On 06/05/2016 01:46 AM, Le_Forgeron wrote:
> Le 05/06/2016 00:48, William F Pokorny a écrit :
>> On 06/04/2016 04:47 PM, Le_Forgeron wrote:
>>> The computation of texture, when interior texture is used, has a bug for the lower
side of the prism.
>>>
>>> Attached the scene to reproduce.
>>>
>>
>> I think you might be finding too the issue I opened for Christoph at:
>>
>> https://github.com/POV-Ray/povray/issues/65
>>
>> but not sure as the your example code is for a clipped lathe?
>
> Oups, wrong include, still an issue.
>
Thanks Jérôme. Good catch & what you are seeing is different than issue 
65. Issue 65 has to do with the bezier prism sides having normals 
flipped relative to all other spline types in prisms and all spline 
types in lathes.

It also seems I might have only looked at the prism tops in my test 
cases... I'll update those to look at both top and bottoms (both ends). 
In any case, I know this bit added to the documentation just added for 
the prisms end normals is not right for the current state of the linear 
spline version :

"The surface normal determination for prism sides depends upon the order 
in which the splines points are specified.  Prism ends have normals 
which always face outward.  The following code, where the linear spline 
point order is counter-clockwise in the x-z plane, will render sides 
with the color Red on the outside and the color Blue on the inside.  The 
prism ends will also be Red on the outside and Blue on the inside."

I'll have to update that when I know what the other spline types do on 
BOTH ends.

Bill P.

Post a reply to this message

Attachments:
Download 'linearprismtopbotnrmldiff.jpg' (43 KB)

Preview of image 'linearprismtopbotnrmldiff.jpg'