 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Not sure what might be causing the spots on the image below, any ideas would be
appreciated. This is actually the best of a bad bunch of renders some of which
had lots more black spots. Render took 14 hours so trying to fix the issue is
time consuming. Might try a 3.6 render to see if it is 3.7 related..
Rendered with 3.7.
image uses focal blur and radiosity.
Focal blur settings:
#declare useFocalBlur = 4; //(0-99)
camera {
location <0, -32, -55>
direction 1.5*z
right x*image_width/image_height
look_at <0, -25, 40.0>
#if(useFocalBlur)
aperture 0.005
blur_samples useFocalBlur*2
focal_point <-22, -23, 20>
confidence 0.9+(0.001*useFocalBlur)
variance 1/(800*useFocalBlur)
#end
}
Images uses following radiosity settings, I am loading pre-calculated radiosity
using +RF"filename" and +RFI:
#declare useRad = 1;
#declare radQuality = 10;
#declare PRETRACE_END=2/image_width;
global_settings {
assumed_gamma 1
max_trace_level 25
#if(useRad)
ambient_light 0
radiosity {
#if (useRad=2)
pretrace_start 1
pretrace_end PRETRACE_END
error_bound 0.5
brightness 0.7
count 5*radQuality
nearest_count 15
low_error_factor .8
gray_threshold 0
recursion_limit 2
adc_bailout 0.01/2
media on
normal on
#end
#if(useRad=1)
// load settings
pretrace_start 1 pretrace_end 1
error_bound 0.5
brightness 0.7
recursion_limit 2
media on
normal on
always_sample off
#end
}
#end
}
Post a reply to this message
Attachments:
Download 'blackspots.jpg' (276 KB)
Preview of image 'blackspots.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Not sure what might be causing the spots on the image below, any ideas would be
> appreciated. This is actually the best of a bad bunch of renders some of which
> had lots more black spots. Render took 14 hours so trying to fix the issue is
> time consuming. Might try a 3.6 render to see if it is 3.7 related..
>
I've found another black spot in the trees...
Do you use transparency? If so, try increasing max_trace_level around 100.
Do your tests without focal blur. Faster and will show if it's related
to foral blur.
Also, do a test without radiosity to verify if it's radiosity related or
not.
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Alain <aze### [at] qwerty org> wrote:
> I've found another black spot in the trees...
> Do you use transparency? If so, try increasing max_trace_level around 100.
>
> Do your tests without focal blur. Faster and will show if it's related
> to foral blur.
>
> Also, do a test without radiosity to verify if it's radiosity related or
> not.
>
>
>
> Alain
Hi Alain,
OK, the no focal-blur and no radiosity images did not have black spots so it
seems like a combination of the two.
I do have transparency and reflection in the scene but am not sure about the
max_trace_level as I thought there is normally a warning output during tracing
if max_trace_level has been reached and I did not see any warnings.
However, I have increased the max_trace_level to 100 and have re-calculated
radiosity data with some improved settings, the no AA trace looked like it was
black in several areas but due to lack of AA it is difficult to tell for sure. I
am re-running a trace with loaded data and AA 0.01 to see if I get any black
spots.
I am starting to suspect this may be a memory problem (not limit but physical)
as my PC seems to be periodically hanging during traces. I think some more tests
are in order..
Sean
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 7/29/2011 6:00 AM, s.day wrote:
> Not sure what might be causing the spots on the image below, any ideas would be
> appreciated. This is actually the best of a bad bunch of renders some of which
> had lots more black spots. Render took 14 hours so trying to fix the issue is
> time consuming. Might try a 3.6 render to see if it is 3.7 related..
other than the barely noticeable black spots that is one incredible
image. Almost indistinguishable from a photo. Great work!
--
~Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Mike Raiford <"m[raiford]!at"@gmail.com> wrote:
> On 7/29/2011 6:00 AM, s.day wrote:
>
> other than the barely noticeable black spots that is one incredible
> image. Almost indistinguishable from a photo. Great work!
>
>
> --
> ~Mike
Thanks, I cannot take too much credit for this so far as it is mostly generated
by arbaro and placed randomly with macros. I am hoping to fill the sky with hot
air balloons but the trace time is really slow so any changes are taking a long
time to implement.
I would not mind if the black spots were like the ones on the previous image but
they appear to be random and often much worse (not what you want to see after
running a render for several days.
Sean
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Mike Raiford<"m[raiford]!at"@gmail.com> wrote:
>> On 7/29/2011 6:00 AM, s.day wrote:
>>
>> other than the barely noticeable black spots that is one incredible
>> image. Almost indistinguishable from a photo. Great work!
>>
>>
>> --
>> ~Mike
>
> Thanks, I cannot take too much credit for this so far as it is mostly generated
> by arbaro and placed randomly with macros. I am hoping to fill the sky with hot
> air balloons but the trace time is really slow so any changes are taking a long
> time to implement.
>
> I would not mind if the black spots were like the ones on the previous image but
> they appear to be random and often much worse (not what you want to see after
> running a render for several days.
>
> Sean
>
>
During the placement and test phase, replace your grass by a simple
texture and don't use radiosity. replace the trees by place holders. It
will be MUCH faster...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
