Reset to defaults
Need help?Turning Minolta RAW Frames Into a Transport Stream
MRW is the raw format Minolta and later Konica Minolta wrote — a block-structured file that opens with the signature bytes 00 4D 52 4D and holds the sensor data alongside separate blocks for the picture dimensions and Bayer pattern, a full TIFF 6.0 header carrying the Exif tags, the white-balance gains measured at capture, and the camera's requested saturation, contrast, sharpness, ISO and colour mode. Konica Minolta announced its withdrawal from cameras on 19 January 2006 and left the business on 31 March that year, consigning service and the SLR line to Sony, so nothing has written a new .mrw in two decades.
TS is a completely different animal: an MPEG-2 transport stream, the broadcast container built to be joined mid-stream and to survive packet loss. This page walks a folder of MRW stills into one, holding each frame on screen for as long as you tell it to.
How to Convert MRW to TS
- Upload Your MRW Files: Drag them onto the page or click "Add Files". Upload order is playback order, so name the files so they sort the way you want them shown.
- Choose a Merge Strategy and Image Duration: Merge strategy opens on Merge images, which builds one transport stream containing every frame; Video per image writes a separate
.tsfor each MRW instead. Image Duration opens at 5 seconds per frame and runs from 1/60 second (for a 60 fps timelapse) up to 10 seconds. - Set the Codec, Quality and Frame Size: Video Codec opens on H.264, with DivX, H.265, MPEG-1, MPEG-2, MPEG-4 and Xvid alongside. File Compression opens on Quality Preset; Constant Quality exposes a CRF slider running 16 to 51 where lower is better. Video resolution opens on Keep original — use Fixed Resolutions for an exact frame such as 1920 x 1080, since Preset Resolutions set only the height.
- Set the Background Colour and Convert: Background Color opens on White and fills the bars wherever a 4:3 or 3:2 still does not match the output frame. Click Convert. Files are uploaded over an encrypted connection, processed on our servers and deleted automatically after a few hours — no sign-up, no watermark.
Walk-through: How the Timing Actually Works
The encoder runs at a fixed 1 frame per second and repeats each still for as long as its Image Duration says. That sounds like an odd design until you see what it buys: a 5-second hold is five identical encoded frames rather than 125, so a slideshow of stills compresses to almost nothing beyond the first keyframe of each image.
It also means Image Duration is the only timing control, and it applies to every frame in the batch equally. There is no per-image timing and no transition or crossfade stage — one still cuts straight to the next.
Choose it by what you are making:
- A slideshow to look at: 3 to 5 seconds per frame. The default of 5 is a comfortable reading pace if the images have detail worth studying.
- A timelapse from interval shots: 1/24, 1/30 or 1/60 second per frame, which give 24, 30 and 60 frames per second of playback respectively. A hundred MRWs at 1/30 second is a little over three seconds of footage.
- A product turntable: 1/10 or 0.2 seconds per frame reads as a deliberate rotation rather than a video.
Two things the page cannot do for you. The output is silent — the audio controls are not rendered on an image source at all, because there is nothing to encode — so a soundtrack has to be added afterwards in an editor. And the aspect ratio of your stills is unlikely to match a 16:9 video frame, which is what the Background Color control exists for: leave it on White for a clean gallery look, or switch it to Black if the result is going on a television.
What Survives the Trip Out of a RAW File
| Held in the MRW | What happens here |
|---|---|
| Sensor data with the Bayer pattern intact | Demosaiced on our servers into an ordinary RGB image before encoding |
| White-balance gains measured at capture | Applied during that development step; you cannot change them afterwards |
| Requested saturation, contrast, sharpness and colour mode | Camera settings recorded in the file, not adjustments you can revisit once the frame is video |
| Full Exif in the embedded TIFF header | Discarded — a transport stream has no place for per-frame camera metadata |
| Highlight and shadow latitude beyond the visible range | Gone. Video is 8-bit and this is a one-way development |
| Frame dimensions | Preserved unless you resize. No MRW body went beyond roughly eight megapixels, so these frames sit comfortably under the 4,096-pixel ceiling the image-to-video path applies |
If you might want to grade or re-crop these images later, develop them to a still format first and keep those as your masters: MRW to TIFF writes a full-resolution still, and MRW to JPG is the compact option.
Codec Choice on a Still-Image Source
| Codec | Behaviour with stills | Verdict on this page |
|---|---|---|
| H.264 (default) | Constant-rate-factor encoding works properly; a held still costs almost nothing after its first frame | The right choice unless something specific demands otherwise |
| H.265 | Same CRF behaviour, smaller files, narrower hardware support | Good if you control the playback device |
| MPEG-2 | No constant-quality mode at all; a resolution-derived minimum bitrate applies, around 7.5 Mbps at 1080p | Only if a broadcast or DVD workflow demands MPEG-2 |
| MPEG-1 | Same limitation, older and larger | Rarely worth it |
| MPEG-4, Xvid, DivX | Quantiser-based rather than CRF-based. Quality Preset has no source bitrate to scale from on a still image, so use Constraint Quality if you choose one of these | Only for legacy hardware targets |
The short version: leave it on H.264 unless a specific device is dictating the answer, and if you do switch to one of the quantiser codecs, move File Compression to Constraint Quality so you still have a working control.
Frequently Asked Questions
Why choose TS rather than MP4 for a slideshow?
Because a transport stream is designed to be broken into pieces and played from anywhere in the middle, which is exactly what streaming servers, broadcast chains and some set-top boxes need. HLS, for example, is built on transport-stream segments. If none of that applies to you, MP4 is the better everyday choice — smaller for the same picture and accepted by more software — so MRW to MP4 is probably the page you want instead.
Which Minolta cameras wrote MRW files?
The published format documentation covers the DiMAGE bridge cameras — the 5, 7, 7i and 7Hi, then the A1, A2 and A200 — along with the Dynax/Maxxum 7D digital SLR. After Konica Minolta left the camera business at the end of March 2006 and Sony took over the SLR line, the Alpha bodies that followed used Sony's own ARW format instead, so MRW stopped at that point.
Can I put a soundtrack on the result?
Not on this page. An image source produces silent video and the audio controls are not rendered at all, because there is no audio stream to configure. Add music afterwards in a video editor, which is the better workflow anyway — you can audition several tracks against the same render instead of re-encoding each time.
Why is there white space around my photographs?
Because MRW frames are 4:3 from the DiMAGE bodies and 3:2 from the DSLRs, and a video frame is usually 16:9, so the difference has to be filled with something. That something is the Background Color, which opens on White. Switch it to Black for a cinematic look, or choose a Fixed Resolution that matches your stills' shape — 1280 x 1024 is close to 4:3, for instance — and the bars mostly disappear.
How many frames do I need for a smooth timelapse?
Multiply the seconds of footage you want by the playback rate you choose. At 1/30 second per frame you need 30 MRWs per second of output, so a ten-second timelapse takes 300 frames. Below about 12 frames per second the result reads as stop-motion rather than motion, which may be exactly what you want for a turntable or a construction sequence.
Will the video look as good as the RAW file?
No, and it is worth being clear about why. A raw file holds a wider range of tones than any display can show and lets you decide afterwards how to map them; video is 8-bit with the mapping already baked in. On top of that the picture is compressed. What you get is a faithful rendering of one particular development of your raw frames — good enough to watch, not a substitute for keeping the originals.
Does the upload order really decide the playback order?
Yes, and the reliable way to control it is filenames. Cameras number frames sequentially, so files usually sort correctly on their own, but if you have renamed anything or mixed material from two cards, add a numeric prefix before uploading. There is no reordering control after upload, so a batch that sorts wrongly has to be re-uploaded.
Will the file be huge?
Usually not, because of how the encoder handles held frames. Each still is repeated at one frame per second for its Image Duration, and the repeated frames compress to almost nothing — nearly all the bytes go into the first frame of each image. A hundred-image slideshow at 5 seconds a frame is therefore roughly the cost of a hundred compressed stills, not the cost of eight minutes of moving video. Transport-stream packet overhead adds a few per cent on top.
How are my files handled during conversion?
Your MRW files are uploaded over an encrypted connection, developed and encoded on our servers, and both the uploads and the finished stream are deleted automatically after a few hours. Nothing is shared or made public, and there is no sign-up, no watermark and no account required.