RE: [css3-images] aliasing and repeating-linear-gradient

Looking at the first example...

 background: -ms-repeating-linear-gradient(left, red 0%, red 0.1%, lime 0.1%, lime 0.2%, blue 0.2%, blue 0.3%);

When the IE PPB2 window is resized to  ~1300 pixels, the rendering is approximately this pattern repeated:
http://lists.w3.org/Archives/Public/www-archive/2011Oct/att-0008/AlanRepeatingPattern1.png

While there is a cyclic bright/dark effect occurring, I don't think this is a non-conformant relative to the current WD.

Alan, do you have any suggestions for spec change(s) to consider that address it?

Given the nature of CRT, LCD, etc. display technology as it stands today I think we'll have these issues to at least some degree, unless we simply disallow sub-pixel color transitions.  I don't think we want to go that far though (we lose a lot of good use cases).

-Brian


-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf Of Alan Gresley
Sent: Wednesday, October 12, 2011 6:05 AM
To: CSS 3 W3C Group
Subject: [css3-images] aliasing and repeating-linear-gradient

Hello all,

Although I am intrigued by what is generated (color wise), I do believe
that it should not be possible to be able to create such aliasing
effects by using very small steps in color stops with
repeating-linear-gradient. Test case.

http://css-class.com/test/css/3/colors/rgb-aliasing-gradient.htm

One example use these color stops:

   repeating-linear-gradient(left, blue 0.0%, blue 0.1%, red 0.1%, red
0.2%, lime 0.2%, lime 0.3%)


Both Gecko and IE10 show rainbow effects when resizing the viewport at
particular widths.

http://css-class.com/test/css/3/colors/rgb-ff7.png



--
Alan Gresley
http://css-3d.org/
http://css-class.com/