All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=912291

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e8f7a295445897b52125cec5ba85562ad070dfa95096c441d575ac3dcc3d132d


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

linux-perf

media.desktop - Benchmark documentation link:
  None

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/164f16dc140000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/164f16dc140000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/13b8307a140000

^^ trying again with a wider range

From the graph this is a pretty clear regression. 

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/13b8307a140000

Graphs from repeated runs show a range of values that is as wide as the regression. In perf dashboard the line has been super consistent for a while, but a bit more volatile since this spike. Maybe just noise, but not enough data to call yet - lets wait a few days and see if the graph bounces down again. 

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/1191c9d8140000

After brief bounce down, graph has persisted up (in "bad" territory). Running once more with "bad" range after the bounce.

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/1191c9d8140000