Glass test

Kosmokrator · Post by **Kosmokrator** » Wed Nov 08, 2006 4:35 am

yea this is the reason but....tha same hapent when i need a color fluid inside a transparent glass....i dont think this is physicaly correct!!!when u have a blue coctail drink inside a glass tha coctail still blue not full transparrent...the image above has a very very red dark fluid inside glass...only in coustics i can see some red...

i think its modeling problem becouse the fluid overlapping glass surface...look at the botom the ono s concept model...my glass follow this modeling consept....

tungee · Post by **tungee** » Wed Nov 08, 2006 9:43 am

yamas! is this ouzo?

Kosmokrator · Post by **Kosmokrator** » Wed Nov 08, 2006 9:46 am

stin ygeia mas tungee!!!
yes is uzo.....

afecelis · Post by **afecelis** » Wed Nov 08, 2006 10:53 am

somehow when I try to model Ono's glass following his tech drawing, I always end up with problems at the center. I tried a couple of times in max and the same with blender. Then I found a nice "glass modeling" tut for blender and gave it a shot.
1st image: with exr lighting; too much DOF I think (don't ask me what kind of wines they are, hehehhe

) and looks a bit strange (not on Indigo's side, but the image itself)

So then I tried a black bkg version with a mesh emitter that came out a bit better:

this exercise showed me what I was doing wrong with Ono's model so next I'll try his glass

afecelis · Post by **afecelis** » Wed Nov 08, 2006 4:26 pm

ahhh! I was finally able to get Ono's glass modeled properly in blender. Here's a quick 7 min render:

dunno if this is the effect he's expecting to see

I'll post clenaer versions as they get "cooked"

afecelis · Post by **afecelis** » Wed Nov 08, 2006 5:18 pm

the "obliged" mesh emitter version:

Big Fan · Post by **Big Fan** » Wed Nov 08, 2006 5:51 pm

edited to remove all

afecelis · Post by **afecelis** » Wed Nov 08, 2006 5:57 pm

why does the glass have a top on it?

check Ono's profile and measures on 1st post:
http://www.indigorenderer.com/montecarlo/glassdiag.png
I think the thickness and the overlapping of the liquid are deliberately made that way to see a certain aspect of refraction

Big Fan · Post by **Big Fan** » Wed Nov 08, 2006 7:29 pm

edited to remove all

Kosmokrator · Post by **Kosmokrator** » Wed Nov 08, 2006 9:42 pm

well i found the reason of my problem....is the mesh overlaping between glass and water surface...Ono's technique dosent work with color dielectrics...instead the old method works fine as u see in the image....

Ono what u think?will fix this in the future releases?

Post by **OnoSendai** » Wed Nov 08, 2006 10:26 pm

For the record:
there is no way to model/render liquids in a glass entirely correctly in any released version of Indigo.
Depending on how this new technique works out, the next version should have support for this.

Kosmokrator · Post by **Kosmokrator** » Wed Nov 08, 2006 10:30 pm

thnx Ono...

filippo · Post by **filippo** » Wed Nov 08, 2006 10:56 pm

as you have created the liquid material?
you excuse my English

come avete creato il materiale liquido?
grazie
filippo

afecelis · Post by **afecelis** » Thu Nov 09, 2006 8:28 am

Code: Select all

<material>
		<name>liquid01</name>
		<specular>
			<ior>1.330</ior>
			<transparent>true</transparent>
			<cauchy_b_coeff>0.000</cauchy_b_coeff>
			<rgb_absorptivity>0.191 0.738 0.738</rgb_absorptivity>
		</specular>
	</material>

Becco_UK · Post by **Becco_UK** » Sun Nov 12, 2006 4:22 am

I doubt over-lapping geometry will give accurate refraction in an accurate render engine such as Indigo.

The liquid/glass surface needs to be a seperate geometry, the normal direction changed and given an IOR value that is an average of the liquid and the type of glass being portrayed.

Is that about right?

Glass test

as you have created the liquid material?

Who is online