All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=648625

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-dP6vAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICghfiTtwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-bmqoAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-ZzHpwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg-fXgsgoM


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

webrtc-android-tests-nexus72
webrtc-android-tests-nexus9
webrtc-win-large-tests

Seems unlikely. The my cl does touch functions computing the psnr (but with no intended changes to how it's computed), but it shouldn't affect fps or dropped frames. 

But it looks like something did change in the graphs. Any other candidate cls? The "Chromium Git Hash range:" links appear dead.

Closing, I see no reasonable action to take.