Indigo Renderer 2.4.3 Released

[OnoSendai]

This is a Beta release.
If you spot any bugs or problems, please make a post about them in this thread.
Thanks!

Indigo for Windows 32-bit:
IndigoRenderer_2.4.3_Setup.exe

Indigo for Windows 64-bit:
IndigoRenderer_x64_2.4.3_Setup.exe


See the 2.4.2 post at http://www.indigorenderer.com/forum/vie ... f=1&t=9619 for a description of the network manager.

Some more new features:

UV affine transformations

Code: Select all

<material>
    <name>mat1</name>
    <diffuse>
      <texture>
        <uv_set>albedo</uv_set>
        <path>__55_Chevy_by_marpo3.jpg</path>
        <exponent>2.2</exponent>
        <tex_coord_generation>
          <uv>
            <matrix>1 0 0 1</matrix>
            <translation>0.5 0.5</translation>
          </uv>
        </tex_coord_generation>
      </texture>

      <albedo>
        <texture>
          <texture_index>0</texture_index>
        </texture>
      </albedo>
    </diffuse>
  </material>

Access to elevation angles in BRDF shaders



This allows users to create fancy colour changing paints, that vary colour based on viewing angle.

There are two new functions added to the shading language, minCosTheta and maxCosTheta.

minCosTheta returns the minimum of the cosines of the zenith angles of the incident and exitant vectors of a ray scattering off a surface.

Likewise, maxCosTheta returns the maximum of the two.

You can use the return values of these functions to blend between different colours, like in the example below.

E.g.

Code: Select all

<material>
		<name>default</name>

		<phong>
			<diffuse_albedo>
        <shader>
          <shader>
            <![CDATA[	
						def eval(vec3 pos) vec3 :
                  lerp(
                    vec3(1.0, 0.0, 0.0),
                    vec3(0.0, 1.0, 0.0),
                    maxCosTheta()
                   )
						]]>
          </shader>
        </shader>
      </diffuse_albedo>

      <exponent>
        <constant>
          10000.0
        </constant>
      </exponent>
		</phong>
  </material>

More shader lang functions
normalWS() Returns the shading normal for the current surface point in world space.

posOS() Returns the current surface point in object space coordinates.

Heterogeneous participating media
Participating media is stuff that scatters light - smoke, dust, snow, wax etc..
Heterogeneous means that the medium's properites can vary with position, meaning the medium can be denser in some places and less dense in others. Or it can vary in absorption colour etc..

This feature is pretty much still experimental, especially for non-uniform scattering coefficients.
Some parameters of Indigo media can now be controlled with shaders - absoprtion coefficient and scattering coefficient.

This is a pretty powerful feature, and should be able to create some pretty awesome effects.

Example:

Code: Select all

  <medium>
    <name>medium</name>
    <precedence>20</precedence>

    <basic>
      <ior>1.0</ior>
      <cauchy_b_coeff>0.0</cauchy_b_coeff>

      <absorption_coefficient>
		  <shader>
        <shader>
          <![CDATA[	
						def eval(vec3 pos) vec3 :
							vec3(fbm(pos, 3) * 10.0)
						]]>
        </shader>
          <wavelength_dependent>false</wavelength_dependent>
        </shader>
      </absorption_coefficient>
    </basic>
  </medium>

[Zom-B]

cool!

here the missing changelog:

Code: Select all

2.4.3
* Removed dependency of indigo_console on nvcuda.dll
* Made network manager stuff a bit less verbose.

[Pibuz]

Thank you guys!
I'll test this boy in minutes!

[suvakas]

Thanks!
Console and network rendering is working Ok now.

[pixie]

Working fine! :D

[OnoSendai]

Added some more feature descriptions!

[Zom-B]

Whoohoo... this is quite awesome!

Now please, please with sugar on top, let fused work on Cindigo to include some of the new controls ^^

[ENSLAVER]

The Heterogeneous participating media looks awesome.

[Zom-B]

minCosTheta returns the minimum of the cosines of the zenith angles of the incident and exitant vectors of a ray scattering off a surface.

So using this I can generate a ISL based "fresnel gradient"!? This will become very awesome for carpaints and complex blendmaterials!!

I tried using this shader for blending too, but get following error:

Code: Select all

SceneLoaderExcep: Couldn't find function eval()

btw: would it be possible to include the line in the igs where the Error appeared into this kind of messages?


Also existentially found out that a "empty function like this let Indigo crash & Burn:

Code: Select all

def eval() real :
(
)

[Zom-B]

I had some problems with the Heterogeneous participating media.
The following code let Indigo Crash & Burn:

Code: Select all

            <absorption_coefficient>
				<shader>
                    <shader>
                        <![CDATA[def Vec3ToReal(vec3 v) real : ((doti(v) + dotj(v) + dotk(v)) / 3.0)
					   def noise01(vec3 v) real : ((noise(v) + 1.0) * 0.5)
					   def noise01(vec2 v) real : ((noise(v) + 1.0) * 0.5)
					   def noise01(real r) real : ((noise(r) + 1.0) * 0.5)
					   def fbm01(vec3 v, int oc) real : ((fbm(v,oc) + 1.0) * 0.5)
					   def fbm01(vec2 v, int oc) real : ((fbm(v,oc) + 1.0) * 0.5)
					   def fbm01(real r, int oc) real : ((fbm(r,oc) + 1.0) * 0.5)
					   def mulCompVec2(vec2 a, vec2 b) vec2 : vec2(doti(a) * doti(b), dotj(a) * dotj(b))
					   def mulCompVec3(vec3 a, vec3 b) vec3 : vec3(doti(a) * doti(b), dotj(a) * dotj(b), dotk(a) * dotk(b))
def eval(vec3 pos) vec3: (lerp(vec3(1.00000, 1.00000, 1.00000), vec3(0.00000, 0.00000, 0.00000), noise01(mulCompVec2(getTexCoords(0) * 10.0, vec2(1000.00000, 1000.00000)))) * 1.00000)]]>
					</shader>
						<wavelength_dependent>false</wavelength_dependent>
				</shader>
            </absorption_coefficient>

[fused]

a point inside a volume probably doesnt have a texture coordinate

better use 3d noise

[fused]

Also existentially found out that a "empty function like this let Indigo crash & Burn:

Code: Select all

def eval() real :
(
)

a function should return a value. but indigo should also give an error

[Zom-B]

better use 3d noise :)

well yes, totally forgot about this -.-

Now I played around a little with my first result of some 3D noise inside a few spheres:

Can you guys see the spooky face in the big sphere to O_o' ???


Here another quick render with IOR = 1:

[fused]

uhhhhhhh nice tests

you should try scattering to make some clouds

edit: that normal test is pretty cool too

[CTZn]

Oh great Ono, thanks for the tests Zom-B !