 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
I have a problem with radiosity at 3400x1920 resolution.
I use 2 pass radiosity: The first pass has deactivated transparency and
reflections. In this pass there is no problem with radiosity and I save
file information "scene.rad".
In the second pass I activate transparency and reflections. If I render a
test at 1700x960 all is correct, but when I render at 3400x1920 there is
splotchiness.
Source code:
global_settings {
ambient_light 1.0
assumed_gamma 2.2
exposure 1.6
max_trace_level 6
#if (RAD)
radiosity {
brightness 2.0 //[1.0]
recursion_limit 2 //[3]
count 500 //[35]
error_bound 0.1 //[1.8]
gray_threshold 0.8 //[0.0]]
#if(SAVERAD)
save_file "scene.rad"
#else
pretrace_start 1
pretrace_end 1
load_file "scene.rad"
always_sample off
#end
}
#end
}
Any comments??
Regards,
Joel.
Post a reply to this message
Attachments:
Download 'image.jpg' (234 KB)
Preview of image 'image.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"JSR" <jsr### [at] gmail com> wrote:
> Hi,
> I have a problem with radiosity at 3400x1920 resolution.
> I use 2 pass radiosity: The first pass has deactivated transparency and
> reflections. In this pass there is no problem with radiosity and I save
> file information "scene.rad".
> In the second pass I activate transparency and reflections. If I render a
> test at 1700x960 all is correct, but when I render at 3400x1920 there is
> splotchiness.
>
> Source code:
>
> global_settings {
> ambient_light 1.0
> assumed_gamma 2.2
> exposure 1.6
> max_trace_level 6
>
> #if (RAD)
> radiosity {
> brightness 2.0 //[1.0]
> recursion_limit 2 //[3]
> count 500 //[35]
> error_bound 0.1 //[1.8]
> gray_threshold 0.8 //[0.0]]
>
> #if(SAVERAD)
> save_file "scene.rad"
> #else
> pretrace_start 1
> pretrace_end 1
> load_file "scene.rad"
> always_sample off
> #end
> }
> #end
> }
>
> Any comments??
>
> Regards,
>
> Joel.
Hi Joel.
I am absolutely not a radiosity expert, lightyears from it, and I was
writing my post concerning a problem I have with rad when you posted yours!
At first look, I would increase the error_bound parameter. Too a low value
implies risk of splotchiness if the count parameter is not high enough. As
I think a count of 500 is very enough for your scene, try increasing
error_bound up to 0.5 or so, and decrease step by step if you really need
it. You may also play with pretrace_start and pretrace_end, the latter
could be determinant. You use default values (0.08 and 0.04 respectively).
Perhaps you could try one more pretrace step by setting pretrace_end to
0.02.
However, the problem might be something more subtle, requiring more
expertise ...
Unless you are a radiosity guru, this feature is a trial-and-fail approach.
Mainly fail for me ;o) ....
Bruno
Post a reply to this message
|
|
| |
| |
|
|
From: Thomas de Groot
Subject: Re: Radiosity problem at high resolution
Date: 6 Mar 2007 08:47:04
Message: <45ed70d8@news.povray.org>
|
|
|
| |
| |
|
|
Radiosity is difficult. I learned from Skip Talbot that :
<quote>
Typically I drop the recursion limit down to 1 or 2 and jack the count
setting way up. Although a significant portion of your image is in the
shadow, and with the nooks and crannies in the architecture you are using,
keeping it at 3 is probably best. You might want to try 2 and see if you
can get away with it. The reason you'd want to lower it is to save time,
because your count is way too low. I'd bump up to 1000 at least, and at a
recursion limit of 3 thats going to take forever.
The biggest factor here might be your error bound. A value of 0.1 is very
accurate and requires a very high count (and ages to render). Maxing out at
1600 you are still going to see artifacts probably. I'd raise that up to at
least 0.5 and you might want to go all the way to 1. Your lighting won't
be as accurate but it will be much smoother. Remember, as your error bound
lowers, the count has to go up.
Nearest count could also be increased to take a few artifacts out.
<unquote>
Maybe you can take profit from this?
Also, visit Tim Nikias' tutorial at
http://www.nolights.de/projects/radiosity/radiosity.html
Hope this helps.
Thomas
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Thomas de Groot" <t.d### [at] internlDOTnet> wrote:
> Typically I drop the recursion limit down to 1 or 2 and jack the count
Following my most recent answer to Bruno's problem, I can report that
recursion_limit needs to be raised in order for radiosity to work correctly
with transparent objects. If you've got glass in your windows I'd recommend
playing with that.
Hope that helps
Bill
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
JSR nous apporta ses lumieres en ce 06-03-2007 04:10:
> Hi,
> I have a problem with radiosity at 3400x1920 resolution.
> I use 2 pass radiosity: The first pass has deactivated transparency and
> reflections. In this pass there is no problem with radiosity and I save
> file information "scene.rad".
> In the second pass I activate transparency and reflections. If I render a
> test at 1700x960 all is correct, but when I render at 3400x1920 there is
> splotchiness.
>
> Source code:
>
> global_settings {
> ambient_light 1.0
> assumed_gamma 2.2
> exposure 1.6
> max_trace_level 6
>
> #if (RAD)
> radiosity {
> brightness 2.0 //[1.0]
> recursion_limit 2 //[3]
> count 500 //[35]
> error_bound 0.1 //[1.8]
> gray_threshold 0.8 //[0.0]]
>
> #if(SAVERAD)
> save_file "scene.rad"
> #else
> pretrace_start 1
> pretrace_end 1
> load_file "scene.rad"
> always_sample off
> #end
> }
> #end
> }
>
> Any comments??
>
> Regards,
>
> Joel.
>
>
>
>
> ------------------------------------------------------------------------
>
I always wonder WHY peoples think that they need very low error_bound? (0.1 IS
very low: 1/18 default value) It increase the rendering time, requier higher
count value (500 is to low in your case) and tend to cause major slpotchiness.
It also greatly increase memory use.
If you need more samples increase count first. Try using a smaler pretrace_end,
like 0.01 or 0.005. You may set low_error_factor to a smaler value than
default, like 0.3 or somewhat less.
Upping nearest_count can help reduce splotchiness, it maxes out at 20.
You may have to increase max_trace_level by 2 or 3, maybe more, during the
second pass.
--
Alain
-------------------------------------------------
I can read your mind, and you should be ashamed of yourself.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Alain" <ele### [at] netscapenet> wrote in message
news:45edf061@news.povray.org...
> I always wonder WHY peoples think that they need very low error_bound?
www.beantoad.com/newimages/shaker16.jpg
versus
www.beantoad.com/newimages/shaker20b.jpg
If you're dealing with a situation where you want deep shadows, a lower
error bound keeps objects from floating on the floor. Unfortunately, it
results in some splotchies, so I then turned on focal blur to smooth them
out a bit. (Yes, in the above, my "sun" was a bit too yellow)
> (0.1 IS very low: 1/18 default value) It increase the rendering time,
> requier higher count value (500 is to low in your case) and tend to cause
> major slpotchiness. It also greatly increase memory use.
> If you need more samples increase count first. Try using a smaler
> pretrace_end, like 0.01 or 0.005. You may set low_error_factor to a
> smaler value than default, like 0.3 or somewhat less.
> Upping nearest_count can help reduce splotchiness, it maxes out at 20.
> You may have to increase max_trace_level by 2 or 3, maybe more, during the
> second pass.
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jeremy M. Praay nous apporta ses lumieres en ce 07-03-2007 10:17:
> "Alain" <ele### [at] netscapenet> wrote in message
> news:45edf061@news.povray.org...
>> I always wonder WHY peoples think that they need very low error_bound?
> www.beantoad.com/newimages/shaker16.jpg
> versus
> www.beantoad.com/newimages/shaker20b.jpg
> If you're dealing with a situation where you want deep shadows, a lower
> error bound keeps objects from floating on the floor. Unfortunately, it
> results in some splotchies, so I then turned on focal blur to smooth them
> out a bit. (Yes, in the above, my "sun" was a bit too yellow)
>> (0.1 IS very low: 1/18 default value) It increase the rendering time,
>> requier higher count value (500 is to low in your case) and tend to cause
>> major slpotchiness. It also greatly increase memory use.
>> If you need more samples increase count first. Try using a smaler
>> pretrace_end, like 0.01 or 0.005. You may set low_error_factor to a
>> smaler value than default, like 0.3 or somewhat less.
>> Upping nearest_count can help reduce splotchiness, it maxes out at 20.
>> You may have to increase max_trace_level by 2 or 3, maybe more, during the
>> second pass.
OK, reducing the value for error_bound help deepen the shadows, but when you
don't need to deepen your shadows, why use a low error_bound?
Also, why use very low values? You should try reducing it slowly to find the
optimum value instead of immediately jumping to a very, probably to small value,
then wondering: WHY do I have so much splotchiness?
You have a very good example of the use of a small value, but you had to do
several test before you got a good error_bound value for your scene, and you
used a large count value.
--
Alain
-------------------------------------------------
EVERYTHING HAS A GENDER
You may not know this but many nonliving things have a gender...
A Subway is Male, because it uses the same old lines to pick people up.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
