New issue
Advanced search Search tips

Issue 668015 link

Starred by 2 users

Issue metadata

Status: WontFix
Owner:
Closed: Nov 2016
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 1
Type: Bug-Regression



Sign in to add a comment

Opacity rendering when child has animation with transform

Reported by david.an...@genero.com, Nov 23 2016

Issue description

Chrome Version       : 54.0.2840.98 (64-bit)
URLs (if applicable) : https://jsfiddle.net/j1w40Lx0/3/
Other browsers tested:
  Add OK or FAIL, along with the version, after other browsers where you
have tested this issue:
     Safari: OK Version 9.1.2 (11601.7.7)
    Firefox: OK 49.0.2

What steps will reproduce the problem?
(1) Go to the provided URL (https://jsfiddle.net/j1w40Lx0/3/)

What is the expected result?
There are two boxes, call them box 1 and 2, each having 1 child box. Both boxes have opacity set to 0.5. The only difference is box 2's child box is being animated to rotate by using `transform` property. Both box 1 and 2's opacity should look identical.

What happens instead?
Box 1 and 2 appears different. In fact, box 2 (the one with rotated child box) is fainter than box 1 as long as its child box is animating. Once animation stops both boxes look identical as they should. This bug only happens when opacity is set using CSS `opacity` property. The bug doesn't happen when opacity is set using the `filter` property.

Screenshot attached.



 
Screen Shot 2016-11-23 at 4.44.08 PM.png
23.9 KB View Download

Comment 1 by ajha@chromium.org, Nov 23 2016

Labels: M-54 Needs-Bisect
Components: Blink>Animation
Labels: -Type-Bug -Pri-3 -Needs-Bisect hasbisect OS-Mac Pri-1 Type-Bug-Regression
Owner: erikc...@chromium.org
Status: Assigned (was: Unconfirmed)
Able to reproduce the issue on  Mac 10.11.6 and using reported version #54.0.2840.98 but the same is not reproducible in the latest canary #57.0.2929.4.

Reverse Bisect Information:
=====================
Good build: 55.0.2883.37 
Bad Build : 55.0.2883.35 

Change Log URL:
--------------- 
https://chromium.googlesource.com/chromium/src/+log/55.0.2883.35..55.0.2883.37?pretty=fuller&n=10000

Possible suspect:
------------------
8bebab8c7bad3e5c3e76ed8b83963ccd561dd522
	
Review-Url:https://codereview.chromium.org/2469893003 

Note: Issue not seen in Windows and Linux

erikchen@ - Could you please check and merge the fix to M54 if it is a valid canadidate.

Thanks...!!


Status: WontFix (was: Assigned)
I don't think we should respin M-54 for this issue. This will be fixed with M-55.

Sign in to add a comment