Split a video into parts
Cut a long recording into pieces short enough to post. The parts are copied, not re-encoded — every one holds exactly the picture and sound that were in the original. Nothing is uploaded.
Nothing is uploaded. Open your browser’s Network tab while you split — you will see the file go nowhere.
Why the parts are not always the exact length you asked for
Video is not stored as a sequence of complete pictures. Every so often there is a keyframe — a full picture — and the frames after it store only what changed. A part that started anywhere else would begin by describing changes to a picture it does not contain, which is what produces the smeared opening you sometimes see on a badly cut clip.
So the cuts land on keyframes. Phone video usually has one every one to four seconds, so parts come out close to what you asked for. This page shows the real spacing in your file, and the exact start and end of every part, before it cuts anything.
The alternative is to re-encode the video so a cut can go anywhere. That is what most tools do silently, and it costs quality on every single cut. We would rather move the cut by a second and tell you.
What happens to the location tag
An iPhone writes the coordinates of the spot you were standing into the
video file, under com.apple.quicktime.location.ISO6709, alongside
the phone’s model and the moment you pressed record. Most people have no
idea it is in there, and every video they post carries it.
Each part this writes does not carry any of it across. A new container is written
from scratch around the compressed picture, and the QuickTime metadata is simply
not copied into it — not the coordinates in the old ©xyz
atom, not the newer Apple keys, not the original recording time. We checked this
rather than assumed it: the automated test builds a file carrying a location in
both formats, runs it through this tool, and then looks for the coordinates twice
— once with a proper parser and once by searching the raw bytes of the
result, because a leak that hides from one can still be found by the other.
Being exact about the limits: the finished file does carry the time you converted it, because the writer stamps one. That says when you used this page, which you already know; it does not say where you were or what recorded it. And this is about metadata, not about the picture — if a street sign is visible in the frame, no tool can help with that.
Common questions
Does splitting reduce the quality?
No. The compressed video and audio are copied across unchanged — the same bytes, in a new container. Each finished part is re-opened and checked, and what it actually contains is listed with the result.
How long does it take?
About as long as reading the file. There is no encoding, so it is limited by disk speed rather than by processor speed. A twelve-second clip splits into three parts in about a quarter of a second.
Is there a size limit?
None is imposed. The practical ceiling is your device’s memory, since each part is assembled there before you save it.
Does my video get uploaded?
No. All of the work happens in this page. The privacy page explains how the browser is prevented from sending it anywhere, and you can watch the Network tab to confirm it.
All 39 HermitFile tools
Every one of them runs in your browser. Nothing you open is uploaded, and nothing is stored.