[Webkit-unassigned] [Bug 211852] New: Media Source Extensions + WebGL video texture degraded playback performance

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=211852

            Bug ID: 211852
           Summary: Media Source Extensions + WebGL video texture degraded
                    playback performance
           Product: WebKit
           Version: Safari 13
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Major
          Priority: P2
         Component: Media
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: dustin.kerstein at gmail.com

It appears that the issue noted in https://bugs.webkit.org/show_bug.cgi?id=211750 is significantly worse when using Media Source Extensions for video playback. Here are two near-identical demos:

   JSFiddle with MSE - https://jsfiddle.net/yj9vpfkr
   JSFiddle with blob as video source - https://jsfiddle.net/knL126st

*Note* that as these JSFiddles use iFrames, you'll need to click into the result iFrame to get Safari to avoid using its low power mode (meant for ads) which limits rendering to 30fps.

Test results - JSFiddle with MSE:

   2011 Macbook Pro - Playback starts to freeze at playbackRate = 0.02 and also causes the render loop to stop running (and the whole tab freezes)
   First generation iPad Pro - Playback starts to freeze at playbackRate = 0.10

Test results - JSFiddle with blob as video source:

   2011 Macbook Pro - Playback starts to freeze at playbackRate = 0.80
   First generation iPad Pro - Playback starts to freeze at playbackRate = 2.00

Note that with blob as video source you'll eventually run into https://bugs.webkit.org/show_bug.cgi?id=211750, but regardless, the maximum playbackRate with MSE is significantly worse.

Also note that these JSFiddles are specifically using Long GOP test video files to avoid this XPC messages issue - https://bugs.webkit.org/show_bug.cgi?id=211295

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20200513/2b2e93ec/attachment-0001.htm>