If your goal is to make an action take longer or shorter, change its timestamps with setpts. If your goal is a higher output frame rate with newly generated in-between frames, assess framerate. These solve different problems: a higher rate alone does not mean interpolation, and changing timestamps does not create new motion samples.
Choose the operation and check its consequences
- State the desired result in seconds and frames per second separately. For example, “make this ten-second action last twenty seconds” calls for a duration change; “target 50 fps with interpolated frames, while aiming to keep roughly ten seconds” calls for a frame-generation decision. Treat that duration as a delivery constraint to verify, not as a guarantee of
framerate. - Confirm the source’s frame rate and whether it is interlaced. FFmpeg documents
setptsas changing frame presentation timestamps; its examples usesetpts=2.0*PTSfor slow motion andsetpts=0.5*PTSfor fast motion. The exact output duration can depend on stream timing and the surrounding output settings, so verify the rendered file rather than relying on the expression alone. See FFmpeg’s `setpts` documentation. - If the intended change is output frame generation, consider
framerate, which FFmpeg describes as interpolating new frames from source frames. Set an explicit target rate, such asframerate=50, and inspect actual frames. Do not substitute thefpsfilter when your requirement is interpolation: FFmpeg documentsfpsas producing constant-rate output by duplicating or dropping frames. See FFmpeg’s `framerate` documentation and its `fps` documentation. - Review representative motion at normal playback and frame by frame. Check fast gestures, limbs crossing or becoming occluded, and every shot boundary. Decide whether the timing change matches the intended rhythm and whether generated images remain plausible. This comparison is an editorial review method, not a guarantee that interpolation can reconstruct missing captured detail.
Hypothetical acceptance worksheet
Assume a progressive 25 fps source lasting 10 seconds. The figures below are planning expectations and review prompts, not measured FFmpeg output or guaranteed frame counts.
| Candidate | Expected timing or output | Observation to record | Decision |
|---|---|---|---|
setpts=2.0*PTS | Timestamp intervals expand; slow-motion candidate is about 20 seconds | Does the action feel twice as long, and does its order remain coherent? | Keep only if the longer rhythm is the goal |
framerate=50 | Target output rate is 50 fps; intermediate frames are interpolated from source frames. Measure the rendered duration; the target rate alone does not establish it. | In a fast hand wave, do edges or finger positions look implausible between source frames? Does the measured duration meet the delivery target? | Keep only if the new frames hold up at normal playback and inspection, and the measured timing meets the brief; otherwise handle timing separately. |
framerate=50 across a cut | The filter may create output frames around the transition | Do any generated frames blend or distort content across the shot change? | Inspect the cut closely; do not assume the rate alone ensures a clean transition |
For an additional obstruction check, note whether one moving subject passes in front of another: reject a candidate if generated boundaries or motion become visibly incoherent. The official documentation describes interpolation and scene-change detection, but does not guarantee artifact-free results for these particular shots.
Limits and what to try next
FFmpeg says framerate is not designed to work correctly on interlaced media. If the source is interlaced, do not apply the progressive-material workflow directly: the documented requirement is to deinterlace before this filter and re-interlace afterward when needed. Confirm the media type before choosing a filter.
If interpolated motion fails review, try a different representative segment or use timestamp retiming when the actual goal is a duration change. If retiming produces the wrong duration, inspect the rendered stream’s timing and output configuration; do not treat that as evidence that interpolation is needed. If neither result meets the brief, retain the original cadence or revise the delivery target rather than claiming synthetic frames restore unrecorded action.
Common questions
Does `setpts` make extra frames? No. It changes presentation timestamps; it is the option to assess when the desired result is a different playback duration.
Does `framerate` change the action’s duration? Its documented purpose is to change output frame rate by interpolating frames. Check the actual rendered duration and timing instead of assuming a particular duration from the target rate alone.
Is `fps=50` equivalent to `framerate=50`? No. FFmpeg describes `fps` as duplicating or dropping frames to reach a constant rate, whereas `framerate` generates interpolated output frames.
Sources and next step
- 20.19 setpts, asetpts — FFmpeg official documentation
- 11.101 framerate — FFmpeg official documentation
- 11.99 fps — FFmpeg official documentation
This guide interprets public documentation and uses explicitly hypothetical acceptance examples, not application tests. Controls and availability can change; verify the cited manual against your installed version.
For work created in Imagild, carry the input requirements and acceptance checklist into the film workspace. The external software controls discussed above belong to their respective products.