[Webkit-unassigned] [Bug 211295] Video buffering and degraded playback performance

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Sun May 17 12:50:15 PDT 2020


https://bugs.webkit.org/show_bug.cgi?id=211295

--- Comment #7 from Dustin Kerstein <dustin.kerstein at gmail.com> ---
I just did a little more testing comparing setting the uhd60fps.mp4 video (164MB with 100% I-Frames) directly as URL src vs. using a blob, and it appears there may be more than one issue at play here. 

With the following code:

      var src = "https://files.panomoments.com/uhd60fps.mp4"
      var video = document.createElement("video");
      document.body.appendChild(video);
      video.controls = true;
      video.autoplay = true;
      video.src = src;

I see a total of 273MB of network requests after the video finished (sluggishly) playing. See the Src.png attachment.

And with this code:

      var src = "https://files.panomoments.com/uhd60fps.mp4"
      var video = document.createElement("video");
      document.body.appendChild(video);
      video.controls = true;
      video.autoplay = true;
      fetch(src).then(function(response) {
        response.blob().then(function(myBlob) {
          video.src = window.URL.createObjectURL(myBlob);
        });
      });

I see a total of 36.1GB of network requests after the video has finished (very sluggishly) playing. See the Blob.png attachment.

In summary, I think I'm seeing these four behaviors in Safari when testing using the uhd60fps.mp4 video linked above: 

1. Play from file:///Users/XYZ/Downloads/uhd60fps.mp4 (ie. dragging the video into the URL bar) - Playback is perfect at 60fps
2. Play via setting src directly to a URL - Playback is degraded (possibly the XPC memory issue)
3. Play with src as blob - Playback is severely degraded (possibly XPC memory issue and additionally the drastically increased 206 network requests)
4. When setting playbackRate < 1 and setting src directly to a URL - Buffering and Playback stall (this feels a bit unrelated compared to 1-3)

Let me know if there's anything else I can get. Are you able to replicate the behaviors above?

-- 
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/20200517/e29a122a/attachment.htm>


More information about the webkit-unassigned mailing list