I have taken Eric Freeman's demo scene for MegaPOV and eliminated all of the
scene contents except the mirrors.  I have seen two anomalies.

First, the initial path from the source to the first mirror shows as
expected for the spotlight.  This portion of the light path is missing from
the cylinder light version.

Second, both the spotlight and cylinder light versions show what seems to be
the presence of a beam along the optical axis of the light source.

I have uploaded files to the beta-test.binaries area.  The files include the
source and .PNG files for the original version, the spotlight version, and
the cylinder light version.  The original source has had the line "#version
unofficial MegaPov 0.4;" removed and the "$"s replace with "#declare "s.

Kynar

Post a reply to this message

"Kynar" <kyn### [at] homexpresswaynet> wrote in message
news:3cf36f09$1@news.povray.org...
>
> First, the initial path from the source to the first mirror shows as
> expected for the spotlight.  This portion of the light path is missing
from
> the cylinder light version.
>
> Second, both the spotlight and cylinder light versions show what seems to
be
> the presence of a beam along the optical axis of the light source.

First of all, cylinder light sources act a bit differently. So, try the
following in place of that:

// cylindrical laser lite
light_source {
  <150,0,0>, color rgb <0,1,0> *15*100
 // spotlight // not needed
  cylinder
  parallel // goes ahead of point_at
  point_at 0
  radius 1/2*9
  falloff 1/2*9
  tightness 1
  media_attenuation off
  photons {
    refraction off
    reflection on
  }
}

Secondly... well, admittedly, I don't know what's going on with that axial
anomaly but it's very possible it could just be a matter of media and/or
photon settings. Although, I couldn't get it to go away in the short time I
was attempting to try other numbers and such.

BTW, default media method 3 is supposed to work best when using intervals 1
and more will automatically be used if needed. Also, sampling isn't
"adaptive" unless the minimum 'samples' is 3 or more; it ignores the max
value.

bob h

Post a reply to this message