All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=702347

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgtNXFggsM


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

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984935879109477152

Ref graph last updated at revision 424054. 

https://chromeperf.appspot.com/report?sid=984fd6d176419670e17174d297e7366523ac4e98c9488418bb9f38fc6b707ac9

Caleb, seems like the build may have been broken around that time - can I bump the ref graph to a new revision to see if it will start working?

The ref build has been broken because the ref build is broken: before sandersd@' fix to prevent us from timing out in the middle of a seek, the bots would time out in the middle of seeks and the tests would break.

Not sure if there's anything we can do... I thought we were just waiting until the reference builds magically start working again once the fix gets streamed up to them as well.

Bisect should still work.

 issue 695960 

The reference build is at M58. Not sure when sandersd@'s change went in.

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : vm_proportional_set_size_delta/tulip2.m4a_browser

Revision             Result                  N
chromium@456337      2137.71 +- 2825.32      21      good
chromium@456352      1947.0 +- 2436.35       21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984935879109477152

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5251432061599744


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984581578714195408

New bisect with wider range.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984485734719209024

Another bisect with wider range - if this one doesn't find I will assume this one is noise (no ref graph to argue either way, only data point we have is failing bisects). 

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : vm_proportional_set_size_delta/tulip2.m4a_browser

Revision             Result                  N
chromium@456200      2168.57 +- 2782.52      21      good
chromium@456400      2016.86 +- 2569.69      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984581578714195408

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5897578668883968


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

Closing per plan in in comment #11.

Makes sense. The problem here is the ridiculously high standard deviations. If we could find a way to make these tests less flaky, it would make these bisects find relevant information. +johnchen@ for reference.

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : vm_proportional_set_size_delta/tulip2.m4a_browser

Revision             Result                  N
chromium@456300      2025.14 +- 2704.93      21      good
chromium@456400      1945.86 +- 2155.72      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984485734719209024

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6430473246474240


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982632482332894112