New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 608399 link

Starred by 3 users

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 3
Type: Bug



Sign in to add a comment

Flickering seen in Canvas on device Kitty

Project Member Reported by srcv@chromium.org, May 2 2016

Issue description

Chrome Version: 51.0.2704.26 / 8172.12.0 dev 
Device: Kitty

What steps will reproduce the problem?
1) Open https://cdn.rawgit.com/uysalere/js-demos/master/intro.html
2) Select any demo video (For example "2d Canvas playing video" https://cdn.rawgit.com/uysalere/js-demos/master/canvascapture.html)

Expected result:
Canvas should show video clearly without any flashing/flickering

Actual result:
Bottom of the canvas is flashing/flickering continuously

Note:
1) This issue is seen with most of the demo videos in https://cdn.rawgit.com/uysalere/js-demos/master/intro.html
2) This issue is seen only on Chrome device Kitty

 

Comment 2 by junov@chromium.org, May 2 2016

Owner: emir...@chromium.org
Status: Assigned (was: Untriaged)
srcv@ can you send a screenshot from this issue? I cannot access the link on #1 for some reason. I am curious if the problem is in the input canvas or output video.

Comment 4 by srcv@chromium.org, May 5 2016

Flickering_in_canvas_bottom_device_kitty.jpeg
1.5 MB View Download
Owner: ----
Status: Available (was: Assigned)
It looks like the problem is in the input canvas (the upper content on the picture) and captured stream has it too as a result. Somehow the original video doesn't get drawn completely onto canvas. It might be a padding issue. 
junov@ can you re-propagate it?

Comment 6 by junov@chromium.org, May 10 2016

Owner: dcasta...@chromium.org
Status: Assigned (was: Available)
@dcastagna: Any idea what may have changed in video to canvas draws that could be causing this?
The biggest change I'm aware of is crrev.com/1869793002. That seemed to have landed after this problem was reported though.

A few related changes have landed on CrOS side a little bit earlier.
Am I right assuming kitty an X86 device?

srcv@, is it possible to bisect to find out when the regression happened?

Cc: -junov@chromium.org

Sign in to add a comment