Erik Harris wrote:
On 9/6/2010 9:23 AM, David Laakso wrote:
PS In the world of geometry is that a sphere or is it known as a simple
circular object?

I think the idea is that the gradient makes it appear spherical. Since it uses a linear gradient and not a radial gradient, though, the effect is somewhat primitive.


It is our perception that makes it appear spherical. A perception of depth that we develop from children. Young babies perceive a sphere in the outside world (beyond ones head) as just a circle with a gradient. Unlike our eyes that allow stereoscopic 2D images to appear 3 dimensional [1], a gradient on a 2D surface (the canvas) with the right background causes something that is just a circle to appear 3 dimensional.

To prove this, try looking at the test, first with one eye covered and then swap and then cover your other eye. Apart from causing eyes strain, the affect is more 2D.


So in the worlds of both geometry and computer graphics, that's a simple circular object. You'd need to change the gradient and apply a transform to the text that's supposed to be on the object to make it a sphere. :)


Yes, the non gradient text partially destroys the optical illusion.


Alan, I'm curious - your "CSS Test" link goes to a page that lists Firefox above safari on your "Honor Roll" (presumably for CSS compliance). Why is that? Firefox is my preferred browser, but my understanding is that WebKit (Safari and Chrome) has had better standards compliance than Firefox by pretty much any measure for quite awhile now.


That was last revised on the 21st of Oct 2009 (so outdated). I consider layout, formatting, selectors and accessibility among other things.


1. <http://en.wikipedia.org/wiki/Stereoscopic_vision>


--
Alan http://css-class.com/

Armies Cannot Stop An Idea Whose Time Has Come. - Victor Hugo
______________________________________________________________________
css-discuss [cs...@lists.css-discuss.org]
http://www.css-discuss.org/mailman/listinfo/css-d
List wiki/FAQ -- http://css-discuss.incutio.com/
List policies -- http://css-discuss.org/policies.html
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/

Reply via email to