Electrolysis/Release Criteria/UI Smoothness: Difference between revisions

← Older edit

Electrolysis/Release Criteria/UI Smoothness (view source)

Revision as of 15:29, 21 April 2016

851 bytes added , 21 April 2016

→‎talos CART (customization mode animation test): CART stuff

Mconley

Confirmed users, Bureaucrats and Sysops emeriti

998

edits

@@ Line 14: / Line 14: @@
 * e10s comparison validated: mconley
-* Current e10s diff: Possibly worse by ~6% on OS X with APZ disabled (see [https://bugzilla.mozilla.org/show_bug.cgi?id=1251699#c2 this analysis])
+* '''signed off by barbara over email on April 20th, 2016'''
-* {{bug|1251699}} - Investigate potential e10s TART regression on OS X
+* Current e10s diff: '''No worse than non-e10s with APZ disabled''' (see [https://bugzilla.mozilla.org/show_bug.cgi?id=1251699#c5 this analysis])
+* {{bug|1251699}} - 5.7% TART regression on OS X with APZ disabled
+** Closed as WORKSFORME - this regression seems to have disappeared.
+* {{bug|1253412}} - APZ causes 4% - 12% TART regression - Event Region Regression
+** This has been WONTFIX'd by the APZ team. See [https://bugzilla.mozilla.org/show_bug.cgi?id=1253412#c10 this comment] and [https://bugzilla.mozilla.org/show_bug.cgi?id=1253412#c19 this comment].
 == talos tpaint (first paint of a new window opened from content) ==
 * e10s comparison validated: jimm
-* Current e10s diff: worse by 30-40% on all platforms, with APZ on or off.
+* Current e10s diff: '''worse by 3-8% on Linux 64, Windows 7 and Windows 8 platforms, with APZ on or off. 11% improvement on OS X'''
-* {{bug|1174770}} - tpaint regressions (3%-33%) in e10s mode compared to non-e10s mode
+* {{bug|1174770}} - tpaint regressions (3%-8%) in e10s mode compared to non-e10s mode
 === Notes ===
@@ Line 37: / Line 41: @@
 * We suspect there is a second slow-path here that Talos is not measuring, and that's when the user clicks on a link that opens a new tab.
 ** A bug has been filed to add such a test ({{bug|1253382}})
+** Preliminary local results from that test show that '''e10s does not seem to regress this case'''
 * A bug has been filed to get Telemetry on how often we hit the ContentParent::RecvCreateWindow codepath, which we believe is the slow path. ({{bug|1253388}})
@@ Line 43: / Line 48: @@
 * e10s comparison validated: jimm
 * Current e10s diff: mixed
-** '''Windows 8 64: better by 12%'''
+** Linux 64: ~7% regression
-** '''Windows 7/XP: worse by 7-12%'''
+** OS X: ~2% regression
-** OS X: worse by 3%
+** Windows 7: ~2% regression
-** Linux: worse by 1%
+** Windows 8 64: ~17% improvement
+** Windows XP: ~9% regression
+* [https://bugzilla.mozilla.org/show_bug.cgi?id=1252947#c23 Signed off by canuckistani here.]
 * {{bug|1179732}} - 11-147% linux*/winxp tresize regression when comparing e10s to non e10s on the same build
+* {{bug|1252947}} - ~5% winxp tresize regression when comparing e10s to non e10s on the same build
 == talos tps (tab-switch speed) ==
 * e10s comparison validated: jimm
-* Current e10s diff (with APZ disabled - see [https://bugzilla.mozilla.org/show_bug.cgi?id=1186585#c22 this analysis])
+* Current e10s diff:
-** '''Win 8:  19.66% worse'''
+** '''Linux 64: 19.38% improvement'''
-** '''Win 7:  23.30% worse'''
+** '''OS X:  35.59% improvement'''
-** '''OS X:  44.37% worse'''
+** '''Win 7:  26.61% improvement'''
-** '''Linux 64:  44.84% worse'''
+** '''Win 8:  33.78% improvement'''
-* {{bug|1186585}} - 28.2% to 48.5% tps talos regression with e10s enabled
+** '''Win XP:  26.26% improvement'''
 == talos CART (customization mode animation test) ==
@@ Line 67: / Line 75: @@
 ** '''Linux 64:  6.8% worse'''
 * {{bug|1252768}} - 2–7% e10s CART regression
+** The regressions on non-Linux platforms is due to the "New non-e10s window" button that is added to Nightly and Aurora builds
+** The regression for Linux 64 is not yet understood.
 = Bugs =