[Bf-committers] Re: [Bf-blender-cvs] CVS commit: blender/release/scripts uv_vcol_to_image.py blender/release/scripts/bpymodules BPyRender.py

[Campbell Barton]

jmsoler at free.fr wrote:
> Selon jmsoler at free.fr:
>
>   
>> Selon Campbell Barton <cbarton at metavr.com>:
>>
>>     
>>> campbellbarton (Campbell Barton) 2006/07/26 04:07:12 CEST
>>>
>>>   Added files:
>>>     blender/release/scripts uv_vcol_to_image.py
>>>     blender/release/scripts/bpymodules BPyRender.py
>>>
>>>   Log:
>>>   Added a script that makes an image from vertex colors using UV coords.
>>>       
>> the
>>     
>>> advantage this has over other existing scripts is that it bleeds the colors
>>> from edge polygons so you dont get mipmapping errors.
>>>   usefull for baking radiosity back into an image (with non overlapping
>>>       
>> UV's)
>>     
>>> and also has an option for including the texface image, and using normals
>>> (instead of colors)
>>>       
>> There is already 2 scripts to  do that in te bundle scripts .
>> Unfair method but it is not the first time .
>>     
>
> Tested .  To make the "bleeding", this script creates one copy per egdge of the
> original mesh object 
Right, it seems stupid,
A while ago Theeth added this functionality in C (Still in tuhopuu) and 
never quite got the bleeding woking (Though everything else worked great).
Anothe example of bad bleeding is in models we buy that were made in 3ds 
Max. these have bleeding enabled with (I assume) a commercial plugin, 
and have the same kind of artifacts, they are noticable and ugly..
bleeding in Python per pixel would be slower then rendering multiple 
objects.

Basicly Blenders optimized for this stuff and python inst. - and Id 
rather have a working bleeding then artifacts.

- Cam

-- 
Campbell J Barton

133 Hope Street
Geelong West, Victoria 3218 Australia

URL:    http://www.metavr.com
e-mail: cbarton at metavr.com
phone: AU (03) 5229 0241