Initializing... drag & drop files here
Supports: MRW
This page turns a Minolta RAW photograph into an AVI clip that holds that one frame on screen. AVI is the format people reach for when something old has to accept the file — a legacy editor, a DVD-authoring tool, a Windows workflow that predates modern containers. That legacy character is also why this conversion has more sharp edges than the modern video targets.
Three of them matter enough to lead with. The default codec is MPEG-4, which caps output at 2,048 pixels and will therefore downscale a full-resolution DSLR frame. The "Quality Preset" control does nothing useful on this page. And the quality slider is labelled CRF but is not a CRF scale at all. Each is explained below, with what to do instead.
.mrw files onto the page or click "+ Add Files". With several queued, "Merge strategy" chooses between one combined clip and one clip per photo..avi. Files upload over an encrypted connection, are processed on our servers, and are deleted automatically after a few hours. No sign-up, no watermark.Because the source is a still rather than a video, every bitrate-based and target-size mode is hidden — there is no source data rate to scale against. That leaves three modes, and on the default MPEG-4 codec they behave very differently.
| Mode | Behaviour on MPEG-4 with an image source | Verdict |
|---|---|---|
| Quality Preset (default) | Works by scaling the source bitrate. A still has no bitrate, so there is nothing to scale | Inert here — moving it changes nothing |
| Constant Quality | Emits a real quantiser value, but the field is labelled "CRF" while the scale is 1–31 with lower meaning better, and it does not open at a value tuned for that scale | Functional, but easy to misuse |
| Constraint Quality | Same 1–31 quantiser plus a Max bitrate ceiling | Use this one |
The labelling trap is worth dwelling on. On modern codecs, CRF runs 0–51 or 0–63 and a value around the middle is a sensible default; on MPEG-4 the encoder takes a quantiser from 1 to 31 where 1 is near-lossless and 31 is heavily compressed. A number that would be unremarkable on H.264 lands deep into the compressed end of this scale. Do not carry x264 habits across — treat whatever the field opens with as a starting point to move downward, and judge by the output.
If none of this appeals, the simplest fix is to change the codec. Select H.264 or MPEG-4 from the "Video Codec" list according to what has to read the file: H.264 inside an AVI is unusual but widely readable, and it makes Quality Preset behave properly because it is a genuine CRF codec.
MPEG-4 is registered in our encoder table with a maximum frame size of 2,048 × 2,048, and image-to-video conversions scale the still down to fit before encoding. Running a 3,008 × 2,000 frame — the native size of a Maxxum/Dynax 7D capture — through the same MPEG-4 settings this page uses produced a 2,048 × 1,360 clip: the width clamped to the ceiling, the height following the aspect ratio and then rounded down to the multiple of four MPEG-4 requires.
That affects every camera in the MRW family, since all of them recorded frames wider than 2,048 pixels at full resolution. If losing that detail matters, either set "Video resolution" explicitly so you control the scaling, or pick a codec without the restriction.
| Codec in the dropdown | Frame-size ceiling | Real CRF codec? |
|---|---|---|
| MPEG-4 (default) | 2,048 × 2,048 | No — quantiser 1–31 |
| Xvid | 2,048 × 2,048 | No — quantiser 1–31 |
| MPEG-1 | 4,095 × 4,095 | No |
| MPEG-2 | 4,096 × 4,096 | No — quality mode is discarded entirely |
| H.264 | 4,096 × 4,096 | Yes, CRF 0–51 |
| H.265 | 8,192 × 8,192 | Yes, CRF 0–51 |
| AV1 | 8,192 × 8,192 | Yes, CRF 0–63 |
| H.261 | 352 × 288 | No — frame snapped to a fixed size |
| H.263 | 1,408 × 1,152 | No — frame snapped to a fixed size |
If the goal is to look at the photograph, AVI is the wrong destination — a video container adds nothing and costs you resolution under the default codec. Convert to a still format instead: MRW to PNG at the Highest preset for a lossless render, or MRW to TIFF with LZW compression for archiving. And if the AVI is only needed because something downstream is old, check first whether that tool accepts MP4 — most software from the last fifteen years does, and MP4 avoids every trap on this page.
Because it works by scaling the source's bitrate, and a photograph does not have one. On video sources that fallback has real numbers to work with and the control behaves normally; on a still it has nothing to scale, so the setting has no effect. This is specific to non-CRF codecs like MPEG-4, Xvid and MPEG-2 — on H.264, H.265, VP9 or AV1 the preset maps onto a real CRF value and works as expected.
Because MPEG-4 has no CRF mode at all. What the encoder actually accepts is a quantiser scale running 1 to 31, and the interface reuses the CRF label across every codec rather than renaming the field per encoder. The direction is the same as CRF — lower means better quality and a larger file — but the range is much narrower, so a value borrowed from H.264 habits sits far further toward heavy compression than you would intend.
Depends entirely on what has to open the file. The list holds twenty-one options; only a few are sensible here. MPEG-4 is the default and is what DivX-era players expect. Xvid targets the same generation of software. MJPEG suits very old capture and editing tools and keeps each frame as an independent JPEG. H.264 is the best choice if the receiving software is merely old rather than ancient. Ignore H.261 and H.263, which snap the frame to a handful of fixed videoconferencing sizes and will distort a photograph.
It is a genuine AVI. That is worth saying because our .av1 and .hevc targets are Matroska files wearing a codec name and need renaming to .mkv before some players will touch them. AVI is a real container with a real muxer behind it, so the file you download is what the extension claims.
Because MPEG-4 Part 2, the codec behind the default option, is roughly two decades older than H.264 and considerably less efficient. AVI as a container also lacks modern indexing and metadata efficiencies. For the same visual quality you should expect a noticeably larger file. If size matters and compatibility does not, an MP4 or an HEVC clip will be much smaller.
No. The RAW decoder develops MRW's 12-bit sensor readout down to 8 bits per channel before any video encoding happens, and MPEG-4 in an AVI carries 8-bit video. The highlight and shadow latitude that made the RAW worth shooting is spent during development. Keep the original .mrw if you may want to re-develop the shot with different exposure or white balance.
Yes. Upload them together and leave "Merge strategy" on merging images, and each still is held for the Image Duration in turn within a single clip. Switch it to one video per image if you would rather get separate files. Note that the codec ceiling and quality settings apply identically to every file in the batch.
Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours. No sign-up, no watermark, and files are never shared or made public. RAW files are large, so the upload usually takes longer than the encode of a single-frame clip.