Initializing... drag & drop files here
Supports: DNG
DNG (Adobe Digital Negative) is a still camera-raw photo; MJPEG (Motion JPEG) is a video format where every frame is an independent, fully self-contained JPEG. So this is a still-to-video conversion: your single DNG becomes a short, silent MJPEG clip that holds that one frame on screen for a set duration. The raw file's wide tonal latitude is baked down to standard 8-bit video in the process. If you just want a normal viewable photo, DNG to JPG or DNG to TIFF is almost always the better target — read on for when MJPEG actually makes sense.
| Property | Value |
|---|---|
| Full name | Adobe Digital Negative |
| Type | Still image — camera raw container |
| Released | September 27, 2004 (Adobe) |
| Based on | TIFF / TIFF-EP (TIFF 6.0) |
| Data | Lossless raw sensor data, typically 12–16 bit per channel |
| Carries audio | No |
| Best for | Archiving raw captures; a vendor-neutral home for CR2/NEF/ARW/RAF |
| Common next step | Develop to JPG, TIFF, or PNG for viewing/sharing |
| Property | Value |
|---|---|
| Full name | Motion JPEG |
| Type | Video — intra-frame only (each frame a standalone JPEG) |
| Compression | DCT per frame; no interframe prediction |
| Typical ratio | ~1:20 (vs ~1:50 for H.264), so files are large |
| Common containers | AVI, QuickTime/MOV |
| Carries audio | Not inherently; audio is muxed separately when present |
| Best for | Frame-accurate editing, IP cameras, machine vision, low-latency capture |
.dng onto the page or click "Add Files." Add several at once if you want a slideshow-style clip.No. DNG stores 12–16 bit lossless sensor data, but MJPEG frames are ordinary 8-bit JPEGs. Highlight and shadow latitude that you would normally recover in a raw editor is baked in permanently here. If you need to keep that latitude for editing, develop the DNG to a 16-bit TIFF instead and edit there first.
MJPEG is intra-frame only — every frame is a full, independent JPEG with no interframe prediction, so it can't reuse pixels across frames the way H.264 does. That caps its efficiency at roughly 1:20 versus about 1:50 for modern codecs. A clip that fits comfortably as an MP4 can be several times larger as MJPEG. For a small, shareable clip, DNG to MP4 is the more practical target.
No. A DNG carries no audio, and this conversion produces a silent clip. MJPEG also doesn't carry audio inherently — when a camera records MJPEG with sound, the audio is muxed in separately, usually as PCM or ADPCM. Here the result is video-only.
A few real cases: feeding a single high-quality frame into editing or playback software that only accepts MJPEG, building a slow slideshow where each frame is held for several seconds, or testing a machine-vision or surveillance pipeline that expects an MJPEG stream. For ordinary viewing or sharing, converting to JPG is simpler and far smaller.
It's driven by the Image Duration setting, not a separate frame-rate field. A 5-second duration holds the frame for 5 seconds; picking a fraction like 1/24s renders the frame at roughly 24 fps. In our testing, a single DNG held for 5 seconds produces a short MJPEG clip of that one frame repeated — useful as a leader or placeholder, not as motion footage.
MJPEG is commonly wrapped in AVI or QuickTime/MOV. The frame data — independent JPEG images — is the same either way; the container mainly affects which players and editors open it without extra codecs. If your editor rejects the file, re-muxing into a different container usually resolves it without re-encoding.
Essentially, yes. Motion JPEG encodes each frame as a complete, standalone JPEG with no prediction between frames. That makes every frame independently decodable — handy for frame-accurate editing and low-latency capture — but far less space-efficient than codecs that reuse pixels across frames, which is why MJPEG files run large.
Because MJPEG is intra-frame only, every frame stands alone, so an editor can cut on any frame without decoding its neighbours and there is no interframe artifacting to worry about. That is exactly why capture cards, IP cameras, and machine-vision rigs still favour it. The trade-off is size: expect a much larger file than an equivalent H.264 clip.
It develops the raw. The DNG's sensor mosaic is demosaiced and rendered to RGB, then encoded as JPEG frames, rather than lifting the small preview thumbnail that some raw files embed. That is why the tonal latitude is baked in and why the output reflects a default rendering instead of a raw editor's adjustments.