On Aug 8, 2009, at 11:14 PM, Harry Underwood wrote:

So what you're also saying is that the combination of CSS 3D Transforms in Animations with SVG (resulting, I'm assuming, in SVG- transformed animation through CSS styling) does not translate or correlate to what either WebGL or O3D are meant to do, even though O3D and SVG are both retained mode APIs?

And you're also saying that it isn't the intention of WebKit's implementation (or Apple's Working Draft) of SVG Transforms (as per http://www.w3.org/TR/SVG-Transforms/ ) to allow for the construction and styling of full scenes and models like those in O3D? The draft has a combination of CSS transitions, CSS Animations and CSS 3D to plug into SVG, so it would seem like any composition using SVG Transforms could accomplish mostly with CSS what O3D accomplishes with JavaScript (unless I'm not reading it correctly).


As has been said before, O3D is much more related to WebGL than to CSS Effects or SVG. If you look at the O3D Beach Demo (http://www.youtube.com/watch?v=uofWfXOzX-g ) you can see that O3D (and WebGL) is used to render 3D textured and lit scenes. It has the upside of being very efficient at showing very high quality and high polygon count 3D objects, using the modern shading techniques available on most of today's graphics cards. But it has the downside that it does now really live in the same DOM as the rest of the page. So you could not, for instance, show an HTML page in the scene (not yet, anyway). And you can't use CSS to apply style, and things like that.

CSS Effects can be used to make simple 3D object, like cubes, and even spheres if you REALLY worked at it. It has the upside of being fully integrated with the HTML DOM, but the downside that doing anything more complex than the simplest shape is really hard, and the "scene" is not lit or textured, or shaded in any way.

So they really are two distinct capabilities.

-----
~Chris
cmar...@apple.com




_______________________________________________
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Reply via email to