Gaussian XY scattering

  • 13 January 2021
  • 4 replies

  • Infrared
  • 7 replies

We have some asymmetric scatterers in the lab that I would like to model in zemax.

They have different gaussian scattering properties in horizontal and vertical direction.

So i tried to use the Gaussian XY DLL to model it, but it behaves weired. A ray hitting the surface perpendicular is scattered as intended, but rays under even small angles are changing it completely. The scatter pattern for a ray with 1° in x-direction is twisted 90° to a ray under 1° in y-direction.

Is there a mistake or do I something wrong?

I have attached an example file.



4 replies

Userlevel 6
Badge +2

Hi Steffen

The Gaussian_XY scatter function desribes the distribution about the projected specular ray. The scattering does not occur along global X and Y axes.

This can be seen by looking at the documentation provided in the source code, as well as by reading the following article: How to create a user defined scattering function – Knowledgebase (

So when you send a collimated source without any tilts vs the scattering surface, you will find the expected distribution. But when you add a tilt, the scattering pattern will rotate with the angle of incidence.

So I can see two options: modify the current dll or create a BSDF file with your measurement. Let us know if we can help any further and do not hesitate if you have any other questions.


Hi Sandrine, thank You for Your awnser!

So this scattering function is only usable for a collimated beams...

what a pitty because in real life 100% collimated sources are rare but scatterers with asymmetric charachteristiks are plenty. And there are mostly no BSDF measurments available.

Is there a quick possability to get to a BSDF file from just two values for X and Y gaussian scatter angles?




Userlevel 6
Badge +2

Hi Steffen

Yes correct. Modifying the dll is not straight forward.

For the BSDF, you would need an external program like excel or Matlab to create the distribution.

Then for the BSDF format, have a look at those articles:

It is probably something we need to investigate as these kind of characteristics are common. 


Hi Sandrine,

thanks for the links.

Yes I am struggeling allready with excel on that... ;)

Asymmetric diffusers are quite common in Laser Beamshaping, Illumination and Display Technologies. So it might be a good idea to look into it...

It is very frustrating that the implemented Gaussian_XY DLL that is allways cited as 'the example' ist not describing anything existant in the real world, exept it is illuminated perpendicular with a collimated beam. 

Have a nice Weekend!