(WIP) China caffè

8bstudio · Post by **8bstudio** » Thu Mar 15, 2007 9:01 pm

@Ono:
Here same scene with 1 meshlight and everything into a cube!
Good I think but orange juice is a little bit dark, what do you think about?

@Kram1032:
Hi! Next test I try to reduce tiling table texture, I think is too visible!
The hole... Maybe it shall good for the scene only, like a click with camera! Eheheh!

suvakas · Post by **suvakas** » Thu Mar 15, 2007 9:28 pm

hmm... I think you're going to the wrong direction with that image. The wood texture has gone much worse under that indoor lighting (tileing is strong and visible). I think it's because it doesn't have much to reflect. Those background walls are also not good. Looks very un-natural and CG'ish.

I'd say stick with the outdoor environment (HDR map) or if you want to go indoor, then give some depth to the scene so you could blur the background a bit using DOF.

Suvakas

Colezero · Post by **Colezero** » Fri Mar 16, 2007 3:57 am

DoF would do a great job here, and change the background as suvukas said,
anyway the models and the idea is great.

8bstudio · Post by **8bstudio** » Fri Mar 16, 2007 9:35 pm

Hi!
A little DOF added and changed background and table!
Orange juice effect is changed and I don't know why! Maybe meshlight stronger?? Doesn't matter!
Here image:

Here orange juice

Code: Select all

<medium>
		<name>arancia</name>
		<precedence>10</precedence>
		<basic>
			<ior>1.4</ior>
			<cauchy_b_coeff>0</cauchy_b_coeff>
			<absorption_coefficient_spectrum>
				<rgb>
					<rgb>1.52221 17.7282 89.1913</rgb>
				</rgb>
			</absorption_coefficient_spectrum>
			<subsurface_scattering>
				<scattering_coefficient_spectrum>
					<uniform>
						<value>150</value>
					</uniform>
				</scattering_coefficient_spectrum>
				<phase_function>
					<uniform/>
				</phase_function>
			</subsurface_scattering>
		</basic>
	</medium>
	<material>
		<name>arancia</name>
		<specular>
			<transparent>true</transparent>
			<internal_medium_name>arancia</internal_medium_name>
		</specular>
	</material>

mrCarnivore · Post by **mrCarnivore** » Fri Mar 16, 2007 10:08 pm

The wood table looks better.

The orange juice does look way too dirty. But looking at the material I think the error is obvious: You should turn down the absorbtion and use non-uniform sss instead. Choose a yellow-ish colour for the sss. The uniform sss also scatters blue values which makes the juice look dirty.

8bstudio · Post by **8bstudio** » Fri Mar 16, 2007 10:49 pm

Thanks a lot mrCarnivore! I'll try soon!!!
I saw that many of you have always answers for any questions... I'm so newbie!!!

Kram1032 · Post by **Kram1032** » Sat Mar 17, 2007 1:18 am

That's no problem

I already can answer quite many answers, although, I'm a newbie too! I'm learning nearly every single day

I really like the new version! Looks great!
The tiles aren't noticeable, if there are any...
The light is much more blueish than the liht in the pic before... is it, 'cause of the cooler colour of the texture, or did you change the lights aswell?
If the light is that cool, orange juice may looks like this, anyway...

8bstudio · Post by **8bstudio** » Mon Mar 19, 2007 9:13 pm

Hi Kram1032.
I'm returned at image with .exr map.
Here two solution with candels added:
Next time I'll try another wip.

And

mrCarnivore · Post by **mrCarnivore** » Mon Mar 19, 2007 9:36 pm

Did you already change the orange-juice? If you did, I would recommend to turn down the absorbtion values even more and to increase the sss-values...

The candle that is in the background looks nice, but the one in the foreground doesn't look like wax at all. Did you use different materials?

8bstudio · Post by **8bstudio** » Mon Mar 19, 2007 9:44 pm

Hi mrCarnivore!
Orange juice changed with SSS non uniform like before! Candle in first plane has same material but the model is different!! I prefer candle in second plane too!

Candle:

Code: Select all

<medium>
		<name>candela</name>
		<precedence>10</precedence>
		<basic>
			<ior>1.4</ior>
			<cauchy_b_coeff>0</cauchy_b_coeff>
			<absorption_coefficient_spectrum>
				<rgb>
					<rgb>26.6121 35.1887 20.4238</rgb>
				</rgb>
			</absorption_coefficient_spectrum>
			<subsurface_scattering>
				<scattering_coefficient_spectrum>
					<uniform>
						<value>475</value>
					</uniform>
				</scattering_coefficient_spectrum>
				<phase_function>
					<uniform/>
				</phase_function>
			</subsurface_scattering>
		</basic>
	</medium>
	<material>
		<name>candela</name>
		<glossy_transparent>
			<exponent>125</exponent>
			<internal_medium_name>candela</internal_medium_name>
		</glossy_transparent>
	</material>

Orange juice

Code: Select all

<medium>
		<name>arancia</name>
		<precedence>10</precedence>
		<basic>
			<ior>1.4</ior>
			<cauchy_b_coeff>0</cauchy_b_coeff>
			<absorption_coefficient_spectrum>
				<rgb>
					<rgb>1.0148 11.8188 59.4609</rgb>
				</rgb>
			</absorption_coefficient_spectrum>
			<subsurface_scattering>
				<scattering_coefficient_spectrum>
					<rgb>
						<rgb>98.4307 93.5408 71.978</rgb>
					</rgb>
				</scattering_coefficient_spectrum>
				<phase_function>
					<uniform/>
				</phase_function>
			</subsurface_scattering>
		</basic>
	</medium>
	<material>
		<name>arancia</name>
		<specular>
			<transparent>true</transparent>
			<internal_medium_name>arancia</internal_medium_name>
		</specular>
	</material>

mrCarnivore · Post by **mrCarnivore** » Mon Mar 19, 2007 10:10 pm

I can't help you much with the wax, I'm afraid, since I never tried to experiment with those kinds of materials before.

About the orange juice: decrease to absorbtion by one order of magnitude (meaning to divide all values by 10).

Also I would reduce the scatter-value for blue much more. And decrese the value for red a bit.

Kram1032 · Post by **Kram1032** » Tue Mar 20, 2007 3:06 am

I really like you new juice!
The only thing, that still is missing, is proper sugar...
Maybe you can add a very little surface tension to both the juice and the coffee! But these are just last touches...

(WIP) China caffè

Who is online