All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=750672

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=64c7b573035e1ba1914809bd8a4bfd21771951fc30e95069ae08e95e10db1762


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

chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972545674532757712

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Justin Schuh
  Commit : 4a572da99ce704e632221bd05e6d894d49a73dcc
  Date   : Mon Jul 24 14:37:24 2017
  Subject: Replace Saturated*() calls with Clamp*() templates

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : speedometer2
  Metric       : Inferno-TodoMVC/Speedometer2
  Change       : 8.44% | 980.57325 -> 1063.37275

Revision             Result                  N
chromium@488938      980.573 +- 8.95047      6      good
chromium@488952      974.61 +- 15.9323       6      good
chromium@488959      970.988 +- 10.6605      6      good
chromium@488963      971.261 +- 13.2336      6      good
chromium@488965      977.522 +- 4.49072      6      good
chromium@488966      1057.45 +- 16.8918      6      bad       <--
chromium@488994      1063.37 +- 27.2524      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=Speedometer2 speedometer2

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8972545674532757712


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972477518104831552

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Hans Wennborg
  Commit : d2c91228a51bdf37ae3b2e501fb53c0528f1629c
  Date   : Fri Jul 28 20:11:05 2017
  Subject: win: Set is_clang=true by default

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : speedometer2
  Metric       : Inferno-TodoMVC/Speedometer2
  Change       : 14.96% | 1069.26183333 -> 909.255083333

Revision             Result                  N
chromium@490485      1069.26 +- 5.91835      6      good
chromium@490490      1066.8 +- 9.82967       6      good
chromium@490492      1066.1 +- 9.64611       6      good
chromium@490493      1074.66 +- 9.41858      6      good
chromium@490494      914.524 +- 9.28185      6      bad       <--
chromium@490502      905.234 +- 4.14881      6      bad
chromium@490518      907.695 +- 11.0604      6      bad
chromium@490550      906.294 +- 11.2515      6      bad
chromium@490615      909.255 +- 8.91557      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=Speedometer2 speedometer2

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8972477518104831552


For feedback, file a bug with component Speed>Bisection

Please note that #6 is not a regression but an improvement.

Let's not dupe it, because  Issue 750923  is about a completely different test.

And are you saying this isn't a regression but an improvement?

Re #8 indeed, I was investigating a regression on Win32 (see #4) and noticed a improvement afterwards and started another bisect.

Sounds like we're all good then :-)

 Issue 752089  has been merged into this issue.