited by tomNL : 07 July 2010 at 10:34 PM.
 
  07 July 2010
My mental ray result is very close to swag's. It seems much softer than the others. I really wish we had photo reference because I doubt a purely diffuse glass has as much facing reflectivity as we are trying to match.

Last edited by scottsch : 07 July 2010 at 01:01 AM.
 
  07 July 2010
is it possible to add further information to the renderings like "time" and if not all the same "the render resolution" It is not all about which renderer makes the most believeable image frosty glas image. I would go with the 6th best looking solution if I could get the rendering in a half amount of time.
 
  07 July 2010
Originally Posted by scottsch: My mental ray result is very close to swag's. It seems much softer than the others. I really wish we had photo reference because I doubt a purely diffuse glass has as much facing reflectivity as we are trying to match.
We are not doing random things, nor rendering of frosted glass is a misery. You can find another test using a reference photo, here:
http://www.maxwellrender.com/forum/...t=34574&start=0
http://communities.bentley.com/prod...54/t/57475.aspx

Originally Posted by Fahrija: It is not all about which renderer makes the most believeable image frosty glas image. I would go with the 6th best looking solution if I could get the rendering in a half amount of time.
This is not a benchmark test. It's only for demonstrating raw accuracy/quality in producing the frosted glass. There's no point trying to compare biased and unbiased rendertimes. Later we can start another challenge for rendertime comparison. Rendertimes alone are not expressive enough. 20 min or 20 hrs on which hardware also matters. It's not as easy as it seems and completely irrelevant in this challenge.
__________________
Next Limit

Last edited by tomNL : 07 July 2010 at 06:59 AM.
 
  07 July 2010
fryrender 1.5 has a bug in the fresnel algorithm when a material becomes transmissive, that's true. That has been fixed internally some time ago but accurate validation takes its time against real world references.
Here is Arion, rendering well and fast, with similar results to the reference image from Henrik Wann Jensen.

 
  07 July 2010
Thanks, Erwan. Arion image is added.
__________________
Next Limit
 
  07 July 2010
Well, I'm curious to know why tom's mxm scene straight out of the maxwell demo is noisier than his maxwell version.

__________________
- Fran
Surreal Structures
Pirates don't skip.
 
reply share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright 2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 01:59 AM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.