Top priority for HD acquisition is file-based workflow support, says Nordahl
Industry veteran, consultant and HD expert Tore Nordahl recently released an update to his 142-page “HD ENG/EFP & HDV Camcorders for Broadcasting & Production” report.
Providing an in-depth look at where HD camcorder technology is headed and providing insight into developments in digital ENG transmission technology for camera backs, the report offers a unique perspective on the rapidly unfolding developments in HD acquisition.
HD Technology Update turned to Nordahl to glean some of the knowledge upon which he’s built the update to his report.
HD Technology Update: What should be the top ENG camera and recording format characteristics that networks, station groups and stations use in settling on an HD field acquisition camera and format?
Tore Nordahl: The top underlying issue to look at is where the infrastructure is going within TV facilities and TV stations. Everything will be file-based and networked in the future. That is the key issue for a TV station to look at first. They need to plan for the eventual transition to a file-based infrastructure, meaning that all of the servers and editing systems will be based on a file-based workflow. We have realized for sometime that the old tape cassette is now finally on its way out. And although people are still using a lot of tape, the emerging models of HD camcorders are all without internal tape cassette drives.
There are basically three major non-tape systems now emerging in the marketplace. That includes Sony’s optical disk, the PD cartridge, with the advantage that it can be used directly as archive; the disadvantage is that the sustained record bit rate is relatively low, but it is more than sufficient for good quality HD ENG acquisition.
Panasonic’s flashable RAM, the P2 card, is extremely fast, very portable. The disadvantage of course is that you would need to transfer to an archival medium for the longer term. But with a file-based workflow, material that is acquired by an HD camcorder is, of course, edited in a workstation attached to a storage area network presumably with archival facilities.
We mentioned that the P2 RAM is very fast and you can transfer much faster than real time, and this is particularly the case now that Panasonic is coming out with the AVC-Intra compression format, which is reported to be about 50Mb/s. That’s half of DVCPRO HD’s 100Mb/s. So, material of, let’s say, the same HD quality, being DVCPRO HD at 100Mb/s before, in the future being AVC-Intra at 50Mb/s and thus actually flowing through the file-based system twice as fast.
The third emerging system is the (Thomson) Grass Valley Infinity HD camcorder, featuring two internal storage subsystems, both removable. One is the CompactFlash, which is flashable RAM like the Panasonic P2, and the second is the Iomega REV Pro removable hard disk cartridge, which is similar to Sony’s optical PD cartridge except it is a hard disk. Just like the PD, the REV Pro cartridge may be used for archival purposes.
So, I think the decisions there between these three systems really will have to be largely based on a TV facility’s analyses as to how they would like to treat their file-based workflow and also how they would like to organize their longer-term archive. The HD ENG camcorder supplier with the most compelling story will get the largest share of the business.
HDTU: What do you envision for a timetable for the implementation of HD field acquisition?
TN: Looking at the DTV channel transition, of course, the magic year is 2009. So, you are really looking at two to three years from now. I predict that most, if not all, of the professionals shooting by then for TV-type program requirement will be shot in HD, including ENG.
Few, if any, professional facilities in the TV business will buy a new interim SD camcorder in 2007. They will buy HD camcorders with the capability of shooting SD, but, once they have it, they’ll start shooting HD sooner than they think. There is the opportunity to make a smooth transition from SD to 16:9 SD to full HD. I would say that over the time frame from now through 2009, nearly all professional TV facilities will have transitioned to full HD acquisition.
HDTU: Won’t the fact that a large audience is watching conventional NTSC sets — either via cable or an over-the-air D-to-A converter box — require stations to continue to shoot with their 4:3 audiences in mind?
TN: That is certainly an important consideration, and they would have to consider the framing of the scenes.
I think that by 2009, a majority of the home audience will actually have some sort of 16:9 set somewhere in the home. But, it certainly is one of several transitioning problems for the TV stations.
In my local market here in Los Angeles, KABC Channel 7 started full local HD news early this year, and when you watch it on a real HD set it is absolutely fabulous — particularly the quality from their news set, but also some of the shots from the field. The HD field acquisition problem is that you may not start out with enough HD equipment for the ENG trucks and helicopters, but that will fill in reasonably soon. But personally, watching KABC-7 news sometimes on my HD set and sometimes on my SD sets, I don’t see any significant problems with the way they are handling the 4:3 and 16:9 framing compromise.
HDTU: Once the 2GHz BAS relocation is complete and stations are contributing field footage on a 12MHz-wide digital channel with an 8MHz pedestal, will some feel pressure to move to 7GHz and 13GHz for ENG to relieve crowding in the 2GHz band or to use wider channels?
TN: I think there is crowding in both 2GHz and the 7GHz band, particularly in major markets. Nevertheless, I think that in the longer term broadcasters should try to move service to the 7GHz band in order to make it easier or even possible to get high-quality live HD footage from the field. The life blood of local TV stations is local news, and to be competitive, they need to deliver all of their news programming in real HD at the earliest possible time, including live from the field.
To transmit real HD over 2GHz in 8/12MHz bandwidth is possible, but you have to compress the HD signal quite significantly to a bit rate of less than 25Mb/s, introducing long GOP and sometimes unacceptable latency. It would be much better if you could utilize the 25MHz bandwidth in the 7GHz band to transmit up to 60Mb/s both from the cameraman to ENG van and from the ENG van to the TV studio, which will accommodate both Grass Valley’s Infinity HD ENG JPEG2000 mode and Panasonic’s new AVC-Intra, both at 50Mb/s. And it will very comfortably accommodate Sony’s XDCAM HD 35Mb/s.
HDTU: What role might MPEG-4 AVC H.264 or JPEG 2000 play in the future field contribution — especially given the interest in alternate means of backhaul to supplement traditional point-to-point microwave? (Specifically, I have WiMAX and EV-DO in mind even as HD acquisition grows in importance.)
TN: The data rate offered by EV DO is not sufficient for live HD ENG transmission, not by a long shot. WiMAX has possibilities with its theoretical 30-mile range and 70Mb/s data rate, but there are many other issues that may seriously limit the practicality of WiMAX for HD ENG live backhaul. For some time to come, I believe TV stations must rely on dedicated licensed microwave links.
At NAB, Panasonic announced their professional ENG format called AVC-Intra, compressing HD to 50Mb/s intra-frame, and in early May, Sony and Panasonic surprised us with their joint announcement of a consumer format called AVCHD, which has the potential of delivering very good HD picture quality at a data rate of less than 20Mb/s.
Panasonic AVC-Intra is intra-frame, so you don’t have the long GOP issues, which you try to avoid in professional ENG. The AVCHD format will soon creep into the professional domain, I predict, in the longer term taking over the market now being developed by the HDV camcorders that started to ship two or three years ago. So, AVC — or MPEG-4 Part 10 H.264 — is very important in the long term future for broadcasters both as an intra-frame ENG format, and as an auxiliary shooting format with less expensive HD cameras with AVCHD.
My personal favorite is still the JPEG 2000 format announced by Grass Valley at 50Mb/s and 75 Mb/s, both Intra-frame, which will also play a significant role in the future of professional ENG.
HDTU: The McGraw-Hill stations just announced they’ve committed to DVCPRO P2 HD for ENG acquisition and also will use the HVX200 for certain news situations, such as locker rooms, assignments abroad and investigative reporting. I know initially you weren’t enthused about professional HDV and HVX cameras because of their 1/3in sensors and low-light performance. What’s your opinion today about HDV and HVX for some ENG applications?
TN: Not exactly true. I was very positive about the HDV and HVX models for a wide variety of applications, but not for professional HD ENG applications for two primary reasons: 1/3-inch imagers are not sensitive enough for the variety of shooting conditions experienced by live news crews, and long GOP compression introducing latency. And, I qualified that by saying that Panasonic’s HVX200 was better suited for HD ENG than the HDV models because of the intra-frame DVCPRO-HD compression and the larger pixel area in their 960x540 imagers. The HDV products offered by Canon, JVC and Sony deliver remarkable HD performance. As long as the shooting conditions are relatively good, each will provide, I think, excellent images to supplement local HD news footage.
Tell Us What You Think!
HDTU invites response from our readers. Please submit your comments to firstname.lastname@example.org. We'll follow up with your comments in an upcoming issue.
For more information, visit: www.nordahl.tv/.