Digital Video: DV or not DV?
- 24 November, 2004 07:51
That is the question. And it seems that there's a considerable amount of confusion about the answer, so this month I'll be taking a quick look at the two main digital video recording formats, what they are, and what they're good for.
Put simply, digital video is a term that can be used to describe any video content that's stored as a series of zeroes and ones. This is instead of the older analog waveform found in your trusty old VHS video, or the photographic film of the Super-8 camera your dad's got lurking in his loft. This method of storing video has a number of advantages, not least of which is the ability to simply move this data from a tape to a hard disk for editing without the need for conversion or specialised capture hardware. It also helps to eliminate generation loss - in which repeated capturing, copying and editing degrades picture and sound quality - which was a common problem with analog video.
Will the real DV please stand up?
The first digital video standard to be introduced outside the broadcast industry was DV25 and, along with the MiniDV tapes it uses, it's still the most common format in consumer and semi-professional camcorders. It's often referred to as plain old DV - for Digital Video - but as this makes things a little confusing, it's best to give it its full name. The '25' indicates the bit rate of the video in Mbps, and this remains constant throughout your recording, unlike more contemporary formats.
This relatively high bit rate is both an advantage and a limitation. It makes it less prone to compression errors like blockiness or noise in fast-moving or detailed scenes, but also means that it takes up a lot more space on your hard disk. DV25's biggest benefit is less obvious, and you need to look a little closer to find it. Unlike more space-efficient video compression routines, DV25 only uses intraframe compression. This means that it reduces the data held in a single frame, on a frame-by-frame basis. When editing, this makes it much easier for your PC to rebuild a frame, or mix frame data with other material like video overlays or effects, and makes editing and rendering a much faster process.
Originally conceived as a means to compress data onto an optical disc, MPEG-2 uses a more efficient compression system. In addition to the intraframe compression of DV25, MPEG-2 applies a technique called interframe compression. This means that, as well as discarding redundant data within a single frame, MPEG-2 compression throws out video information from preceding and successive frames, replacing it with pointers and placeholders instead.
As well as being the standard used by all DVD movies and digital TV broadcasts, MPEG-2 is building up quite a following in the camcorder community, and is used by DVD camcorders, Sony's MicroMV camcorders and, more recently, JVC's rather lovely GR-PD1 high definition camera.
Being a very flexible format, MPEG-2 doesn't have a set bit rate or resolution, and can significantly reduce the amount of data required to store a video clip, which makes things a lot easier on your hard drive. However, it also means that your PC has to work much harder to rebuild a single frame of video data, adding a considerable amount to the time it takes to render even a simple edit. Not only that, but overly-compressed MPEG-2 video (which you'll often find in DVD-based camcorders) can look blocky and indistinct, particularly in those high-detail, poorly-lit or fast-moving scenes that DV25 handles so much better. Though, in its favour, camcorders that support variable bit rate (VBR) encoding can try to avoid this by applying less compression to frames that need more data, balancing this with more compression for simple or relatively static scenes.
The real irony here is that MPEG-2 can introduce its own form of generation loss if you're not careful, with your camcorder recording at one bit rate, and your editing system recompressing it to another. Do this too often and you'll see the kind of image noise that digital video was designed to avoid.
As a regular video editor, I strongly recommend using DV25. MPEG-2 is an excellent format when it comes to fitting video into a much smaller space, but only when you've done editing your footage and want to fit the finished masterpiece onto a DVD to share with other people.