All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=623976

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg4uWvvAoM


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

chromium-rel-win7-x64-dual

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Trying again with another bisect.

The bisect has failed again :(. However this seems to be a regression that follows a previous improvement.

In particular, the improvement happens in range 401756:401809 and regression occurs in 402287:402324. This exactly coincides with landing (401796) and reversion (402294) of "Turn on enable begin frame scheduling by default". 

The patch modifies begin main frame logic which is relevant to this metric that measures FPS. /cc enne@ who authored the patch.

I suggest closing this bug and not sinking more time on it.

What does -20ms mean on this graph?

alancutter@ is the owner of the test. I do not know what this test is doing. We are just using these benchmarks to measure the Chrome's performance.

The metric has removed. I think we can close this.

Erm I meant to say: The metric has recovered.