Initializing... drag & drop files here
Supports: DV
A .dv file is a raw DV-DIF stream lifted off a MiniDV, DVCAM, or Digital8 tape over FireWire — an intraframe format from 1995 that spends about 25 Mbit/s on standard-definition video and works out to roughly 11.9 GB per hour. AV1 sits at the opposite end of the timeline: a royalty-free codec finalised in 2018 that is one of the most efficient ways to store that footage long term. This page re-encodes the DV video to AV1 and writes it into a Matroska container with the tape's audio track intact.
| Property | Value |
|---|---|
| Standard | IEC 61834 family (the DV "Blue Book" specification) |
| Released | 1995 |
| Compression | Intraframe only, DCT-based — every frame is a keyframe |
| Video data rate | ~25 Mbit/s for DV25 |
| Frame size and chroma | 720×480 with 4:1:1 (NTSC), 720×576 with 4:2:0 (PAL) |
| Scanning | Interlaced standard definition on consumer camcorders |
| Audio | 48 kHz 16-bit stereo (the usual mode), 32 kHz 12-bit four-channel, or 44.1 kHz 16-bit stereo |
| File form | Raw elementary stream, .dv or .dif; no container |
| Typical size | ~11.9 GB per hour of tape |
| Property | Value |
|---|---|
| Developed by | Alliance for Open Media |
| Version 1.0.0 released | 25 June 2018 |
| Licensing | Open and royalty-free |
| Efficiency | Facebook's 2018 tests measured roughly 34% lower bitrate than VP9 and about 50% lower than x264 at comparable quality |
| Standard containers | ISO base media file format (MP4), Matroska, RTP |
| Browser playback | Chrome 70 and Firefox 67 (May 2019) onward |
| Container written on this page | Matroska |
| Default audio codec on this page | Opus |
| Encoder used here | SVT-AV1 |
.dv or .dif capture, or click "+ Add Files" to browse. Batch upload works, and every capture in the queue takes the same settings..av1 is a codec name, not a container format, so there is no such thing as a native AV1 file structure. Our pipeline resolves that the way the codec's own ecosystem does: the output is a Matroska file that carries the AV1 video track alongside the Opus audio track decoded from the tape. It is a complete, playable file rather than a raw elementary bitstream — VLC and mpv open it as-is.
The one practical wrinkle is the extension. Some players and media libraries decide what to do by file extension rather than by inspecting the container, and they do not recognise .av1. If a player refuses the download, rename it to .mkv; nothing inside the file changes and the same bytes then open normally. If you would rather not deal with that at all, DV to MKV writes the same container with a conventional extension, and DV to MP4 produces the most broadly compatible file of the three.
No. AV1 is a video coding format, and the AV1 specification defines a bitstream, not a file wrapper. The standardised carriers for AV1 are the ISO base media file format, Matroska, and RTP. When you ask for .av1, our converter picks Matroska as the container and writes the AV1 track into it, which is why the file plays properly in VLC or mpv instead of behaving like a raw stream.
Yes. The Matroska file carries an audio track alongside the video, encoded as Opus by default. That is a real re-encode of the tape's PCM audio, not a copy, so the DV audio is decoded and compressed once. If you would rather have a codec your other tools already handle, the Audio Codec control on this page also offers AAC, MP3, FLAC, Vorbis, AC3, EAC3, MP2, and DTS.
It depends on what you are optimising. AV1's efficiency advantage is real and it is the right answer if the goal is the smallest possible archive of a stack of digitised tapes. But AV1 encoding is far slower than H.264 or H.265, and SD footage is small to begin with, so the absolute bytes you save on a 720×480 source are modest compared to what you would save on HD material. For everyday viewing and sharing, DV to MP4 is the pragmatic pick; for a compact long-term master, AV1 earns its keep.
The page starts on the Quality Preset mode at "Very High (Recommended)", which the encoder translates into its own quality target. If you want to set the number yourself, switch File Compression to Constant Quality — the CRF field starts at 30, and AV1's CRF scale runs from 0 (largest, highest quality) to 63. Lower values mean bigger files.
Modern ones, usually. Chrome and Firefox have decoded AV1 since 2018–2019, recent Android phones and newer smart TVs include hardware AV1 decoders, and desktop players like VLC and mpv have supported it for years. Older TVs, older phones, and set-top boxes very often do not, and software decoding AV1 on weak hardware stutters. If the file has to play on unknown equipment, convert to DV to MP4 instead — H.264 in MP4 is still the safest bet by a wide margin.
Because AV1's compression gains come from a much larger search space than H.264 or VP9, and the encoder has to explore it. We use SVT-AV1, which is the fastest of the mainstream AV1 encoders and is specifically built for threading, but it is still slower than an x264 encode of the same clip. A full-tape DV capture is also a big upload before any encoding starts, so trimming the section you actually need is the single biggest time saver.
Yes. Consumer DV is interlaced standard definition, and this conversion does not deinterlace — the field pattern from the tape is carried into the AV1 frames rather than removed, which is the right behaviour if the material is heading back into an editor. On a progressive display, fast motion may show combing. If you need a clean progressive master, deinterlace in an editing application or a desktop encoder before or after this step; there is no deinterlace control on this page.
The file is uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared or made public. The conversion is always a re-encode: our AV1 path decodes the DV frames and encodes them fresh with SVT-AV1, so this is one lossy generation away from the tape, not a rewrap of the original data.