"dan B hentschel" <dan### [at] alumritedu> wrote...
> I'm assuming this is a typo. Since the ratio needs to be 5:1, then it
needs
> to be either spacing 0.1; radius 0.5 OR spacing 0.01; radius 0.05. I'm
> guessing the latter. It also seems that with the above (#2) comment, my
> resulting block should look something like:
> photons
> {
>   spacing 0.01
>   radius 0.05, 5
>   autostop 0
> }

Almost.  I meant that you should use either #1 or #2.

So, if you want to use count, then do something like this
photons
{
  count 100000000  // use more photons, because POV's estimate is off
  radius ,5  // tell POV to multiply it's radius guess by 5
  autostop 0
}

Alternatively, you can remove two of the guesses that POV does:
photons
{
  spacing 0.1
  radius 0.5
  autostop 0
}

In the first example, POV will take a look at your count and try to estimate
the angular density of photons that it sends from the light sources,
attempting to get as close to the final count as possible.  Depending on the
scene, this estimate can be very poor.  Next, POV shoots the photons using
that angular density.  POV then analyzes the spacial density of the photons
that have been deposited on the surfaces and tries to choose a good radius
to use when gathering photons.  Depending on the distribution of the
photons, this is also susciptable to poor results.

In the second example, you specify a target spacial density.  POV computes
an angular density for shooting photons in an attempt to get close to the
desired spacial density.  Generally, this is a relatively good estimate
(though it can be poor sometimes).  Because you specify the gather radius,
POV doesn't have to try to estimate anything there.

The numbers for spacing and radius are given in the same units, so they are
directly related to each other.  For this reason, if you choose to hard-code
the gather radius it is usually best to specify the spacing manually, too.

> Yes, I have read, and re-read the FAQ, but comments here have helped to
> clarify some of my questions. Most importantly, I was under the impression
> that count and spacing were interchangeable, in other words, increasing
the
> count is pretty much the same thing as decreasing the spacing. This
doesn't
> seem to be the case. I really appreciate the help, and I will let you know
> the results of my experiments.

Actually, count and spacing are basically interchangable.  I suggest using
spacing instead of count because it makes it easier to achieve the desired
"5:1" ratio of gather radius to average spacing between photons.

-Nathan

Post a reply to this message