Key Points
We recently attended a demonstration of the reception of digital television in Baltimore given by Sinclair Broadcast Group. Using simple antennaes and current generation consumer grade receivers, the European digital television standard, COFDM, consistently delivered pictures, while the current U.S. standard, 8VSB, often failed to deliver any picture.
8VSB's problem is multipath; the inability for receivers to "reassemble" signals received from many different incoming paths. We believe that multipath is more common in urban markets. ***
While we ultimately do not know whether there is any real momentum to change the U.S. standard within the industry itself, we believe the Sinclair test may apply serious pressure to manufacturers of 8VSB receivers.
* If a better receiver can not be developed, the value of digital television may be seriously compromised. Changing the standard, which we believe is not likely, could delay the roll-out of digital television for years.
* We believe the success of digital television is important to local broadcasters and the U.S. government.
Summary - Update on Digital Television. You Have to Receive to Believe - 8VSB Versus COFDM.
We recently attended a demonstration of digital television in Baltimore that was sponsored by Sinclair Broadcast Group. The demonstration was compelling in the simplicity of its message; using simple antennae and currently available consumer grade digital television receivers, one transmission standard, COFDM, now common in Europe, consistently delivered pictures, while the U.S. transmission standard, 8VSB, consistently failed to deliver pictures.
Sinclair's demonstration was made with the consumer in mind; the company believes that unless the technology can be accessed easily and cheaply, that it may be rejected.
We believe that Sinclair's motive in providing this demonstration is also simple to understand. While fully supportive and appreciative of all of those that have helped broadcasters secure a piece of digital spectrum, most notably the Advanced Television System Committee (which was ably led by Richard Wiley, the former Chairman of the Federal Communications Commission) and the National Association of Broadcasters, Sinclair wants to ensure that broadcasters will be able to take full advantage of the digital spectrum's potential.
We believe that Sinclair's message is simple; if current 8VSB receivers can not be improved dramatically, then the industry may have to consider a technology (such as the European technology, COFDM) which demonstrates that it can be received now. In any case, the development of digital could slow (or not according to how quickly new technology is developed) while waiting for better 8VSB receivers (more likely scenario) or conversion to another standard (presumably COFDM, a less likely scenario).
We believe that the 8VSB issue may not have as much impact on the networks as it will on non-network owned broadcasters. Two networks, NBC and CBS, have already negotiated cable carriage agreements for digital offerings with cable operators.
The broad digital television issue is key to the television industry; television broadcasters will have to spend billions of dollars to develop digital television and will not be able to recover this investment in any meaningful way if the transmission standard does not allow broadcasters to deliver high definition or multicasting video services consistently to the consumer. Also, for those broadcasters who would consider delivering non- video ancillary services, a weak digital signal (or a standard which is not appropriate for mobile services) could adversely impact broadcasters' ability to earn incremental revenues in these services as well.
Lastly, we believe that the government may be interested to see digital television do well. The government should be motivated by creating current revenue from ancillary fees (television broadcasters which create subscription services with their digital spectrum must pay the U.S. Government 5% of gross revenues) and future revenues from the auction of the analog spectrum in 2003. We believe the government would ultimately receive higher bids if the broadcasters, who are permitted to bid on their digital licenses, have already developed robust digital services.
Within the context of these issues lie some very basic, but unanswered questions.
What are the capabilities of the 8VSB standard?
How much better can 8VSB receivers become?
Under what environments (urban, suburban, rural) and conditions does 8VSB actually work well in video delivery?
Should the Government intervene and impose standards on receiver manufacturers?
How long will it take for adequate 8VSB receivers to be commercially available?
If necessary, what cost in time and money would a conversion to another standard take?
From the Consumer's Perspective, the 8VSB Digital Television Standard Still Needs Work. Recently, we attended a demonstration that Sinclair Broadcast Group sponsored in Baltimore which provided a side by side demonstration of digital television standards from Europe and the United States in Baltimore, Maryland. The company has been providing these demonstrations for broadcasters, various government offices and other interested parties for a few weeks. The European standard is COFDM (Coded Orthogonal Frequency Division Multiplex), while the U.S standard is 8VSB (8-bit vestigial side- band).
In its demonstration, the company emphasized the use of commercially available consumer-grade receivers and antennae. This is significant for one simple reason; the company believes that unless digital "over-the-air" technology is made extremely convenient for the average consumer to use, that adoption and usage rates of the digital spectrum may be impacted measurably. While we were limited to viewing the digital television test to two locations in one city, we believe that Sinclair's demonstration clearly shows that 8VSB, in its current state, may not be commercially viable.
(Anecdotally, other guests who had attended Sinclair's test in Baltimore found similar problems in tests they had conducted in Washington D.C. and Philadelphia, although we were not there for those tests personally.)
In its test, the company offered two demonstrations of the ability to receive digital television signals and conducted these tests in the city of Baltimore. One demonstration was done in a multi-level parking garage with a view of the company's broadcast tower (4 miles away) and another was conducted in an apartment in Baltimore Harbor that did not directly face the company's broadcast tower (5 miles away).
In tests using simple antennae and commercially available 8VSB and COFDM receivers, the COFDM technology performed much better than did 8VSB; 8VSB failed repeatedly while COFDM succeeded repeatedly. The COFDM signal delivered strong pictures, even when the antennae was rotated away from the direction of the broadcast television tower and even when the antennae was moving. On the other hand, unless it was relatively perfectly aligned, the 8VSB signal continually failed. Even when pointed at a broadcast tower (the test conducted in a parking garage), the introduction of a hand waving near the antennae was enough to distort the image. In the apartment setting, which did not face the tower, the 8VSB picture would not come in at all.
8VSB Receivers Do Not Receive Much.Yet? We believe that Sinclair's focus and concern for the 8VSB digital television standard centers on the consumer "operability" aspects of the emerging technology; they (and we) believe that "ease of use" is the key to consumers.
While fully supportive and appreciative of all of those that have helped broadcasters secure a piece of digital spectrum, most notably the Advance Television System Committee, which was ably led by Richard Wiley (former Federal Communications Commission Chairman) and the National Association of Broadcasters, we believe that Sinclair wants to ensure that broadcasters will be able to take full advantage of the digital spectrum's potential.
When the viability of the 8VSB signal was originally tested five years ago, we believe that tests measuring the ability to receive an 8VSB signal were conducted using a 30 foot high antennae and COFDM technology was in its infancy.
Ultimately, we do not believe that the average consumer would be willing to pay for and install a large antennae to receive 8VSB signals. (This obviously would be complicated further in cities, where antennae could not easily be installed).
Additionally, if the various television towers are not centrally located, the current 8VSB receiving equipment would require the antennae to be rotated so that it is pointed directly at the tower. This again, is not convenient for the consumer.
Lastly, adoption rates would be very low for digital television due to the prohibitive cost of installing an antennae and purchasing a set-top box.
Ultimately, at this point, the industry does not know whether the standard works well or not. Right now, we do know that commercially available 8VSB receivers do not seem to work well, at least in Baltimore. However, it should be noted that 8VSB receivers are essentially in their first generation of development. Newer generations could vastly improve the performance of these 8VSB receivers and eliminate much of the technology's current weaknesses. However, if better 8VSB receivers can not be made, then a review of the standard itself may be necessary.
Multipath is 8VSB's Problem. In traditional analog television, deterioration of the picture can result in an increased level of audio and visual "noise" which will eventually and gradually lead to the failure of the picture. In a sense, an analog television system fails gracefully. However, when a digital system fails, it can fail suddenly and entirely; pictures suddenly fade to black.
In digital television, pixels that comprise the picture are converted to binary numbers and transmitted in sequences. If a certain number of errors occur between the transmission and reception of these binary numbers, a typical receiver may not be able to "reconstruct" the various sequence of bytes, which leads to complete failure of the system. The reality that the system can fail implies that signals must be reliable at higher rates than that of analog television.
In the test of 8VSB that we viewed, we believe that its higher failure rate was caused by multipath propagation problems. When a signal reaches a receiver from many different directions after being transmitted, this is referred to as "multipath". Typically, the main part of the signal may reach a receiver by the most direct path, but the second signal could take a more indirect route after having reflected off a large object, building, water, etc. When the signals do not reach the transmitter simultaneously, this is referred to as "echoing".
In analog television, "echoes" would appear as ghosting; a faint second signal which creates the appearance that the main signal is being shadowed. However, in a digital signal, multipath can prove to be fatal; 8VSB digital television is very sensitive to multipath. We believe that multipath is even more pronounced in cities. This higher current failure rate in cities (if Baltimore is an appropriate benchmark for a city) could become a concern for digital television because the Federal Communications Commission mandated that digital television be built out in the 30 largest U.S. markets initially. The vast majority of these markets are cities and could likely face multipath problems.
Although we do not profess to be engineers, we believe that multipath is not a power issue. In its test, Sinclair used a transmitter with 50 kilowatts of power, which is the minimum power allocation in the digital television world. Although some broadcasters will be allocated up to 1 megawatt of power (20 times the power of a 50 megawatt transmitter), we believe that multipath will not be measurably improved. The fundamental problem is that the signal takes multiple paths to the receiver and that the receiver is not able to "reassemble" the signal to make it viable for display.
Issue May Affect Smaller Broadcasters More than the Networks. We believe that the 8VSB issue may not have as much impact on the networks as it will on non- network owned broadcasters. Two networks, NBC and CBS, have already negotiated cable carriage agreements for digital offerings with cable operators. Meanwhile, we do not know of any non-network broadcasters who has secured such a deal. However, we do believe that affiliates of the networks that have struck deals may be able to "piggyback" to some extent from deals struck by parent networks.
Wireless Cable Signal Propagation was Changed to COFDM; Spectrum Valuations Expanded Significantly. Over the last several years, the wireless cable business, pioneered by companies such as CAI Wireless, CS Wireless Systems, Inc, Wireless One, etc., saw the valuation of their businesses reach all-time lows as the competition from direct broadcast satellites and the true "cable" multiple system operators became overwhelming. However, valuations in these stocks have had a remarkable comeback as of late as interest in wireless cable's spectrum drew investments from Sprint and MCI Worldcom.
The interesting follow-up question would be, why the sudden interest in the spectrum? Recently, we believe that the Federal Communications Commission added the propagation standard of COFDM to the wireless cable's 2 GHZ of spectrum. Since then, interest in this wireless spectrum has increased dramatically. We believe that much of the credit for "wireless cables'' sudden increase in valuation is tied to the change in propagation technology.
The appeal of wireless cable to MCI Worldcom and Sprint is COFDM's ability to work and the fact that COFDM is portable, mobile and reduces the need for line of sight propagation: the technology works effectively on the move.
In terms of mobility, we have also been told that 8VSB is not particularly mobile. In the Telecommunications Act of 1996 and the Rulemaking made by the Federal Communications Commission concerning digital television, it was determined that broadcasters would not be limited in what they could ultimately do with their digital spectrum. If a broadcaster chose to provide a subscription service, than the broadcaster would have to pay the U.S. Government 5% of that ventures' revenues as a fee. With the potential to provide high definition television, digital television (multicasting model), data, audio, etc. over its digital spectrum, we believe that broadcasters may want to have a technology that offers them the most options. Again, however, we have not really seen 8VSB's potential (the standard) or the promise of 8VSB's receivers.
Viability of Digital Signal and the Ability to Ultimately Profit from Digital May be On the Line. The ability for local broadcasters to participate robustly in the development of the digital spectrum is critical for a few reasons. While analog signals are difficult to compress, higher resolution (relative to analog) digital signals can be compressed to permit multi-channel delivery over the assigned 6 megahertz digital channel. Ultimately, we believe that broadcasters' 19.4 million bits per second bandwidth can provide multiple video, audio and data services to the marketplace, assuming again, that the digital standard is robust.
We believe the ability to create economic uses for the digital signal is crucial to the industry for a few reasons.
First, it is important that the industry have a way to create a "return on digital assets". With an estimated $5-20 billion in total capital expenditures for digital television conversion, the industry must try to create incremental revenues from video, audio and data sources in order to not only justify digital expenditures but also to ideally create a return from the development of these services. However, unless the signal is robust, the industry may have a fundamental problem in the adoption rate of digital television and ancillary services.
Second, it is important that the digital television signal support the development of new services. In our estimation, in both the Telecommunications Act of 1996 and a subsequent Rulemaking from the Commission on the subject, broadcasters are allowed to develop high definition television, digital television and any other services the broadcaster wishes to offer over its digital spectrum. (However, if the broadcaster develops any subscription based service, the broadcaster would owe 5% of gross revenues to the U.S. Government). Again, without a robust signal, the development of any video, audio or text businesses may be in jeopardy.
Third, with a significant reliance on advertising dollars, local television stations should find ways to create new revenue streams. Digital television's spectrum may create such an opportunity. The development of new services is especially critical to non-network owned broadcaster's economics. Faced with increasing tension between networks and their affiliates and the likelihood that network compensation will diminish, we believe that it is important for broadcasters to develop alternative revenue streams.
COFDM is Not Perfect Either. While COFDM's relative performance versus 8VSB was compelling in the limited demonstration, it should be noted that COFDM has a few shortcomings of its own.
First, technically, as the technology exists today, COFDM provides only 18.6 million bits per second of "throughput", approximately 4-5% less than that of 8VSB. This is a drawback to COFDM technology. However, most feel that the disparity in bit rates between the two will narrow or be reduced completely. Hopefully, the same will be true of the receivers.
Second, we believe that 8VSB technology was focused on delivering the signal to the ends of the Grade B contour in order to closely replicate the signal "reach" of the analog signal. We believe that COFDM, unless driven by significant levels of power, performs the best within a smaller geographic area. We believe that 8VSB focuses on the fringe (Grade B signal) while COFDM focuses on the more densely populated areas surrounding a television station.
Third, if COFDM requires more power to replicate the same signal contour than
does 8VSB, then a significant change in the table of digital channel allocations would occur, which could prove difficult, especially in the signaled-packed Northeast, Middle Atlantic, Southeast and Central U.S. Fourth, years of development of digital television was made with the 8VSB model in mind and thousands of television sets have already been purchased. (However, to an extent, if changes were to be made, maybe it should occur well before there is any serious level of penetration in the U.S.' 100 million television households. Penetration rates of digital television sets are still extremely low.)
The Government Should Also Want Digital Television to Succeed. As we have alluded to repeatedly, at the end of the day, we also believe that the U.S. Government has an interest in seeing the digital television standard work well from an economic standpoint. If a broadcaster decides to use part of the digital spectrum for subscription services, the Government would be entitled to collect 5% of the gross revenue of that service. A robust digital signal could generate current revenue for the Government. Additionally, we believe that the existing analog spectrum may come up for auction in 2003. The latest FCC Rulemaking will permit existing broadcasters to bid on their analog spectrum. If the local broadcaster is able to create a model that works within an analog/digital environment, then the broadcaster could be a logical participant in the auction process. To the degree the Government can increase the number of bidders for digital spectrum, theoretically, the ultimate proceeds to the Government should increase.
Companies Mentioned in This Report
Sinclair Broadcast Group - SBGI - $17 3/8
MCI Worldcom - WCOM - $86 3/16
Sprint PCS - PCS - $56 7/8
CBS Corp. - CBS - $44 5/8
General Electric Co. - GE - $114
First Call Corporation - all rights reserved. 617/345-2500
