Some simple SSS tests

Show off your final renders
Post Reply
13 posts • Page 1 of 1
User avatar
OnoSendai
Developer
Posts: 6243
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Some simple SSS tests

Post by OnoSendai » Sun Nov 23, 2008 12:36 am

I've been messing around with the SSS code a little, and wanted to test out an interesting SSS phenomenon that's easily visible in real life. This is where you position a thin, opaque object (such as a knife blade) above a SSS object. You can then observe light scattering underneath the occluding object, and exiting the surface on the far side of the occluding object.
Note that the thin occluding object is in exactly the same position in all 3 images.
The first image is a Phong material (non-SSS) reference.

The red material is defined like so, and the white material similarly, except that all absorption RGB coefficients are 10.

Code: Select all

<medium>
		<name>subject_medium</name>
		
		<basic>
			<ior>1.3</ior>
			<cauchy_b_coeff>0.0</cauchy_b_coeff>
			<absorption_coefficient_spectrum>
				<rgb>
					<rgb>10 100 100</rgb>
					<gamma>1</gamma>
				</rgb>
			</absorption_coefficient_spectrum>

			<subsurface_scattering>
				<scattering_coefficient_spectrum>
					<rgb>
						<rgb>10 10 10</rgb>
						<gamma>1</gamma>
					</rgb>
				</scattering_coefficient_spectrum>

				<phase_function>
					<uniform/>
				</phase_function>
			</subsurface_scattering>
		</basic>
	</medium>
	
	<material>
		<name>subject_mat</name>
		<glossy_transparent>
			<internal_medium_name>subject_medium</internal_medium_name>

			<exponent>
				<constant>
					100
				</constant>
			</exponent>
		</glossy_transparent>
	</material>
Attachments
sss2.jpg
sss2.jpg (36.29 KiB) Viewed 12279 times
sss1.jpg
sss1.jpg (37.32 KiB) Viewed 12279 times
sss_ref.jpg
sss_ref.jpg (36.43 KiB) Viewed 12280 times

easyarchitect
Posts: 47
Joined: Fri Jan 12, 2007 1:49 am

Post by easyarchitect » Sun Nov 23, 2008 12:53 am

Indigo renderer 1.1.15 :shock: :shock: :shock:

Delle

User avatar
PureSpider
Posts: 1459
Joined: Tue Apr 08, 2008 9:37 am
Location: Karlsruhe, BW, Germany
Contact:

Post by PureSpider » Sun Nov 23, 2008 2:07 am

Nice test!
Can you explain why that happens in real life?
I'm really interested in that phenomenon!

User avatar
OnoSendai
Developer
Posts: 6243
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Post by OnoSendai » Sun Nov 23, 2008 2:16 am

It's pretty simple. Some light enters the block just to the right of the thin blocking object. After one or more bounces inside the object, some of the light happens to bounce out of the block just to the left of the blocking object.

User avatar
PureSpider
Posts: 1459
Joined: Tue Apr 08, 2008 9:37 am
Location: Karlsruhe, BW, Germany
Contact:

Post by PureSpider » Sun Nov 23, 2008 3:16 am

Ah cool thanks for the explanation :)

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Sun Nov 23, 2008 6:29 am

Just to make things clear: this effect is available in Indigo since sss mats were implemented, it's just a test, nothing new isn't it ?

What is the real distance between the two objects, are their surfaces coplanar (or could they be) ? Or must the distance be above the nudge value for this effect to show ?

[/lazy tester mode]
obsolete asset

User avatar
Zom-B
1st Place 100
Posts: 4701
Joined: Tue Jul 04, 2006 4:18 pm
Location: ´'`\_(ò_Ó)_/´'`
Contact:

Post by Zom-B » Sun Nov 23, 2008 6:38 am

Is this a new example scene for Indigo? ;-)

The Speed for the red example is amazing, very nice :)

I would like to see some speed comparison between old & new code!

looking forward to 1.1.15, great work Ono!!
polygonmanufaktur.de

User avatar
OnoSendai
Developer
Posts: 6243
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Post by OnoSendai » Sun Nov 23, 2008 6:48 am

CTZn wrote:Just to make things clear: this effect is available in Indigo since sss mats were implemented, it's just a test, nothing new isn't it ?

What is the real distance between the two objects, are their surfaces coplanar (or could they be) ? Or must the distance be above the nudge value for this effect to show ?

[/lazy tester mode]
The thin blocking object is exactly just touching the block.

User avatar
OnoSendai
Developer
Posts: 6243
Joined: Sat May 20, 2006 6:16 pm
Location: Wellington, NZ
Contact:

Post by OnoSendai » Sun Nov 23, 2008 6:49 am

ZomB wrote:Is this a new example scene for Indigo? ;-)

The Speed for the red example is amazing, very nice :)

I would like to see some speed comparison between old & new code!

looking forward to 1.1.15, great work Ono!!
Yes, it's a new example / test scene.

User avatar
CTZn
Posts: 7240
Joined: Thu Nov 16, 2006 4:34 pm
Location: Paris, France

Post by CTZn » Sun Nov 23, 2008 12:55 pm

Ok thx. There is a strange but nice effect on the last pic, the left topmost edge is fading from blue to red along its length, is this a jpg artifact or physical effect, like cauchy coeff ?

Or should I not ask at all :mrgreen:
obsolete asset

User avatar
Photoguy
Posts: 78
Joined: Fri Jan 18, 2008 2:44 am

Post by Photoguy » Sun Nov 23, 2008 6:37 pm

ZomB wrote:
The Speed for the red example is amazing, very nice :)

I would like to see some speed comparison between old & new code!
Well what Cpu is he running, then I'll be amazed.
22 minutes for 1000 spp is great!

Hellstorm
Posts: 41
Joined: Mon Sep 22, 2008 9:31 pm
Location: Einbeck, Germany
Contact:

Post by Hellstorm » Sun Nov 23, 2008 11:18 pm

CTZn wrote:Ok thx. There is a strange but nice effect on the last pic, the left topmost edge is fading from blue to red along its length, is this a jpg artifact or physical effect, like cauchy coeff ?

Or should I not ask at all :mrgreen:
Yeah. There's also some cyan color on the lower left corner of the cube, directly next to the watermark.

User avatar
Zom-B
1st Place 100
Posts: 4701
Joined: Tue Jul 04, 2006 4:18 pm
Location: ´'`\_(ò_Ó)_/´'`
Contact:

Post by Zom-B » Mon Nov 24, 2008 1:23 am

Photoguy wrote:Well what Cpu is he running, then I'll be amazed.
22 minutes for 1000 spp is great!
Well a Intel 6600 Quad Core afaik...but the point is that SSS seems to be much slower atm. Therefore a speed comparison would be nice :-)
polygonmanufaktur.de

Post Reply
13 posts • Page 1 of 1

Who is online

Users browsing this forum: No registered users and 60 guests