MXF to MJPEG Converter

MXF to MJPEG: A Broadcast Wrapper Becomes a Bare Stack of JPEGs

MXF is a professional wrapper — SMPTE ST 377-1, currently published as ST 377-1:2019 — designed to carry picture, sound, timecode and descriptive metadata together through a broadcast or post chain. It is all structure.

A .mjpeg file is the opposite of structure. It is a run of complete JPEG images written back to back with nothing around them: no index, no timecode, no track list, no declared frame rate, no audio. Every frame is a self-contained keyframe, which is exactly why frame-accurate editing systems, machine-vision capture tools and old surveillance recorders liked the format — and exactly why it is a poor choice for anything you intend to watch or hand to someone else.

This page explains what comes out, what silently does not, and which single control changes the result.

What Reaches the Output and What Does Not

How to Convert MXF to MJPEG

1. Upload Your MXF File: Drop the .mxf on the page or click "+ Add Files". Several files can be queued and share one set of settings.
2. Set File Compression to Constant Quality: The panel opens on Quality Preset, which has no effect on this route — see the measurements below. Switch to Constant Quality and use its CRF slider, which on Motion JPEG runs 2 to 31 and opens at 5, lower being better.
3. Set Video Resolution and Trim if Needed: Video resolution opens on Keep original, with Resolution Percentage, Preset Resolutions, Width, Height and Width x Height as alternatives. Trim opens on Unchanged; switch it to Time Range to take a start time and duration out of a long master.
4. Convert and Download: Click Convert. Files are uploaded over an encrypted connection, re-encoded on our servers, and deleted automatically after a few hours — no sign-up, no watermark.

The Quality Slider Is the Only Lever That Works

Motion JPEG has no bitrate-targeting mode. The encoder is always driven by a quantizer value, and the pipeline sets one on every job — so the size and bitrate controls in the panel end up sitting on top of a decision that has already been made.

Measured on a 10-second 1280 x 720 MXF (MPEG-2 video, 48 kHz PCM audio, 11,235,897 bytes on disk):

Every entry in the top half is the same file, byte for byte. Seven quality presets and two very different bitrate requests all land on identical output, because the quantizer the encoder actually uses never changed. The bottom half is the same job with only the Constant Quality slider moved, and it spans a factor of four.

So: ignore Quality Preset and the bitrate boxes on this route, and set the number under Constant Quality. Around 2 to 4 for a master you will edit, 5 for general use, 10 to 15 when size matters more than the last of the detail.

Frequently Asked Questions

Why does my .mjpeg play back at the wrong speed?

Because a raw MJPEG file stores no frame rate, and the player has to guess — the usual guess is 25 fps. This is easy to reproduce: a 10-second source encoded at 10 fps produces 102 frames, and every probe of that file reports 25 fps, so it plays in about four seconds instead of ten. If timing matters at all, either keep the source frame rate so the guess is closer to right, or put the Motion JPEG stream in a real container. MXF to MOV writes a container that stores the frame rate properly.

Where did my audio go?

It was never written. The output is a bare JPEG stream, and that format has no concept of an audio track, so the audio essence in your MXF is not mapped into the output at all. A probe of the result shows exactly one stream. The Audio Codec list on this page renders seventeen entries with MP3 preselected, but nothing you pick there can reach a container that cannot hold sound — treat that control as inert on this route. If you need the audio, pull it separately with MXF to AIFF or MXF to MP3, or convert to MXF to MP4 and keep both tracks together.

Is Motion JPEG lossless?

No. Each frame is a baseline JPEG, and baseline JPEG is a lossy DCT codec. What Motion JPEG gives you is not losslessness but independence: no frame depends on any other, so you can cut anywhere, scrub instantly, and decode any single frame without reading its neighbours. That is why it survives in editing, machine vision and instrumentation work. It is also why the files are large — you are paying for a full still image on every frame.

Why is the output bigger than the MXF I started with?

Because MXF masters commonly carry a long-GOP codec like MPEG-2 that predicts most frames from their neighbours, while Motion JPEG cannot predict anything. On the 10-second 720p sample above, an 11.2 MB MXF became a 12.1 MB .mjpeg at the default quantizer even though the audio was thrown away. If growth is a problem, move the Constant Quality slider up rather than reaching for a bitrate box that will not respond.

What actually opens a .mjpeg file?

Anything built on ffmpeg — VLC, mpv, ffplay — plus the editing and machine-vision tools that expect the format. Browsers do not play .mjpeg files as video, though they will happily render a stream of the same JPEG frames served as multipart/x-mixed-replace, which is how IP cameras have done live view for two decades. Because every frame is a valid JPEG, you can also split the file into individual stills with almost any tool that reads JPEG.

Will interlaced broadcast footage be deinterlaced?

No, and this is worth checking before you commit a job. There is no deinterlacing step anywhere in this pipeline, so if your MXF holds interlaced fields, every JPEG in the output will carry visible comb artifacts on anything that moved. Motion JPEG makes this more obvious than a long-GOP codec does, because there is no inter-frame smoothing to hide it. Deinterlace in an NLE first if the master is interlaced.

Can I trim before converting, so I do not encode an hour of programme?

Yes, and you should. The Trim group opens on Unchanged; switching it to Time Range exposes a Start time and Duration pair that each accept plain seconds or HH:MM:SS.sss. Because Motion JPEG spends full-frame bits on every frame, trimming is by far the cheapest way to control the size of the output. For anything more elaborate than one range, Video Cutter accepts MXF directly.

Does this route preserve MXF timecode or the operational pattern?

No. Neither concept exists on the other side. MXF is a wrapper standard built around packages, UMIDs, essence descriptors and an operational pattern that tells a receiving system how the file is structured; a raw JPEG stream has none of that and no place to put it. If your workflow depends on timecode or MXF metadata surviving, this is the wrong conversion — keep the MXF, or re-wrap into a container that carries timecode.

What is the difference between .mjpeg and Motion JPEG inside AVI or MOV?

The codec is the same; the packaging is not. A .mjpeg file is the codec with nothing around it, which is why frame rate, duration and audio all disappear. Motion JPEG inside AVI or QuickTime keeps a real container: a frame index, a declared frame rate, a duration, and room for an audio track alongside. If your reason for wanting Motion JPEG is frame-independent editing rather than a literal .mjpeg filename, MXF to MOV is usually the better destination.

What happens to my uploaded file?

Your .mxf is uploaded over an encrypted connection, re-encoded on our servers, and both the upload and the .mjpeg it produced are deleted automatically after a few hours. Nothing is shared or made public, no account is required, and no watermark is added.