|
|
I changed the settings to
global_settings {
radiosity {
pretrace_start 0.08 // start pretrace at this size
pretrace_end 0.04 // end pretrace at this size
count 60 // higher -> higher quality (1..1600) [35]
nearest_count 5 // higher -> higher quality (1..10) [5]
error_bound 1.8 // higher -> smoother, less accurate [1.8]
recursion_limit 5 // how much interreflections are
calculated (1..5+) [3]
low_error_factor .5 // reduce error_bound during last pretrace
step
gray_threshold 0.0 // increase for weakening colors (0..1)
[0]
minimum_reuse 0.015 // reuse of old radiosity samples [0.015]
brightness 1 // brightness of radiosity effects (0..1)
[1]
// adc_bailout 0.01/2
adc_bailout 1/64
normal on // take surface normals into account [off]
}
max_trace_level 26
This seems satisfactory to me, most disturbing black spots seem to be gone
now.
I plan to do an animation with this scene (hopefully ready by christmas
2007), so I guess I'll have to reduce the trace level to something like 17
make it faster
Thanks for the advice.
> Each time you reflect on a sphere, the trace level goes up by 3: 1 getting
> into
> the transparent outer shell, 1 relecting on the iner reflective part, and
> 1 more
> geting OUT of the transparent shell.
> If the reflection hit 2 balls, it register 6 levels, 9 if it hits 3 balls.
> Add
> one to reatch the background or any surrounding object.
> So, for one ball, you need a Max_trace_level of at least 4. For 2, you
> need at
> least 7, and 10 if you hit 3.
> In your case, I'd go up to arount 15 or 17. For that scene, I'd go up to
> the max
> of 255 and set adc_bailout to 1/128 or even as large as 1/64.
> Note that there are many more max_trace_level related black spots than
> those you
> spotted. I can find at least 8 more, and I probably missed some.
Post a reply to this message
Attachments:
Download 'HDRItest3.jpg' (148 KB)
Preview of image 'HDRItest3.jpg'
|
|