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!

 
Thread Tools Display Modes
  07 July 2010
Fryrender's frosted glass is not correct. Fixing it has been on the to-do list, but there are other things that are higher priority. Please replace the other one with this one:

__________________
- Fran
Surreal Structures
Pirates don't skip.

Last edited by Frances : 07 July 2010 at 03:09 PM.
 
  07 July 2010
fryrender image is updated. Thanks, Frances.
__________________
Next Limit
 
  07 July 2010
Modo

Modo 401 render.

Took about 2h30 but could be faster, just cranked the settings up and didn't optimise.

 
  07 July 2010
Modo looks really good Matt.
__________________
- Fran
Surreal Structures
Pirates don't skip.
 
  07 July 2010
modo image added, thanks!
__________________
Next Limit
 
  07 July 2010
Originally Posted by jwiede: You seem to be implying that yours are somehow better, despite all appearances to the contrary.


[Your post must have been delayed because I didn't see it until now.]

No, that is not what I was implying at all. At this point I had not produced a render and didn't intend to (because I would consider it to be a conflict of interest) until a very poor fryrender example was submitted. It was only then that I was even aware that there was a bug with fry's frosted glass.

We have a lot of talented users and I just presumed that someone would post a nice fryrender frosty dragon and be done with it. Obviously that didn't happen and for good reason. A person should be able to pick a material out of the fryrender library and get expected results. That didn't happen in this case. In a previous release it would have been fine. Until we can correct the dielectric BRDF, we will be working on a way for our customers to achieve a good rough glass.

Unfortunately, as the challenge rules dictate, frosted glass must be achieved solely by the roughness parameter and fryrender fails with that since the higher the roughness value, the larger the reflection and above 20 or so, it looks like metal.

Cheers.
__________________
- Fran
Surreal Structures
Pirates don't skip.
 
  07 July 2010
Kray 2.1

Here's Kray 2.1 render. I hate the blue background, sorry...
30min on core i7 920....
Attached Images
File Type: jpg kray00000_sc.jpg (52.9 KB, 204 views)
 
  07 July 2010
Originally Posted by Jure: I hate the blue background, sorry...


As I'm sure most of us do.. but it's an important requirement for the render.
 
  07 July 2010
Jure, I'd like to include it if you do it over that blue. Thanks!
__________________
Next Limit
 
  07 July 2010
I'd like to add an entry for a usually forgotten render engine.
By the way, I too am sorry about the background color, but this engine is a bit limited in that regard



Please add this notepad-rendered image to the list
 
  07 July 2010
Cinema4D image added, thanks!
Also thanks for the notepad render, Marty. I can't add it to table though...
__________________
Next Limit
 
  07 July 2010
Kray 2.1

Ok here's the blue version.
Attached Images
File Type: jpg kray_blue.jpg (70.9 KB, 125 views)
 
  07 July 2010
Thanks, Jure! Added your Kray entry now.
__________________
Next Limit
 
  07 July 2010
Originally Posted by sundialsvc4: Watching these exchanges with, of course, quite some interest ... I know that it is intended as a blow-by-blow comparison of "straight rendering" algorithms, but I do wonder whether the same results could not well be achieved through compositing. I presume that this is beyond the intended scope of the challenge.

To my eye, and to be quite frank about it, none of these images are outstanding -- not even the first one. And what is "not outstanding" about it, is the one thing that I instinctively zoom-in on from days as a studio photographer: surface characteristics. Even a "frosted glass" object, properly lit, will have clear, crisp surface roughness that must be portrayed in agonizingly-sharp detail. Even if the surface is "matte," it's got surface highlight. (Just grab any copy of, say, Vanity Fair magazine and study the first twenty pages.)

Look carefully at the shadows that are being cast by the dragon. It's at the usual "two o'clock high" (and the dragon, by the way, is magically floating about a quarter-inch high in the air). So, howcum the dragon is glowing as though it had a light bulb stuffed inside its guts? What is the plausible physical explanation of why the dragon's lower lip, and the tip of his tail, and the underside (but not the top) of his body, should be "positively glowing" in this way? (I mean, "which side of that tail is supposedly pointing directly towards the light? Not that one!")

Why is there no surface definition? The reflection off the surface of the tail is muddy and indistinct, when it ought to be crisp and sharp. The strongest visual impression that I think I should be getting off this "frosted glass" is the light that is bouncing directly off the glass surface.

Okay, okay, I won't argue with math and physics. But, there's not a single one of these renders that I have seen yet which says to my practiced eye, "this is a frosted-glass dragon sitting on a studio table, properly lit and in tack-sharp focus."


It would be nice to see a photo of the actual thing to compare to. Looking at the renders, it's hard to say which ones are more accurate since a frosted glass dragon floating against a solid blue background isn't something you come across everyday.

Head to head comparisons like this are helpful to look at. I'd love to see more tests with gems, glasses of water and other misc hard to render items.

Tom, why not add a poll where people can vote for their winners and losers?

Last edited by TJFrame : 07 July 2010 at 06:38 AM.
 
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 05:50 PM.


Powered by vBulletin
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.