All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=599580

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgiLCTpAoM


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

chromium-webrtc-trunk-tot-rel-mac

The original graph is noisy, so I'm not exactly certain about the CL range, but here's a list of the changes in the originally reported range:
http://test-results.appspot.com/revision_range?start=383461&end=383742

As seems to be the common theme today, nothing obvious stands out. There aren't any WebRTC rolls or anything ostensibly related to h.264 in that range. I also associated an increase in chromium-webrtc-trunk-tot-rel-mac/browser_tests / Unique_frames_count / 720p_H264 to this bug since it appears to have happened around the same CL range. The configuration for that test may need to be modified to indicate that an incrase in unique_frames_count should be considered an improvement.

hbos@ - can you take a look?

That Mac machine was upgraded from 10.9 to 10.10 at March 29, which fits with the regression. I'll leave it to hbos@ to close the bug though (in case there may be something else).