⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Changeset 277406 in webkit


Ignore:
Timestamp:
May 12, 2021, 4:18:49 PM (5 years ago)
Author:
mmaxfield@apple.com
Message:

[MotionMark] Temporarily remove the focus test
https://bugs.webkit.org/show_bug.cgi?id=225714

Reviewed by Geoffrey Garen.

The focus test is causing unacceptably high variance on test results. Initial investigation
indicates that this is related to secondary and tertiary interactions between the browser's
runloop and the compositor's runloop. Not only that, but the variable cost of the focus test
seems to affect the variance on the subtests which run after it.

This patch temporarily removes the focus test, just so we can get a stronger signal of
browsers' true performance, while we do a deeper analysis of the focus test.

Because the focus test's score is lower than the average score of the subtests, removing it
causes scores in all browsers to increase:

| WebKit | Chrome | Firefox

================================================
Score with Focus | 1020 | 550 | 962
Score without Focus | 1623 | 874 | 1101
Stddev with Focus | 9.3% | 2.3% | 2.9%
Stddev without Focus | 0.7% | 0.5% | 7.9%

  • MotionMark/resources/runner/tests.js:
Location:
trunk/PerformanceTests
Files:
2 edited

Legend:

Unmodified
Added
Removed
  • trunk/PerformanceTests/ChangeLog

    r276247 r277406  
     12021-05-12  Myles C. Maxfield  <mmaxfield@apple.com>
     2
     3        [MotionMark] Temporarily remove the focus test
     4        https://bugs.webkit.org/show_bug.cgi?id=225714
     5
     6        Reviewed by Geoffrey Garen.
     7
     8        The focus test is causing unacceptably high variance on test results. Initial investigation
     9        indicates that this is related to secondary and tertiary interactions between the browser's
     10        runloop and the compositor's runloop. Not only that, but the variable cost of the focus test
     11        seems to affect the variance on the subtests which run after it.
     12
     13        This patch temporarily removes the focus test, just so we can get a stronger signal of
     14        browsers' true performance, while we do a deeper analysis of the focus test.
     15
     16        Because the focus test's score is lower than the average score of the subtests, removing it
     17        causes scores in all browsers to increase:
     18
     19                             | WebKit | Chrome | Firefox
     20        ================================================
     21        Score with Focus     |   1020 |    550 |    962
     22        Score without Focus  |   1623 |    874 |   1101
     23        Stddev with Focus    |   9.3% |   2.3% |   2.9%
     24        Stddev without Focus |   0.7% |   0.5% |   7.9%
     25
     26        * MotionMark/resources/runner/tests.js:
     27
    1282021-04-19  Kimmo Kinnunen  <kkinnunen@apple.com>
    229
  • trunk/PerformanceTests/MotionMark/resources/runner/tests.js

    r232653 r277406  
    5353        },
    5454        {
    55             url: "master/focus.html",
    56             name: "Focus"
    57         },
    58         {
    5955            url: "master/image-data.html",
    6056            name: "Images"
Note: See TracChangeset for help on using the changeset viewer.