Advise by how long users wait for views, as the benchmarks measure it
- github
- github/openchronology/timeline-app
- commit
- main
- github/athanclark
The browser's capacity advice compared the editor's own drawing time with 150 ms. Users wait from their input to the frame that shows the new view, which is longer: the advice showed too late (#69). The editor now times each zoom, pan, pinch or key from the input event until that view is painted, and opening until its first paint, and advises by those. Measuring this showed that zooming drew the moments a frame after the axis and bounds fields moved, and the UI benchmarks stopped timing at the bounds fields. Zoom frames now draw the moments with the axis, and the benchmarks wait for the moments' view, which the editor records as it draws them. A temporary CI job measures navigation by size on hosted runners, to recalibrate the comfortable sizes; it is removed before merging. Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_011fFY8eruEi1xAkkMBni13B https://github.com/openchronology/timeline-app/commit/89a6a86d1a68315e7e3d42370c6baf16e4f08649
Related
Open Advise by how long users wait for views, as the benchmarks measure it in the timeline