Thomas de Groot <tho### [at] degrootorg> wrote:
> Great minds think alike ;-) My hammer is also an Estwing, and been with
> me for the last 50 years. Never lost it happily. It now gets a bit rusty
> in places... I am now modelling it for the scene; first a low detail
> version, but I intend to make a high resolution one later on for a
> projected close up of sorts. The least I can do for its faithful
> services :-)

Good brand! I've had mine for about 14 years or so now, and it's badly in need
of a second sharpening already. (The ground doesn't dig itself... well sometimes
it does.) I'm looking forward to seeing your model when you finish it!

> I am curious to see what you mean. I am indeed using radiosity. The
> isosurface max_gradient is pretty low, much lower than recommended by
> POV but giving the details I intend to get. With a lower count, I may
> need to increase the max_gradient though.

I'm not sure increasing the max_gradient is necessary, as all that does is fill
in the holes. If anything, you may need to lower the isosurface's accuracy if
you intend to lower the radiosity's count.

One of my isosurface tests has been posted to the TC-EOG-RTC thread:

http://tinachepforum.forumup.co.uk/viewtopic.php?p=3636&mforum=tinachepforum#3636

That image rendered in 5 minutes 39 seconds, which isn't too bad, all things
considered.

Here's the scene's radiosity block:

radiosity{
  count 1
  error_bound .1
  recursion_limit 1
  nearest_count 1
  normal on
}

some isosurface settings:

accuracy 0.001
max_gradient 10

and the command line parameters:

+fn +f +a0.3 +am2 +r2 +j0.2

I can post the entire source (with include files) if needed.

An important thing to keep in mind is to always have enough small details on
your surfaces by using things like surface normals and/or and low isosurface
accuracies. Low aa settings help a great deal as far as speeding things up as
well.

Sam

Post a reply to this message