Re: [Goptical] Intensity mode

[Richard Graham]

Hi Alexandre,

Thanks for your prompt response. I appreciate it as I am working to a
deadline at the moment.

I tried adding transmittance data for my wavelength to each of the
materials I am using, though I still get the "doesn't contains enough
data" error. I checked that the data was going into the transmittance
data set.

In any case, I am not sure if the calculated intensities would be
sufficient for me. In my case I am not to concerned about the internal
transmittance losses of the materials which are negligible. I am
concerned about the reflection losses which vary with angle of
incidence. In material_base, I see only functions for finding the
transmittance at normal incidence. Should I expect the losses to vary
properly with angle of incidence, or is this not implemented?

Thanks,

--
Richard Graham


On 02/18/2013 03:37 PM, Alexandre Becoulet wrote:
> On Monday 18 February 2013 15:15:17 Richard Graham wrote:
> 
> Hi,
>  
>> I am trying to work out the intensities in my system and have some
>> confusion about the status of the intensity calculations. From the
>> documentation it seems that while taking into account polarizations is
>> not yet supported, without polarization should be, which is ok for me.
> 
> It is supported, see OpticalSurface::trace_ray_intensity in the 
> sys_optical_surface.cc source file.
>  
>> But when I do:
>>
>> tracer.get_params().set_intensity_mode(Trace::IntensityMode(Trace::Intensity
>> Trace)); tracer.trace();
>>
>> I get a goptical error "what():  data set doesn't contains enough data".
> 
> You need to add transmitance data to your material first:
> 
> http://www.gnu.org/software/goptical/manual/Material_Dielectric_class_reference.html#__624
> 
>> When I do a spot analysis I see that the intensities are always whole
>> numbers -- it seems like it is just counting intersecting rays rather
>> than actually working out the intensities. I don't see anywhere in the
>> code where the IntensityTrace mode is used.
> 
> It is used in sys_surface.cc to select a call to the trace_ray_intensity 
> virtual function.
> 
>> Are the reflection coefficents (such as worked out from Fresnel
>> equations) actually calculated anywhere?
> 
> Yes they are, see in material_base.cc. Please check the formula is ok for you 
> and let me know.
> 
> Best,
>