Slow the review down to observable moments
Use a local clip to inspect the setup, contact point and follow-through of a movement. Pause near the action, choose a time-step size and compare nearby pictures. Capture the moments you want to discuss, then download them for notes or a side-by-side layout in an editor. Choose recordings where the subject is visible and the camera position is useful; an obstructed view cannot become a reliable comparison just by sampling more stills.
Timestamp steps are not certified frame indexing
The 1/24, 1/30 and 1/60-second controls request small time changes. A variable-frame-rate recording or a browser decoder may not deliver an exact adjacent source frame for every request. Match the step to the approximate recording rate, inspect what changes and keep the timestamps as navigation references. This tool is not a biomechanics measurement system: it does not calculate joint angles, speed, force or injury risk from a two-dimensional picture.
Make comparisons that respect the recording
Changing camera angle, zoom and perspective can make the same movement look different. When comparing sessions, use similar camera placement and choose comparable phases rather than simply matching elapsed seconds. Motion blur and rolling-shutter distortion can also affect apparent positions. Export a few clear PNG images, annotate observations separately and retain the video context. Avoid treating one captured instant as proof of a technical or medical conclusion.
Questions for this workflow
Can this measure an athlete’s speed or joint angles?
No. It supports visual review and still capture, without calibrated movement measurements.
Does 1/30-second stepping always mean one frame?
No. Frame rate, timing and browser seeking can differ. Use a dedicated analyzer when exact source-frame identity is essential.
Continue your workflow
Read practical guides · Enlarge an image · Report a reproducible problem
Capture uses the shared frame-extractor limits: 2 GiB source, 8,294,400 decoded pixels, 120 pictures and 96 MiB retained images. Start with a short playable sample; codec support and device memory vary.
Make a repeatable comparison note
For a swing or jump, enter three requested timestamps around setup, contact and follow-through. Save the original filename, camera position, approximate times and what is visibly different. Use a timestamped contact sheet to keep those pictures together. Compare another session only when view and phase are comparable; this is visual review, not an automatic measurement.