New issue
Advanced search Search tips

Issue 605877 link

Starred by 0 users
Status: WontFix
Owner:
sprang@chromium.org
Closed: May 2016
Cc:
pbos@chromium.org
nyerramilli@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

5.1% regression in webrtc_perf_tests at 12446:12447

Project Member Reported by hlundin@chromium.org, Apr 22 2016

Comment 1 by hlundin@chromium.org, Apr 22 2016 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=605877

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgxPfiuQoM


Bot(s) for this bug's original alert(s):

webrtc-mac-large-tests

Comment 2 by hlundin@chromium.org, Apr 22 2016

Owner: pbos@chromium.org
Possibly related to https://bugs.chromium.org/p/chromium/issues/detail?id=605867.

pbos@ PTAL.

Comment 3 by pbos@chromium.org, Apr 22 2016

Cc: pbos@chromium.org
Owner: sprang@chromium.org
Not really sure what to do with this. sprang@: we're not sending scaled screenshare video in tip of tree, are we?

Otherwise I think this is due to the asynchronous video encoders, but I don't see how that should affect the bitrate like this? Maybe we start sending a bit faster/differently so we get more frames?

Comment 4 by nyerramilli@chromium.org, May 2 2016

Cc: nyerramilli@chromium.org
gentle ping..

Comment 5 by sprang@chromium.org, May 3 2016

Status: WontFix (was: Assigned)
This very much looks to me like similar performance "regressions" in the past, where the timing of call start was offset slightly relative to the timed thread polling stats. 

Sending slightly more data in vp9 svc mode does not seem scary to me.
I'll close this as won't fix, please reopen if you disagree.
►

Sign in to add a comment