Hi, Safari Webkit, Firefox 3b and Opera 9.5b render this test different:
http://www.satzansatz.de/cssd/tmp/opacity/ #stackA { opacity: 0.66; z-index: 2; background: red; width: 300px; height: 300px; } #stackA p { position: relative; z-index: 1; width: 200px; height: 200px; background: yellow; left: -100px; top: 50px; } #stackB { opacity: 0.90; background: blue; margin-top: -150px; width: 150px; height: 300px; } ... <body> <div class="wrapper"> <div id="stackA"> <p><!----></p> </div> <div id="stackB"></div> </div> </body> Screenshots: http://www.satzansatz.de/cssd/tmp/opacity/webkit.png http://www.satzansatz.de/cssd/tmp/opacity/fx.png http://www.satzansatz.de/cssd/tmp/opacity/opera.png The red box and the blue box have opacity set. In addition, the first one gets a z-index. In both Safari and Firefox, opacity forms a stacking context. In Safari, the red box is painted over the blue box. z-index applies. Correct? (Opera seems wrong in letting the blue box crawl between the red box and its child box, since a stacking context should be atomic. In addition, the yellow child element inherits opacity, which appears to be wrong.) I know CSS3 Color Moldule: opacity does not tell about stacking issues, and CSS3 Positioning Module is not written yet (or where is stacking context defined in CSS3 ?). But since these three implementations of opacity exist: should z-index apply here? Another question, probably related: "(2) Elements with opacity < 1.0 establish a stacking context (and a z-index of auto is changed to 0)" Dave Hyatt: http://weblogs.mozillazine.org/hyatt/archives/2004_09.html For the latter part of it, where is that "a z-index of auto is changed to 0" defined? Ingo -- http://www.satzansatz.de/css.html ______________________________________________________________________ css-discuss [EMAIL PROTECTED] 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/
