Author Topic: Spotting with Abbe  (Read 3781 times)

0 Members and 1 Guest are viewing this topic.

Offline NDenekamp

  • Full Member
  • ***
  • Posts: 148
Spotting with Abbe
« on: January 21, 2014, 03:57:18 am »
Hi guys,

I was testing the use of optics on LED's in Keyshot, when I came across some odd behaviour when using emissive materials in combination with Abbe refraction.

The optic I am using here I have downloaded for Carclo: http://www.carclo-optics.com/optic-10417?ledmanufacturer=lumileds&ledfamily=z%20es&opticfamily=10mm&beamangle=1%20to%2010%20degrees (see IGS file)

This is a PMMA total internal reflector lense that culminates the led light source into a spot beam. nifty little things.

For the LED i'm using a Philips Lumileds Rebel Z which I've modelled for another project. http://www.philipslumileds.com/products/luxeon-z-es

In the first image I rendered the LED shining through the optic by applying a diffuse light material to the light emitting area of the chip. The Optic is a Dielectric material with IOR of 1.491. The box around it is a white diffuse material. You see there is some noise left on the faces on the cube, but also some slight spotting visible around the edges of the optic.

In the second image in addition to the IOR of PMMA, I tested also dialling in the Abbe value of 52.6. This time a lot of colourful dots appear on the walls of the box, and a reduction in overall light intensity.

This is mainly just an observation. I"m not saying I necessarily want to render optics with such accurate dispersion (although it would be nice, and the little bit of colour glint in plastics when doing animations looks awesome). Just wanted to make you aware of this.


n



Some further details:

The modelling units are set to MM
The LED is emitting 2.5 lumen (the real thing goes up to 256 Lm)
The colour temperature of the LED is 6500K
The environment is


thomasteger

  • Guest
Re: Spotting with Abbe
« Reply #1 on: January 21, 2014, 05:49:28 am »
You should use the actual light materials like area, point or IES light, and not the emissive material. Can you try ad share the results?

Offline NDenekamp

  • Full Member
  • ***
  • Posts: 148
Re: Spotting with Abbe
« Reply #2 on: January 21, 2014, 07:51:49 am »
Sorry, by 'diffuse light material'I meant the 'area' light material.


thomasteger

  • Guest
Re: Spotting with Abbe
« Reply #3 on: January 21, 2014, 11:24:28 am »
And you are using 4.3?

Offline NDenekamp

  • Full Member
  • ***
  • Posts: 148
Re: Spotting with Abbe
« Reply #4 on: January 21, 2014, 07:30:54 pm »
Yes, 4.3.10

Offline PhilippeV8

  • Hero Member
  • *****
  • Posts: 1914
    • PV8-Portfolio
Re: Spotting with Abbe
« Reply #5 on: January 27, 2014, 04:56:17 am »
I keep getting grainy bright pixels in the rendering I'm working on now too.  Working with area light diffuse and the other materials.
Rendering is so slow however that I can't tell if it's just due to lack of render time ... (at this moment 0.3 FPS)

Offline PhilippeV8

  • Hero Member
  • *****
  • Posts: 1914
    • PV8-Portfolio
Re: Spotting with Abbe
« Reply #6 on: January 27, 2014, 05:40:02 am »
In my case, turning off "global illumination" gives me back the FPS and seems to fix the issues while still getting a rather good look overall .. I think I'll stick with this setting .. 29 FPS now.