HDV’s Coming Up
Issue: Volume: 28 Issue: 2 (Feb 2005)

HDV’s Coming Up

HDV is nothing more than a format-a future format at that-and so it is perhaps not all that interesting to the active videographer or designer. But HDV is arguably the most intriguing format to come along since MPEG-2 (which also got a lot of press before there were really any usable products). Simply put, HDV is an industry-standard way to write high-definition MPEG-2 video, either 720p or 1080i, onto standard DV tape.

The promise of HDV and, thus, the anticipation is very much like that of early DV nearly a decade ago. DV brought professional-quality digital acquisition to handheld camcorders and immediately shook up the high-end versus low-end industry caste system. DV’s quality rivaled that of the then-standard professional tape format, Betacam SP, and while there was debate about which was better, more professional, and appropriate for what purposes, DV was an eye-opener on price alone. In the decade since, regardless of that debate, DV has found a comfortable place among the tools of the trade. Now HDV is likely to travel a similar course for high-definition acquisition.

The basics of HDV are actually remarkably straightforward. Instead of the 25Mb/sec intraframe compression of standard-definition DV, HDV uses the temporal, interframe compression of MPEG-2 to squeeze higher resolution into the same 25Mb/sec (for 1080i, 19Mb/sec for 720p) bitstream of standard DV. There’s greater compression at work, but mostly it’s just more efficient use of the available storage capacity.

Makers of professional editing systems have long favored intraframe compression formats, such as Motion JPEG and DV for editing, and there are a couple of important reasons why. First, since most individual frames don’t exist separately in interframe compression, and instead use data from neighboring frames, they often have to be re-built on the fly for an editor to be able cut on an exact frame, or to step or shuttle forward or backward through a specific clip or transition. Historically, re-creating frames that fast has been difficult to do. But with today’s CPUs churning through data so quickly, it is becoming less of a concern.

Interestingly, it’s the editing systems vendors-especially editing software companies that cater to a more price-sensitive audience-that are the most anxious for HDV camcorders to hit the market. While they focus on what is here today-most of them privately, and some not so privately-all see HDV as a big part of their future.
Sony’s new HVR-Z1 HDV camcorder could help bring about the next major acquisition revolution.

The second reason MPEG and other interframe codecs often get scoffed at for professional editing has to do with the potential loss of quality through the concatenated artifacts and errors that arise from the decompression-recompression cycles of rebuilding frames. On average, only one of every 15 MPEG frames is a self-contained intraframe, or I-frame, and no serious editor wants to be limited to cutting every 15th frame. MPEG-2 editing systems can build new I-frames and reconstruct the MPEG GOPs (Groups of Pictures between I-frames), but that means building new I-frames based on the most heavily compressed frames of the original footage.

Some editing software companies, such as Canopus and Avid, have created proprietary intraframe HD codecs specifically for editing. By transcoding HDV into an intraframe codec during import, they effectively limit that artifact concatenation by working in higher quality throughout, albeit while dramatically increasing the storage requirements compared to native HDV. Still, at a basic level, that is simply fodder for a “what’s good-enough quality” argument much like that of the early DV days. HDV faces the same scrutiny, and that will surely leave many, at least at the higher levels of production, staying with more expensive HD acquisition formats.

On the other hand, at expected camcorder price points under $5000, HDV seems destined to follow a similar path as DV, with a strong appeal to DV filmmakers, documentarians, and small studio professionals, as well as in-the-field news gatherers and other high-end professionals for whom an essentially disposable camcorder has value. Exactly how high HDV goes up the professional video food chain remains to be seen, but the basic level of excitement is obvious. And that’s been true for nearly two years, ever since JVC introduced its single-chip, consumer-focused JY-HD10 using the HDV format.

Essentially, HDV has been all theory for the past couple of years, with the exception of the JVC JY-HD10, although it is considered under-featured by professional standards. But this past fall, Sony introduced not one, but two, sub-$5000 HDV camcorders: the HDR-FX1 and the more professionally oriented HVR-Z1. And while that brings the total number of available HDV camcorders to a mere three, it feels like the start of the next major acquisition revolution.

Both Sony models use the familiar design of the venerable VX1000/2100 series. Both models can record in either standard-definition DV or high-definition HDV. (Sony’s chosen HDV resolution is 1080i, while JVC’s is 720p, and neither supports both formats.) And both capture image data through the same 3 x 1/3-inch native 16:9 CCDs and have the same 0.44-inch viewfinder.

The differences between the two are features that cater to either consumer or professional users. For example, the $3699 HDR-FX1 supports recording in HDV, DV (SP), and DV (LP, or long play) whereas the $4900 HVR-Z1 drops LP recording in favor of DVCam support. The Z1 also shows an action safe zone, as well as 4:3 safe zone, and it has more color adjustment and correction features for set-up levels.

Is HD acquisition finally hitting the mainstream? Will this “low-end” HD be good enough for professional work? These questions have surrounded HDV for the past year, and industry pundits have taken sides. But with the release of these two camcorders, the marketplace is about to speak. It’s time to listen and to look for yourself.

Jeff Sauer is a contributing editor of Computer Graphics World and director of the Digital Video Group, an independent research and testing organization for digital media. He can be reached at jeff@dtvgroup.com.