Splitting a video at specific timestamps is one of the most common editing needs. A teacher needs to divide a 60-minute lecture into four 15-minute segments for an LMS that has a file size limit. A podcaster needs to separate a multi-guest episode into individual interview segments. A videographer needs to extract a specific 45-second window from a 2-hour event recording.
All of these tasks require the same operation: marking time positions in the video and creating separate files from each segment.
Timestamp formats
Most video tools accept timestamps in one of these formats:
- HH:MM:SS (hours:minutes:seconds): "01:23:45" means 1 hour, 23 minutes, 45 seconds
- MM:SS (minutes:seconds): "05:30" means 5 minutes, 30 seconds
- SS.mmm (seconds with milliseconds): "330.500" means 330.5 seconds
- HH:MM:SS.mmm: "00:05:30.500" means 5 minutes, 30.5 seconds
For most splitting tasks, second-level precision is sufficient. For frame-accurate work (syncing to music beats or extracting precise moments), millisecond precision matters. At 30fps, each frame occupies 33.3 milliseconds, so millisecond-level timestamps give you frame-level control.
Single split vs. multiple splits
A single split divides the video into two parts: everything before the split point and everything after. You specify one timestamp and get two output files.
Multiple splits divide the video into three or more parts. You specify multiple timestamps and get one output file per segment. For example, splitting a 15-minute video at 5:00 and 10:00 produces three 5-minute segments.
Some tools require you to specify split points (where to cut). Others ask for segment durations (how long each piece should be). The result is the same, but the input method differs.
Splitting without re-encoding vs. with re-encoding
Splitting without re-encoding (stream copy) is fast because it copies the compressed data directly without processing it. The limitation is keyframe alignment: splits can only happen cleanly at keyframe boundaries, which may be up to several seconds apart. The beginning of each segment after the first may include a few extra frames before the intended start point.
Splitting with re-encoding is slower but allows frame-accurate split points. Each segment is decoded and re-encoded with its own keyframe structure. This guarantees that every segment starts and ends exactly where you specified, at the cost of processing time and one generation of compression.
For most splitting tasks (dividing a lecture into parts, separating interview segments), keyframe-level accuracy is fine. For precision work (music video editing, frame-accurate content extraction), re-encoding is necessary.
File naming conventions
When splitting a video into multiple parts, clear file naming prevents confusion. A good convention is:
- original-filename-part1.mp4
- original-filename-part2.mp4
- original-filename-part3.mp4
Or timestamp-based naming: - original-filename-00m00s-05m00s.mp4 - original-filename-05m00s-10m00s.mp4
This makes it obvious which part contains which portion of the original video. The MediaBrew split tool uses sequential numbering (part1, part2, etc.) for the output files.