- From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
- Date: Fri, 31 Jul 2009 16:17:03 +0200
- To: www-svg@w3.org
Hello, for light source 'feDistantLight' http://www.w3.org/TR/SVG11/filters.html#feDistantLight is specified: ' azimuth = "<number>" Direction angle for the light source on the XY plane, in degrees. If the attribute is not specified, then the effect is as if a value of 0 were specified. Animatable: yes. elevation = "<number>" Direction angle for the light source on the YZ plane, in degrees. If the attribute is not specified, then the effect is as if a value of 0 were specified. Animatable: yes. ' Respectively in the 1.2 draft: http://www.w3.org/TR/2007/WD-SVGFilter12-20070501/#feDistantLightElement ' 7.2 Light source 'feDistantLight' Attribute definitions: azimuth = "<number>" Direction angle for the light source on the XY plane (clockwise), in degrees. If the attribute is not specified, then the effect is as if a value of 0 were specified. Animatable:yes. elevation = "<number>" Direction angle for the light source on the YZ plane, in degrees. If the attribute is not specified, then the effect is as if a value of 0 were specified. Animatable:yes. ' I think, this specifies two directions, not just one (typically implementations behave differently, more in the usual/intuitive way having only one direction), therefore this needs some explanations: Is it correct, that the azimuth = "0" is the direction of the x axis? (I think yes, if elevation is 0) Is it correct, that the elevation = "0" is the direction of the y axis? (I think in general no and not in the z axis too, it is in y axis only if azimuth is 90) Is it correct, that the z direction points towards the person viewing the content as noted in (Filter 1.2) 7.3? Does a positive elevation correspond to a light source in the hemisphere with positive z or elevation = 90 means the light source is on the z-axis? -> if azimuth is an angle in the XY plane and elevation an angle in the YZ plane, how do they create a direction for the light source? What is the direction in usual spherical coordinates (formula? I think the azimuth is or can be the same as in spherical coordinates, but the elevation seems to be not related to the polar angle). -> note, that the formulas given in (Filter 1.2) 13 for an infinite light source fit to the geographic notation of spherical coordinates, in this case is elevation = 90 - polar angle/degree But then normaly the elevation angle is not the direction angle for the light source on the YZ plane. The polar angle is the angle between the z-axis and the the direction to the light source. The elevation angle is given with the formula above or a little bit more complicated description including projections. -> using the same formulas, azimuth is not the 'Direction angle for the light source on the XY plane (clockwise), in degrees.' Lets assume L is a unit vector (from the current point) to the light source and P is (Lx, Ly, 0) the vector of the projection of L on the XY plane, then azimuth is the angle between the x direction and P. If Ly is zero and Lx is negative, the angle is 180 degrees, else if Ly is not negative, the angle is not negative too, else the angle is negative (optionally 360 degrees can be added to get a positive angle again). -> maybe useful to resort some sections for better understanding and to refer for the definition of azimuth and elevation to some better understandable formulas and to fix the wording. Best wishes Olaf
Received on Friday, 31 July 2009 14:21:39 UTC