Re: [SVG1.1] feSpotLight's limitingConeAngle not mentioned in lighting equations

On Mon, 04 Jun 2007 19:26:25 +0200, Tim Rowley <tor@acm.org> wrote:

>
> There is a limitingConeAngle specified as an attribute and DOM
> property of the feSpotLight element, but the lighting specification in
> the feDiffuseLighting and feSpecularLighting sections do not include
> it in the equations given.

The specification already says how limitingConeAngle should behave[1],  
although as you note it does not provide an equation. If you check the DTD  
(or search the document) for spotlight you will find that it can only be a  
child of an feDiffuseLighting or feSpecularLighting element. Now, in the  
case of feDiffuseLighting it says that the variables Lr,Lg,Lb are  
functions of x and y for the feSpotLight case, and for feSpecularLighting  
it refers back to feDiffuseLighting for the definition of Lr,Lg,Lb. So  
what you are looking for is the equations for the Lr,Lg,Lb which are  
functions of x and y. If the limitingConeAngle is not specified, you would  
always get non-zero values back from these functions. When it is specified  
these functions return 0 for everything that is outside of the cone. The  
specification recommends that smoothing of the edges of the cone is done,  
but doesn't say how. That may affect the equation as well, depending on  
how you want to implement the edge smoothing.

The SVG WG discussed this issue, and decided that if this resolves your  
issue, examples will be added to illustrate the limitingConeAngle cases.
If this is not satisfactory please let us know if you have a better  
resolution for us to consider.

Best regards
/Erik, on behalf of the SVG WG

[1]  
http://www.w3.org/TR/SVG11/filters.html#feSpotLightLimitingConeAngleAttribute
[2] http://www.w3.org/TR/SVG11/filters.html#feDiffuseLighting

-- 
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

Received on Monday, 25 June 2007 09:34:30 UTC