NAB 2008: Tape's Terminus Is Postponed But Still Inevitable
Speaking of high-def digital video…and for that matter, the 2008 CES…on Friday evening I’ll be headed to Las Vegas for the National Association of Broadcasters conference, through next Thursday morning. Sin City is, frankly, by no means my favourite place in the world to abide…about the only good thing I can say about it is its relatively close proximity to excellent national and state parks, and national conservation areas (ok, along with Spamalot, which I plan to see Wednesday night!). So why am I spending the weekend there, as a co-worker asked last night? For the technical conferences, of course, notably:
- Broadcast Engineering (where I’ll be specifically focusing my attention on mobile ATSC)
- Broadcast Regulatory & Legislative
- Digital Cinema Summit (which I’ve covered before), and the
- MPEGIF Master Class
But I digress. This post is about high-def digital video cameras, which will be showcased next week and were also present in abundance at CES. Judging from the sub-$1000 prices regularly showcased at sites like Dealnews and Techbargains, along with the burgeoning abundance of suppliers and models, the standard-to-high def transition is finally well underway. The recent resolution of the Blu-ray vs HD DVD format fight will only accelerate this evolution. But which video codec will reign supreme after the conversion is complete, and to what storage technology (or more likely, technologies) will the captured images most commonly archive?
Depends. As regular readers already know, I’ve long closely followed trends associated with the MPEG-2 based HDV and H.264 (aka MPEG-4 AVC, aka MPEG-4 Part 10)-based AVCHD formats. Back in December of 2006, I prophesied that AVCHD would eventually win out over HDV, primarily by virtue of its greater storage efficiency for a given level of video quality. My prediction was a reflection of the fact that H.264 is a more modern codec than MPEG-2, and the attribute is particularly attractive when the storage destination is per-GByte pricey (on a comparative basis, versus HDDs, optical storage and tape) flash memory.
That forecast is definitely playing out as envisioned, at least in the consumer space. To the best of my knowledge, only one new HDV camcorder was released at CES (and re-released a few weeks later at PMA, a fact overlooked by many of my press peers, which I noted with great bemusement), that being Canon’s HV30. Frankly, ‘new’ is a bit of a stretch, since the HV30 is a minor enhancement of the HV20, adding a 30 fps 1080-line progressive-scan capture mode (for that matter, the HV20 was essentially just a form factor re-layout of the HV10 predecessor, although it also supplemented the HV10’s 60fps 1080i capture with a 24 fps 1080p option and had slightly better low-light performance and a HDMI output option). Note, too, that Canon also has both HDD- and optical storage-centric AVCHD camcorders in its arsenal, and plenty of AVCHD-based products from other suppliers also showed up in Las Vegas in early January. AVCHD support is comparatively immature in consumer video packages, but the situation is similar to where HDV was at a comparable stage in its life cycle and will inevitably improve over time.
Things look somewhat different in the prosumer and entry-level professional ranks, at least in the short term. Here, HDV is still the dominant high-def codec (along with Panasonic’s proprietary DVCPRO HD), a quality-driven reflection of the comparative maturity of silicon-based MPEG-2 video processors and software-based algorithms for alternative CPU and DSP processing engines. Canon, for example, will be launching two new HDV camcorders at NAB next week, and I anticipate similar unveilings from other suppliers. Over time, I’m confident that H.264 quality-vs-bitrate metrics will improve (Ambarella, for example, is undoubtedly hard at work on the problem) and, in conjunction, that encoders supporting the full 24 Mbps bitrate potential of AVCHD will emerge. However, I doubt HDV will completely disappear in this particular market segment; it’s already well entrenched from hardware/software investment and workflow standpoints, and its use of the same MPEG-2 codec as ATSC digital television largely precludes the need for time-consuming and quality-degrading transcoding as part of editing, and prior to broadcast.
Continue reading with ‘High-Def Video: Storage Thoughts‘…