152221 2015-12-12 21:40:54 -0800 Web Inspector: Don't render console messages until Console is opened 2016-05-02 11:50:32 -0700 1 1 1 Unclassified WebKit Web Inspector WebKit Local Build All All RESOLVED DUPLICATE 155629 InRadar P2 Normal --- 152220 1 nvasilyev nvasilyev bburg giovanni.piller graouts joepeck mattbaker nvasilyev timothy webkit-bug-importer oldest_to_newest 1148780 0 nvasilyev 2015-12-12 21:40:54 -0800 Running the following code make Web Inspector unbearably slow in a couple of minutes: setInterval(function() { console.log({x: Math.random()}) }, 200); The low hanging fruit would be to simply not update Console's DOM when it isn't visible, e.d. the split console is closed and the active tab isn't Console. 1148783 1 nvasilyev 2015-12-12 22:14:35 -0800 When the console isn't visible, the console content view is already not present in the DOM. However, https://github.com/WebKit/webkit/blob/master/Source/WebInspectorUI/UserInterface/Controllers/JavaScriptLogViewController.js#L276 WebInspector.JavaScriptLogViewController#_appendConsoleMessageView keeps appending elements regardless of that. When console isn't visible, it appends to a DOM element that is detached from the DOM: >> this._currentConsoleGroup.element.parentNode.parentNode.parentNode -> null I wonder if modifying detached DOM is still making Web Inspector slow. 1148787 2 nvasilyev 2015-12-12 22:27:27 -0800 After profiling in Timelines I noticed that I get 4 paints (!) on every console.info call when the split console is closed! I intentionally used console.info instead of console.log that doesn't modify console message counts on the dashboard. 1148791 3 nvasilyev 2015-12-12 22:36:02 -0800 I ran the following code in Firefox DevTools and Chrome DevTools for a couple of minutes: setInterval(function() { console.log({x: Math.random()}) }, 1); Both performed well — responsive and snappy. I profiled Chrome DevTools and it seems that they use a custom requestAnimationFrame batching. Firefox DevTools use React.js. 1189352 4 webkit-bug-importer 2016-05-02 02:05:14 -0700 <rdar://problem/26037238> 1189378 5 timothy 2016-05-02 06:05:58 -0700 Isn't this a dupe? 1189401 6 timothy 2016-05-02 09:10:28 -0700 *** This bug has been marked as a duplicate of bug 155629 *** 1189448 7 nvasilyev 2016-05-02 11:50:32 -0700 (In reply to comment #5) > Isn't this a dupe? Yes, it is.