(In reply to Markus Stange [:mstange] from comment #0) > https://speedometer-preview.netlify.app/?suite=Perf-Dashboard [...] > If we can match Chrome's time on these two buckets, our Perf-Dashboard score should improve by 6.8%. Is this still an issue? Running the ["Perf-Dashboard" in current Nightly with 60 iterations](https://speedometer-preview.netlify.app/?suites=Perf-Dashboard&iterationCount=60) to reduce noise, I get consistently better (higher) scores in Firefox Nightly vs. Chrome. On Linux ThinkStation: Firefox 126.0a1 (2024-03-25): 18.2 ± 0.33 Chrome 123.0.6312.58 (Official Build): 17 ± 0.32 If I click Details and then expand the table to look specifically at the "SelectingRange-sync" subtest referenced in comment 4, I see we're faster than Chrome on that specific subtest, too. Firefox: SelectingRange/Sync 21.77 ± 1.19 (5.5%) ms Chrome: SelectingRange/Sync 22.92 ± 0.33 (1.5%) ms Maybe this has been fixed by other speedometer work that we've landed, which got us better batching of invalidations here, or something along those lines?
Bug 1850834 Comment 7 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
(In reply to Markus Stange [:mstange] from comment #0) > https://speedometer-preview.netlify.app/?suite=Perf-Dashboard [...] > If we can match Chrome's time on these two buckets, our Perf-Dashboard score should improve by 6.8%. Is this still an issue? Running the ["Perf-Dashboard" in current Nightly with 60 iterations](https://speedometer-preview.netlify.app/?suites=Perf-Dashboard&iterationCount=60) to reduce noise, I get consistently better (higher) scores in Firefox Nightly vs. Chrome. On Linux ThinkStation: Firefox 126.0a1 (2024-03-25): 18.2 ± 0.33 Chrome 123.0.6312.58 (Official Build): 17 ± 0.32 If I click Details and then expand the table to look specifically at the "SelectingRange-sync" subtest referenced in comment 4, I see we're faster (lower time in `ms`) than Chrome on that specific subtest, too. Firefox: SelectingRange/Sync 21.77 ± 1.19 (5.5%) ms Chrome: SelectingRange/Sync 22.92 ± 0.33 (1.5%) ms Maybe this has been fixed by other speedometer work that we've landed, which got us better batching of invalidations here, or something along those lines?