Video Format Overview

by Dave Brewis

We are often asked about video formats so I thought I'd summarise some of the technical aspects of the common formats in use today.

DVCAM

DVCAM is Sony's prosumer video format which continues to serve the corporate communications business well. Used most appropriately where capturing content is more important than the finite quality of the image, DVCAM provides a flexible format in which to work, and is a great base from which to quickly create DVDs and video for the web, due to the relatively small file sizes it generates.

DVCAM captures at pixel resolutions of 720x576 (PAL) and 720x480 (NTSC), Standard Definition in other words, and generates files of around 3.6MB/s having used a 5:1 compression ratio and a sampling ratio of 4:1:1.

DVCAM is certainly for the more cost-conscious when choosing format. Our rates for DVCAM work reflect the fact that we haven't had to make such a large investment in storage, processing and playback equipment compared with high-end SD, HDV and HD formats.

We hire cameras to suit the project, but typically we'll use the Sony DSR-570 as our mainstay, and the Sony PD170 when mobility and speed are paramount, both of which support a 16:9 aspect ratio.

Digibeta

Digital BetaCam or 'DigiBeta' is Sony's digital format which up until recently has been the de facto post production standard in most professional facilities.

Digibeta also captures in Standard Definition... at pixel resolutions of 720x576 (PAL) and 720x480 (NTSC).

Despite having been around for some time, the quality of good DigiBeta cameras and lenses means that this is a cost-effective format in which to shoot, especially if you dislike the 'dv look', and are not concerned with HD.

For instance, if image detail and colour depth is important, but future proofing your content for higher resolution distribution (HDTV or HD-DVD) is less important, then DigiBeta can offer higher quality over DVCAM, albeit at a higher price.

DigiBeta supports a 4:3 or a 16:9 aspect ratio. It is a 10-bit format meaning it has a possible range of 1024 discrete gradations of colour or brightness compared with only 256 (8-bit). It samples colour at a ratio of 4:2:2, and uses a 2:1 compression ratio, which generates files of around 9MB/s.

HDV

HDV (not to be confused with HD) is High Definition Video recorded on the same digital videotapes that miniDV is recorded on today. HDV is recorded to tape in the same format that HD is currently broadcast in, namely an MPEG-2 transport stream, which provides the amount of compression needed to get high definition resolution images on to a small miniDV tape.

High definition images appear crisper and more detailed because they possess a higher resolution than Standard Definition.

JVC produce HDV cameras which operate at ë720p' (a resolution of 1280x720 pixels with a progressive signal). Sony on the other hand produce the HDR-FX1 camera which operates at ë1080i' (a resolution of 1920x1080 pixels with an interlaced signal).

The JVC's 720p HDV format captures in 8-bit 4:2:2 and is compressed using MPEG-2 resulting in datarates of around 2.4MB/s. The Sony 1080i format captures in 8-bit 4:2:2 and is compressed using MPEG-2 resulting in datarates of approximately 3.1MB/s.

The advantage to shooting in HDV is that your project and footage will still be usable in its native resolution in the coming years, as television sets and DVD players all are made primarily for High Definition material.

For instance, the event video you shoot now will be able to be enjoyed in a HD-DVD player in the future, whereas if you were to shoot the event today in SD, it would be necessary to upscale the video, with a resultant loss in image quality. SD video upscaled to HD generally looks poor.

HD

HD is without doubt the highest quality video format.

There are many flavours of HD, such as Panasonic's DVC Pro HD and Sony's HDCAM. However, HD always refers to an image with a 16:9 aspect ratio at a pixel resolution of either 1920x1080 or 1280x720.

The captured HD image may be either be interlaced (i) or progressive (p) and depending on the camera may support frame rates of 24, 25, 29.97 or 30 fps.

Often an HD format will be described as ë1080i50' or ë720p30'Ö for example, 1080i50 means the image was shot at a resolution of 1920x1080 pixels, is interlaced (has two fields per frame) and was shot at a rate of 25fps (frames per second).

Most HD video cameras acquire the source image in 4:4:4 RGB, then reduce it to a 4:2:2 YUV image before further downsampling, bit depth reduction and additional compression is applied. This brings the datarate down to something recordable onto videotape, somewhere in the 14 to 23 MB/sec range.

If we're shooting a commercial we recommend shooting in HD and creating both a SD and HD master of the Final Cut. That way the TV station can broadcast the commercial in SD up until the time the broadcaster switches to broadcasting in HD.