Initializing... drag & drop files here
Supports: RW2
RW2 is Panasonic's Lumix raw photo format — a single still image, not a video. This converter renders that photo and encodes it as an AV1 video clip, so the page is really about three decisions: how long the frame is on screen, what fills the space around it, and what to do with the unusual .av1 file you get back. It also covers when you should not be on this page at all, because for most reasons people want a photo converted, a still-image format is the better answer.
.rw2 onto the page or click "+ Add Files". You can add several. Merge strategy starts on "Merge images", which builds one combined clip; switch it to "Video per image" for a separate clip per photo.The option surface for a photo-to-video conversion is narrower than a video-to-video one, because several controls have nothing to act on.
| Control | Starting value | Notes |
|---|---|---|
| Merge strategy | Merge images | Only meaningful when you upload more than one photo |
| Image Duration | 5 seconds per frame | The whole clip length when there is a single photo |
| Background Color | White | 25 colours: Black, White and 23 chromatic entries |
| Video Codec | AV1 | AV1 is the only entry — this is an AV1-specific page |
| File Compression | Quality Preset, on "Very High (Recommended)" | Constant Quality and Constraint Quality expose a 16-63 CRF slider |
| Video resolution | Keep original | Fixed Resolutions, height presets, or explicit Width / Height |
Two things are deliberately absent. There is no Audio Codec control, because a photograph has no soundtrack — the output is silent, and merging several photos does not change that. And there are no bitrate or target-file-size modes, because a still has no source bitrate to scale from; quality-based modes are the only ones that make sense and the only ones offered.
AV1 is a codec, not a container, so the encoder has to put the video inside something. Here that something is Matroska — the same format as an .mkv file — written to a file named .av1. That extension is not a standard container extension, and most players decide what to do with a file by looking at its name, which is why a perfectly valid clip can be refused before anything tries to decode it.
| If you want to… | Do this |
|---|---|
| Play it locally | Open it in VLC or mpv, which sniff the contents rather than trusting the extension |
| Make it behave like a normal file | Rename it to .mkv — the bytes are already Matroska, so nothing needs re-encoding |
| Put it on the web or in an editor | Convert it with AV1 to MP4, which most browsers and editors accept |
For reference, caniuse puts AV1 decoding at roughly 94% of tracked traffic, starting at Chrome 70, Firefox 67 and Edge 121, with Safari and iOS Safari from version 17 listed as partial. Note that Edge dropped AV1 entirely for a while before re-enabling it at 121. None of those browsers will load a file named .av1 directly regardless — the codec support and the container naming are separate problems.
Image Duration is the control that decides whether the clip is usable. A single photo held for 5 seconds is a sensible default; anything under about a second can flash past before some players have drawn their first frame. If you are merging a set of photos into a slideshow, the duration you choose applies to each frame, so ten photos at 3 seconds each is a 30-second clip.
Background Color matters more than it sounds. A camera raw has no alpha channel, so there is nothing transparent in the photo itself to fill — but the moment you pick an output shape that differs from the photo's, the picture is letterboxed and this colour is what fills the bars. That happens with the Fixed Resolutions entries and with an explicit Width x Height, since both name an exact pixel size; the height-only Preset Resolutions keep the source aspect ratio and so produce no bars at all. It starts on White, which is right on a light page and glaring in a dark video timeline. Set it to Black before you convert if the clip is going anywhere cinematic.
The resolution ceiling is the part nobody expects. Before encoding, every uploaded photo is rendered to a clean PNG frame — the step that turns Panasonic's raw sensor data into an ordinary RGB image using a neutral default development — and that step caps the long edge at 4096 pixels. It applies even on "Keep original", so a 5184 x 3888 raw is encoded from a 4096 x 3072 render. Frames already within 4096 px pass through at native size. Since AV1 is spending its bits on a single unchanging frame, this is rarely visible, but it does mean you cannot use this page to produce a 6K still-frame clip.
.av1 extension is the problem, not the video. Rename to .mkv, open it in VLC, or convert it with AV1 to MP4.Turning one RW2 into AV1 serves a narrow set of jobs: dropping a still onto a timeline that expects AV1, or merging several photos into a silent slideshow. If your actual goal is to view, print, edit or share the photograph, a still-image format wins every time — it keeps more of the render, opens everywhere, and skips the video-codec overhead entirely. Use RW2 to JPG for something shareable or RW2 to TIFF for something you will edit further. If you want a video but with sane playback, RW2 to MP4 produces a file that opens by double-clicking, and the RW2 converter lists the other targets.
No, and there is no setting that changes it. RW2 is a still photo, so there is no soundtrack to encode, and the Audio Codec control is hidden on this conversion entirely. Merging several photos into one clip produces a longer silent video, not a scored one. If you need audio, add it afterwards in a video editor.
Because this page's target is the codec rather than a container, and the file is named after it. What you actually receive is a Matroska file — the same structure as .mkv — carrying an AV1 video track. Renaming it to .mkv costs nothing and makes it behave normally, because no re-encoding is involved in the rename; only the name changes.
Whatever Image Duration says, which starts at 5 seconds per frame. One photo at the default gives a 5-second clip. If you upload several photos with Merge strategy on "Merge images", the duration applies per frame, so the total is roughly the number of photos multiplied by the duration.
Letterboxing. A camera raw has no alpha channel, so the photo itself never has holes to fill — but if the output resolution has a different aspect ratio from the photograph, the picture is centred and the leftover bars are painted in this colour. It starts on White, which is the wrong choice for most video work, so switch it to Black unless you specifically want a light frame.
Up to 4096 pixels on the long edge. Each photo is rendered to a clean frame before encoding, and that step clamps the long edge at 4096 px whether or not you touched Video resolution — so a 5184 x 3888 raw is encoded from a 4096 x 3072 render. Anything already inside 4096 px is encoded at its native size.
Because they would be meaningless. Bitrate modes work by scaling against a source's existing data rate, and a still photograph does not have one. That is why this page offers Quality Preset, Constant Quality and Constraint Quality but hides the constant-bitrate, variable-bitrate and target-size modes you would see converting one video into another.
Only if something downstream requires it. AV1's advantage is compression efficiency across many changing frames, and a clip of one unchanging frame is trivially small in any modern codec — so you pay AV1's much slower encode for a saving measured in kilobytes. If the destination accepts H.264, RW2 to MP4 finishes faster and plays in more places.
No. RW2 stores minimally-processed sensor data, and it is developed with a neutral default before encoding, so adjustments made in Lightroom, Capture One or SILKYPIX are not applied. Raw also holds far more tonal latitude than the encoded video can carry, so do your exposure and white-balance work first, export a finished image, and convert that.