|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hello,
I would like to know if there is a way of reading the resulting pixel
value in the rendered image as it is being rendered and if it is
possible to overwrite it with another value. Possibly based on the
original color.
This is for testing lighting on a scene. If the color at the pixel is
above a certain value it should be marked with another color and would
let me know that the image is being saturated with too much light. Not a
scientifically rigorous method but it works well for most cases.
I can do this manually by just rendering, in EXR or HDR format, the
scene using only a simple diffuse white material for all. Then, after
rendering, load the render into a high-dynamic image viewer or editor to
see if there are any places on the scene that are too bright.
I already looked at ImageMagick but it just overkill for what I need.
FlyerX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Hello,
>
> I would like to know if there is a way of reading the resulting pixel
> value in the rendered image as it is being rendered and if it is
> possible to overwrite it with another value. Possibly based on the
> original color.
>
> This is for testing lighting on a scene. If the color at the pixel is
> above a certain value it should be marked with another color and would
> let me know that the image is being saturated with too much light. Not a
> scientifically rigorous method but it works well for most cases.
>
> I can do this manually by just rendering, in EXR or HDR format, the
> scene using only a simple diffuse white material for all. Then, after
> rendering, load the render into a high-dynamic image viewer or editor to
> see if there are any places on the scene that are too bright.
>
> I already looked at ImageMagick but it just overkill for what I need.
>
>
>
> FlyerX
Reading a rendered pixel DURING render? Not possible without patching
and recompiling.
I can propose some alternatives:
Render the scene with a lower light level, say, dividing all
light_source intensity by 10 or some other value.
Have some mirror with a reflection value of 1/8, 1/32 or less showing
the areas where you suspect are affected by clipping.
Alain
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 3/27/2014 9:43 PM, Alain wrote:
>> Hello,
>>
>> I would like to know if there is a way of reading the resulting pixel
>> value in the rendered image as it is being rendered and if it is
>> possible to overwrite it with another value. Possibly based on the
>> original color.
>>
>> This is for testing lighting on a scene. If the color at the pixel is
>> above a certain value it should be marked with another color and would
>> let me know that the image is being saturated with too much light. Not a
>> scientifically rigorous method but it works well for most cases.
>>
>> I can do this manually by just rendering, in EXR or HDR format, the
>> scene using only a simple diffuse white material for all. Then, after
>> rendering, load the render into a high-dynamic image viewer or editor to
>> see if there are any places on the scene that are too bright.
>>
>> I already looked at ImageMagick but it just overkill for what I need.
>>
>>
>>
>> FlyerX
>
> Reading a rendered pixel DURING render? Not possible without patching
> and recompiling.
>
> I can propose some alternatives:
> Render the scene with a lower light level, say, dividing all
> light_source intensity by 10 or some other value.
>
> Have some mirror with a reflection value of 1/8, 1/32 or less showing
> the areas where you suspect are affected by clipping.
>
>
>
> Alain
Thanks for the tip. I think I found an easy way to get an idea of
saturated colors: rendering in HDR/EXR format with Display_Gamma=0.01.
Any colors above 1 will still show.
FlyerX
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 29.03.2014 11:06, schrieb FlyerX:
> Thanks for the tip. I think I found an easy way to get an idea of
> saturated colors: rendering in HDR/EXR format with Display_Gamma=0.01.
> Any colors above 1 will still show.
Holy cow, here's some smart thinking!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|