Video Trim Case: Why a Two-Second Cut Can Export Three Seconds
Compare actual browser downloads from Fast trim and Accurate MP4 on the same four-second synthetic clip: 3.104341 seconds versus 2.000000 seconds.
Same range, different downloaded duration
Selecting start 1 and end 3 asks for two seconds of video. In our September 6, 2026 browser test, Fast trim downloaded a 3.104341-second file. Accurate MP4 downloaded a 2.000000-second file from the same source and selected range.
This is why a successful download is not enough to verify a cut. Video Trimmer offers stream copying for speed and re-encoding for more accurate boundaries. We kept both outputs so you can inspect the difference yourself.
Source and settings
The source is a synthetic four-second moving test pattern, 320 × 180 pixels at 24 frames per second, with a 440 Hz tone. It was generated with FFmpeg as H.264 video and AAC audio; it is not customer footage or a performance benchmark for phone recordings.
The source and both browser downloads are linked below. Keep the source unchanged between runs.
- Open Video Trimmer and select the source MP4.
- Set Start time to 1 and End time to 3.
- Choose Fast trim, run Trim video, review the preview and download.
- Repeat from the original source with Accurate MP4. Keep the default re-encoding settings.
- Play both exports and inspect their durations, rather than relying only on the requested range.
Measured outputs
| Setting | Fast trim | Accurate MP4 |
|---|---|---|
| Requested range | 1–3 seconds | 1–3 seconds |
| Requested duration | 2 seconds | 2 seconds |
| Download duration | 3.104341 seconds | 2.000000 seconds |
| Download size | 140,081 bytes | 71,727 bytes |
| Video | H.264, 320 × 180 | H.264, 320 × 180 |
| Audio | AAC | AAC |
Both files were produced through automated Chrome sessions using the site's file selector, controls and download buttons. We then decoded the complete downloaded files independently with FFmpeg and read their container durations with ffprobe. Fast trim was exercised on the deployed site; Accurate MP4 was exercised on the release candidate with the same trimming implementation. This is a specific fixture result, not a claim that every cut has this timing difference.
Why stream copy can include extra time
Fast trim avoids decoding and re-encoding the entire selected video. Compressed frames can depend on earlier frames, and a usable start may be tied to a nearby keyframe. Packet timestamps and audio boundaries can also affect the reported container duration. A requested start time therefore does not always become an exact new first frame.
Accurate MP4 decodes and re-encodes the selected range. That costs more CPU and memory but can make the requested boundaries much closer. Here it produced exactly two seconds at the container level. For frame-critical editing, still inspect the first and last visible frames in the receiving editor; a duration alone does not prove that the right content was selected.
Which mode to choose
Fast trim is useful for a rough draft when retaining a little extra context is acceptable. Accurate MP4 is the better starting point when a title card, spoken sentence, or publishing duration limit depends on the cut boundary. Keep mobile ranges short because re-encoding can exhaust browser memory.
Do not infer that Accurate MP4 always makes smaller files from this table. This source is simple and its re-encoding settings differ from the original encoding. Another clip can become larger. After trimming, use Video Info Inspector and, if needed, Video Compressor to check the actual file you plan to publish.
The media publishing workflow keeps video steps separate from image preparation and explains which downloaded file should enter the next step. Keep the original until playback, sound and the final destination have been checked.
Download the case files
Synthetic samples and actual browser exports from September 6, 2026. Keep the original when comparing results.