Any Colour You Like

[Skywise]

My first contribution to the gallery.

gallery2.php?g2_itemId=19801

Brian

[B.Y.O.B.]

I like these renders

What light did you use for the input beam (sunlight, blackbody, RBG 1,1,1, ...)?
And how did you display all the beams, is there fog in the air or are they on a surface?

[wouter215]

could you please post the .blend file?

[Skywise]

I like these renders

What light did you use for the input beam (sunlight, blackbody, RBG 1,1,1, ...)?
And how did you display all the beams, is there fog in the air or are they on a surface?

I used the lamp spectrum texture, "Natural Daylight". It's an area light using the 'laser' method and ies file from this thread: viewtopic.php?f=36&t=6485

To make the light visible, I used a homogeneous volume for all exterior volumes, "Air @ STP" preset, scattering HSV value of about 0.02, and 0.1 for the scattering scale.

I used a homogeneous volume for the Glass2 prism, "Glass Fused Silica" preset, scattering HSV value of about 0.1, and 0.1 for the scattering scale. I also applied a Cauchy texture with 1.42 IOR and a 'B' value of about 0.07 to get enough dispersion. Setting the IOR and B values were a matter of trial and error in order to match the album cover. (in case anyone doesn't recognize the image, I emulated the album cover of Pink Floyd's Dark Side of the Moon)

The green heartbeat object also uses a lux emission material, and there is also a dim ambient mesh light off to the top right out of frame to give a slight wash to the matte black background. I have all three light sources on their own light groups so I could adjust them to taste in Lux, along with manual linear tonemapping settings.

could you please post the .blend file?

I would share this file but I want to keep my commercial options open for a larger and improved version of the image. I was so impressed with the results so far in the project that I wanted to share the results. I really think this showcases Lux's ability to work with true spectral dispersion and caustics, which is one of the factors that brought me to this render engine.

But tell ya what. I'll make a new file that demonstrates these settings so you can still learn from what I found. That ok? I'll attach it to this thread within a day or two.

Brian

[B.Y.O.B.]

Thanks for the description

[wouter215]

But tell ya what. I'll make a new file that demonstrates these settings so you can still learn from what I found. That ok? I'll attach it to this thread within a day or two.

i'm mostly interested in the prism

[Skywise]

OK, here's a zip. It contains a .blend, the .ies file, and a sample .png.

The prism object is pretty easy to make. Just make a cylinder and reduce the segments to 3.

If you don't want the light splash inside the prism, just change it's material from homogeneous to clear.

Let me know if you have any problems or questions.

Brian

prismdemo.zip

(232.48 KiB) Downloaded 76 times

[hoodedmanwithsythe]

LOL a "Day Light Laser", not exactly what I had in mind when I wrote that IES file., but it's a cool effect.

I'd like to see someone set up a physics experiment type scene some time, if I had the time I might do something myself.

[wouter215]

thanks a lot for this
edit:

after on hour of rendering i got this:

prismdemo.Scene.00001.png (1.13 KiB) Viewed 2265 times

does it take very long to render or is something wrong with my setup?

[Skywise]

after on hour of rendering i got this:

does it take very long to render or is something wrong with my setup?

That sample image only took me about 45 minutes, give or take. Don't recall exactly. Definitely under an hour to get to 1000 s/P on my quad core Core2Duo 3.0Ghz under WinXP.

I'm still fairly new to both blender and lux, so I'm not sure what settings get saved or not. As far as I know, it seems everything gets saved. Although we may be dealing with differences in versions. I'm using Blender 2.6, lux 0.8, and luxblend for 2.5.

A first thought is to check in your render settings, under volume integrator, that it is set to 'single'. I know that if 'multi' is used, the volume scattering takes much longer to integrate. However, I tested on my system with 'multi' and I still had a recognizable image in 30 seconds even though it was much less than with 'single'.

Next, check the 'laser source' hemi light settings. Check that the 'IES Data' path is correct. Since the .ies file is in the same folder as the .blend file, it should be "//laser-final.ies". Even if it looks OK, try resetting it. ISTR having to do that before. If the .ies file can't be found, you don't get the laser 'beam'.

Beyond that, we may have to compare other settings step by step. But we'll figure it out.

Brian