New issue
Advanced search Search tips

Issue 849699 link

Starred by 3 users

Issue metadata

Status: WontFix
Owner:
Closed: Jun 2018
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

1.2%-21% regression in webrtc_perf_tests at 23511:23511

Project Member Reported by asapersson@chromium.org, Jun 5 2018

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=849699

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=948f0b2fc677d0c0b1313524942bd8dda695a91910532cb06e86788773df2891


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

webrtc-linux-large-tests
webrtc-win-large-tests
Owner: sprang@chromium.org
Status: WontFix (was: Untriaged)
Most of these are improvements. We're using an extra thread in the encoder now, which of course has some effect.
CPU usage up, but encode time is down, which makes sense.
Interestingly BWE has increased and send time has decreased in many tests.

There's only one test where encode frame rate has actually gone down. The qp has actually gone down, so maybe this causes some overshooting leading to dropped frames. Change is small enough that I wouldn't worry.

Overall I'll say this is working as intended.

Comment 4 by sprang@chromium.org, Jun 15 2018

Cc: brandtr@chromium.org sprang@chromium.org
 Issue 853129  has been merged into this issue.

Sign in to add a comment