Am 13.04.2017 um 04:56 schrieb Cousin Ricky:
> On 2017-04-12 10:45 PM (-4), Cousin Ricky wrote:
>> I am working on a diamond cut.  A while ago, in comparing ior.inc
>> dispersion with Lilysoft's SpectralRender ("Native dispersion v.
>> spectral render" in p.b.i), I noticed a discrepancy in the apparent
>> index of refraction of BK7 glass between the two renders.  So, just to
>> make sure that wasn't a fluke, I tested both methods with a diamond
>> prism.

The `ior.inc` include file isn't perfect, and it can't be, for
fundamental reasons:

The `ior` parameter is computed so that when used without dispersion,
you get the correct refraction angle at what is typically the peak
brightness (yellow, about spectral line D) (1).

The `dispersion` parameter is computed so that you get the right
difference in refraction angles between the red and violet ends of the
spectrum (2).

POV-Ray, on the other hand, naively presumes that the refractive index
of dispersive materials is a linear function of wavelength varying from
IOR-CONST to IOR+CONST, implying that IOR is specified at the wavelength
halfway between red and blue, meaning you can't get both properties (1)
and (2) at the same time.

To fix that, POV-Ray itself would have to be amended. But that would
still be a kludge, since refractive indices typically aren't a linear
function of wavelength anyway.


Another reason for `ior.inc` values differing from SpectralRender may be
that some materials exhibit a rather wide variation in refractive index.
To my knowledge, SpectralRender's dispersion is based on measurements of
individual samples, while the data for `ior.inc` was taken from
literature, which can be presumed to represent average values for the
corresponding materials.

Post a reply to this message