Your Time Warner Congress & DVD

Orwell's Dream Team Hits The Field

Francis Vale


A Cultural Clash

The avowed goal of the audiophile high end is to render a 'perfect' sonic reproduction in the home of what was originally recorded, whether it be it on CD, vinyl, or tape. In this rarefied quest, cost is often considered to be no object. The high end is thus at the opposite end of mass market audio gear. But like any phenomenon with grossly exaggerated features, the high end also helps make clearly visible all the nasty cracks and fissures within the consumer electronics industry.

There has been an ongoing debate -- bordering on a cultural war -- between many in the audio high end about the relative merits of digital vs. analog. Many in the latter camp are convinced that vinyl records have yet to be surpassed in their ability to provide a life-like facsimile. On the other hand, digital sources, like CDs, are often characterized as being 'lifeless' and 'dry,' and causing 'listener fatigue.' Thus, neatly encapsulated in the high end, we have a societal rearguard clash, as epitomized by one disaffected group's sometimes searing critiques of all things digital.

To overcome these supposed digital audio limitations, a fairly large cottage industry has sprung up in the high end. Its sole purpose is to make digital sound as good as those ancient, rotating bč.Zes noire. To reach that end, these high end vendors create custom designed, proprietary technology gadgets, costing anywhere from several hundred dollars to more than $10,000. Some of their products are complete CD playback systems. But many other high end approaches split the CD player (the transport) and digital processing functions (the digital analog converter) into two separate components. But regardless of all the hard effort and expense that goes into creating these high end devices, the technology for processing the bits into audio analog waveforms has essentially been stagnant for more than a dozen years; since the inception of the CD format.

The Culprit Caught

The real culprit behind this glacial rate of techno-change is the entertainment industry, and not the consumer electronics manufacturers. The former exerts a vice-like grip over its content. And as a class, the entertainment companies are deathly afraid of the new digital technologies. Moreover, any consumer device with digital recording capabilities that can produce studio master quality duplicates is immediately pounced upon by the entertainment industry, which sees to it that the new device is functionally and/or legally crippled.

To the bootleg paranoid entertainment execs, these new digital technologies only spell trouble. Somehow, they must envision every mom, pop, and kid in America, all slaving away in their basements, making millions of illicit copies of Guns & Roses, Pearl Jam, etc. An unfortunate effect of this industry paranoia was to keep audio DAT drives out of the U.S. for a prolonged period of time. The same extended fate seems to be happening with the new digital VCRs, already on sale in Japan. The entertainment people went on a similar fevered rampage against our current VCRs .(Remember the industry's spokesman Jack Valenti, and his VCR tirades?). The movie moguls were convinced that VCRs meant the end of their Olympic swimming pools and Mercedes roadsters. But in a complete contradiction, the home video market has become as financially important to the studios as the large silver screen. (In 1994 alone, 497 million prerecorded video cassettes were sold, and 3.5 billion rental transaction occurred.)

Nonetheless, the industry's collective paranoia has continued to overrule objective reality. As a consequence, we have audio's infamous Serial Copy Management System, or SCMS (pronounced, appropriately enough, 'scams'). With SCMS, you cannot make multiple backups of a digital copy. Thus, second generation duplicates is as far as SCMS allows you to go. So, if your digital copy made from a master source starts to rot, deteriorate, or is otherwise damaged, and/or the master is lost, you are out of luck. The computer software industry effectively banished such brain dead copy protection schemes years ago. But the zombified SCMS technology is still the entertainment's industry's preferred reaction to home audio DATs, CD-R drives, and likely, to the new digital VCRs; all of which have the ability to make excellent second, third, fourth, etc., generation copies off the same master.

What is fascinating to behold is the legal hocus pocus that has produced such a procrustean digital bed. For example, if a digital audio recording device is labeled as being for the 'consumer' market, then it legally -- as mandated by your U.S. Congress -- must have SCMS. This is a draconian piece of legislation, make no mistake. The Congress had decided that the fine for violating a SCMS system should be a cool $1 million. Who knows, if the entertainment industry truly had its way, they would probably have Federal ATF storm troopers breaking down suspect consumer's doors.

If, however, the very same consumer-labeled device is intended for business computer use, then it avoids the dreaded SCMS. Audio consumers thus get stuck with scammed DAT drives, CD-R's, etc., while all those with 'business' computers have no such restrictions. Same device, same functionality, same technology; the only difference is that the consumer version is sold under the imprimatur of paranoids.

This diseased digital attitude is also manifesting itself in how the entertainment industry is looking at copyright payments. The industry collects a royalty every time content is transmitted, such as when played over a radio station, or is piped into a restaurant. However, if the content is instead played back from a CD, like in your home, the industry collects another type of royalty. Now, what to do when the content is first transmitted in digital format, say via the Internet, and is then played back at home via your multimedia PC? Well, let's royally bang 'em twice, says they! Many in the industry think they are now entitled to collect two concurrent royalties: one royalty for transmission while the digital content is en route to you, and another royalty when played back after arrival in your home. Fortunately, early indications are that the courts seem to be showing more common sense in this matter than the industry's legal eagles (vultures?).

The consumer electronics vendors' desires for making zillions of cheap cookie cutter products and the entertainment industry's fears have thus become neurotically conjoined. The unfortunate result of this pathological pairing has effectively stymied innovation in digital consumer electronics. And into this manipulative morass has stepped the high end equipment manufacturers, who, with their hand-crafted, custom designed, assembler coded computer devices, have attempted to transform all these digital sow ears into your silken listening pleasure.

In many ways, the high end vendors' livelihood rests -- indeed, relies -- on such an ongoing stream of kludged formats and devices. For if CD was indeed Perfect Sound Forever, who would need these specialty vendors? This an unholy techno-trinity, if ever there was one. Ironically, the high end not only helps construct these audio Potemkin Villages, it simultaneously exposes their phony facades.

Enter DVD Dementia

But lo, a new beast has entered the digital playback scene: DVD. This topic was covered at length in other Impact articles, but to briefly reprise: Digital Versatile Disc is an amalgamation of two competing camps' technologies: Sony/Philips v. Toshiba/Warner. In the end, IBM played the role of enforcer/broker and got both sides to agree on a common format. DVD holds forth the promise of entirely new ways to distribute movies, computer-ROM multimedia content, and lastly, audio-only playback.

On the DVD-video and DVD-ROM side, both MPEG-2 video compression and Dolby AC-3 multi-channel surround sound were selected as the enabling multimedia technologies. Despite what you may have heard, DVD MPEG-2 picture quality is only about as good as current laser discs. So why switch playback horses? The answer is that you can cram a lot more compressed digital content onto DVD than a laser disc. The other reason, of course, is marketing. Better to have you chuck those big old laser disc platters, and buy a new 5" DVD player.

In addition, Dolby Pro Logic, which synthesizes multichannel surround sound from two specially encoded stereo channels, has been passed over for AC-3 from Dolby Labs, a fully discrete five channel system. AC-3 is a standard feature (in the U.S. market) for both DVD movies and DVD ROMs. But from the computer vendors' point of view, AC-3 on DVD ROMs has its own problems. AC-3 processing sucks up computer cycles like a strung out alcoholic. On average, AC-3 will consume from 35 to 45 DSP MIPS to process this Dolby-ized audio. So where does the extra processing horsepower come from on your PC? From your already OS/application strained host CPU? Or do you add in special multimedia functions to the host CPU chip, as Intel is now in the process of doing? Which means, of course, adios to your existing PC. Or do you add a DSP on the motherboard? Say good-bye once again. Or do you add-in a special card? Any and all of these AC-3 DVD ROM processing options add cost and complexity to your PC.

In the cut-throat PC business, where the margin between profit and loss is razor thin, such extra costs and complexity only mean fiscal trouble. Moreover, the current sole source for Dolby AC-3 decoding chips is a company called Zoran. Although Dolby Labs is making supposed moves to ease this sourcing choke point, Dolby is notoriously slow for taking its time about certification. This painstaking approach to quality control is consumer-laudable, but it is abhorrent from the time-to-market-driven PC vendors. In addition, if there is anything the computer industry cannot stand, apart from a single source, is a single standard. The computer industry manipulates standards for selfish market positioning the way World Champion Wrestling pretzelizes arms and legs. Add it all up, and the prospect of DVD-ROMs with AC-3 are so much PC molto agida.

Finally, for the audio-only application of DVD, the high end industry sees all that new gigabit beach front, and dreams of a new generation of 'super' CD's that might yet achieve perfect sound; if not forever, then at least for a couple of years. Fomenting this rush to the Bit Bastille was "A Proposal for the High Quality Audio Application of High Density CD Carriers." This DVD proposal was fathered by the Acoustic Renaissance for Audio (ARA). Among other things, the text of the original ARA proposal called for sampling rates up to 96kHHz with (minimum) 14 bit precision, or 20 bit precision at 48kHz; eight full bandwidth channels of surround sound; and perhaps most critically, that a lossless compression scheme be used.

The primary founders of ARA included: Tony Griffiths, Technical Director, Decca Recording; Professor Malcom Hawksford, University of Essex; David Meares, R&D Manager, BBC Radio Research & Development; and Bob Stuart, Chairman and technical director, Meridian Audio Visual, Ltd. Acting as ARA group advisors are: Peter Craven, Consultant, Michael Gerzon, Consultant, Hirokazu Negishsi, Director D&D Center, Canon, Francis Rumsey, University of Guilford, and Chris Travis, of Division, Ltd., and some senior staff from Pioneer. The inclusion of this last group was strategically important from ARA's point of view.

Indeed, Pioneer has recently come forward and proposed an audio-only, 'Super DVD' playback system very much like that called for by the original ARA agenda. It would sample the sound at 96,000 bits/second, and each sample would be represented by a 24 bit number. This Pioneer scheme would extend the dynamic range to 144dB. Naturally, there is a competitive super DVD proposal on the table; and also quite naturally, it comes from Sony, who is arguing for a sampling rate of 2.8 million bits/sec; but with each sample represented by a single bit. This audio format would offer, says Sony, much greater processing flexibility; but on the downside, it would also reduce the amount of playback time. The industry cynics say this alternative scheme is merely a stalking horse for Sony, who makes much of the audio recording equipment. If the Sony format were adopted en masse, then the recording studios would, like PC users, have to chuck their equipment to accommodate this new type of audio-only DVD.

And finally, there are the entertainment companies. Their take on this new Super Audio DVD is not surprising: They are terrified. Change makes these technology-adverse companies very nervous. They like their CD cash cow just the way it is, thank you. A new format introduces questions, such as consumer acceptance, and perhaps most important, new copyright issues. Once you begin processing the bits in a new way, does that mean recording artists/lawyers will demand new kinds of DVD audio royalties? Most assuredly. In short, given a choice between 'don't rock the boat' and better audio playback, the entertainment companies will Valenti-ize (as in demonize) the new Super Audio DVD like it did the VCR. Why jeopardize a good thing, especially when playback time will be nearly identical between CD and Super Audio DVD? After all, the present CD = Perfect Sound Forever, right?

Why You need a New kind of Computer

Of course, there is an easy solution to all of this format nonsense -- Forget about closed, proprietary audio hardware (in both consumer electronics and the high end), and follow the tremendously successful 'open' PC model. PC users have found digital freedom via several synergistic components: a) Faster, cheaper, standardized microprocessors, executing b) better, smarter, software-working on c) format-liberated content. What digital audio users need is: 1) A standardized system box; 2) sporting the new 32 bit, 33 MHz, 132 megabyte/sec PCI local bus now being adopted by PC manufacturers; into which PCI bus 3) could be added standardized, plug and play system cards offering specialized or enhanced functions; 4) all administered by a blazingly fast host processor; and 5) integrated with digital content sources via 'FireWire', the new IEEE P1394 consumer electronics/computer standard. This high speed serial interface provides isochronous service and guarantees latency for audio, imaging, video, and other streaming data.

With an audio PC, you could forget about which Super DVD format is better, and instead try to make sure that any such new scheme be completely open to all types of digital processing. E.g., PC users routinely transform multimedia data files from one format to another. If such an 'open audio' approach was to win the day, then the focus would be on better audio processing algorithms, sold on plug and chug software; and not on spurious and distracting format wars. Of course, open formats mean much greater user accessibility to content; which in turn begets the entertainment's industry's ire and paranoia. A mobius marketing strip, if ever there was one.

One chip set that can form the basis for a home entertainment multimedia PC is from Philips Semiconductor, and goes under the name of TriMedia. The TriMedia chip can process several multimedia types concurrently, at the rate of five operations per individual instruction. Philips' new chip is a hybrid device, combining elements of both a DSP and a regular computer CPU. The TriMedia has been designed to operate as either a multimedia co-processor in a PC; or standalone, as an integrated DSP/CPU in a consumer electronics device. In a nutshell, the TriMedia combines the multimedia power of a next generation DSP with the high level programmability of a regular CPU.

The TriMedia directly interfaces into the PCI local bus, as well into a digital camera, video encoder, stereo A-D/D-A converters, and a V.34 analog modem or ISDN digital interface for telephony. Its 'glueless' interface means no extra circuitry or support chips are necessary. This feature greatly cuts down on system cost and complexity. At its heart is a 400 MBps internal bus that links together autonomous modules.

A single TriMedia instruction can do several things at once, like opening up data paths between main memory and to any of its modules. These multimedia modules provide such functions as digital video, audio, MPEG, and image processing. A 120-MHz Intel Pentium cruises along at 200 million operations per second (MOPS); while the 100MHz TriMedia screams at up to 4 billion operations per second (BOPS). Comparing DSP processing power with regular CPUs is always a case of apple and oranges, but generally speaking, one DSP operation is equivalent to three CPU operations. In other words, a CPU needs to run 3x faster to compete with a DSP device at a given performance rating. Significantly, these TriMedia processing speeds are being quoted by Philips for C code. In fact, all performance optimization is done in C. Unlike prior generations of DSPs, no assembly or machine language is required with the TriMedia. Finally, the projected price for the TriMedia chip is less than $50 in quantity.

Your Time Warner Congress & Myopic Money

Be's exciting efforts notwithstanding, there is little question that the home entertainment 'consumer computers' of the future will be made and marketed by Philips, Samsung, Pioneer, Sony, and others of their consumer electronics ilk, and will utilize the already familiar TV as their user interface. These new consumer systems are the next generation of 'TV mainframes.' The multimedia PCs that use these same technologies, like Be, will go into a different market segment: the Congressionally-defined computer market. Apart from the U.S. Congress trying to decide, Solomon-like, what turns an identical device into two separate and distinct products, there are also economic forces at work which help perpetuate this increasingly artificial market division.

Just because the Intel Pentium costs many times more than the TriMedia, don't think for a minute that this has anything to do with its complexity or fabrication costs. Intel makes a bundle of money because its chips are so cheap to make. All things being equal, once you have the fabrication process down pat, and all the quality control issues worked out, modern day CPUs or DSPs cost just a few bucks to make.

So how can Intel get away with charging hundreds of dollars for one of its PC CPUs? It's because Intel sells into a market -- the PC market -- where complete desktop systems typically retail from a low of $999, to a high $9,999, a 10X price point spread. Consumer electronics, on the other hand, are moved through a low margin, high volume distribution channel that usually has retail price points of between $50 and $500. To play in the consumer game, products have to be cheap, and closely track a radically altered10X pricing spread.

The PC market is thus a completely different ballgame than consumer electronics. In the PC world, you have to offer perceived (as in costly) functionality if you want to stay alive, never mind win. 'Value add' in PCs typically translates into higher retail price points. And the distribution channel for PCs, like the PC vendors, desperately needs those higher price points -- and higher margins -- if they are to survive. (Apple's recent problems are a case in point. It sold lots of low margin product, but failed to cut its high overhead. But even if it drastically cuts its costs, Apple, like all the other PC makers, cannot profitably survive a steady stream of heavy margin hits.) Therefore, the PC market enthusiastically supports higher margins, which in turn gives Intel its own margin fix, which in turn keeps PC prices up. Does all this sound familiar? It should. In many ways, the PC market mirrors the audio high end: expensive boxes, artificially expensive core technology.

This is why Philips, a classic consumer electronics company, offers TriMedia chips for fifty bucks in quantity -- It has to. Intel, on the other hand, can sell its top of the line PC CPUs for $700, and the day after a new generation is announced, boom! The old chip's price suddenly falls like a stone, and the new one costs $700 again. And Intel thereby stays consistently profitable, as does (hopefully) its distribution channel partners.

[An Aside: The current technology focus of making the intelligent TV set-top box a 'PC' running some type of MS Windows variant, (e.g., Microsoft's newly announced S.I.P.C.) or other such traditional PC operating system, is dead wrong. Somewhere along the Digital Convergence way, HDTV/Interactive/Intelligent TV have become confused with PCs. Overzealous hypermarketing by the computer industry has only helped create this delusion. The TV, be it Interactive, HDTV, or otherwise, is not a PC, and never will be. It is a consumer appliance. The tremendous upsurge in home PCs does not mean that they are about to take over the TV; no more than the latter took over radios. Both TVs and radios now comfortably coexist. The same easy association will likely happen with PCs and 'intelligent' TVs. Just as is done today with a PC, the consumer wants the option to radically modify or significantly enhance their electronic product's functionality, and without having to throw out the whole device. However, the consumer also wants these functional changes to be excruciatingly simple to accomplish. The goal should therefore be to make the task of performing such complex machine modifications 'invisible'. -- A highly sophisticated set of functions, ones that can dynamically be regrouped and merged together, but posing as a stone simple set of options, is what the consumer really desires. -- Thus, the logic implicit in most advanced Windows-style TV set-top boxes is flawed because it assumes that all Consumers are PC Users, and that all PC Users are Consumers. Not so. Consumers are not Users. When the average consumer goes shopping for a TV or a DVD player at Sears or JC Penny, it is very unlikely that a complex PC system like Windows, stripped down or otherwise, is what they have in mind. Until this basic truism is found out, these TV set-top vendors and their erstwhile suppliers will travel down a very rocky road.]

Orwell's Dream Team?

In sum, if a TV-oriented home entertainment box is to succeed as a consumer level product, it has to sell for $500 or less, and offer an entirely new user paradigm. Philips, as evidenced by the TriMedia's $50 price point, is already geared up for this market. Moreover, companies like Philips have been playing the consumer game for years. But even if Intel et al succeeded in this new consumer arena, would these marketing re-christened PCs suddenly be branded by Congress as 'consumer computers'? And if artificially-contrived market forces win the day (as is likely) then so does the entertainment industry. Which means, hello SCMS, and who knows what new types of brain dead schemes the suits will dream up on those warm Hollywood nights; which is a nightmare, indeed. The DVD Animal Farm obviously has a very bizarre bunch of unlikely stable mates. Even Orwell would have been envious of DVD's convoluted politics.

Regardless of the techno-business-vested interest politics and the financial fate of the PC market, what still matters most is that these powerful new TV mainframes could radically shift the audio emphasis away from fulminating formats and hand-crafted, high end hardware, and over to smart software applications. Both the high end and consumer audio future belongs to a new generation of software-only companies who are clever enough to leverage their talents and knowledge off these cheap consumer computers. And once the software fox is in the audio roost, vendor feathers will surely fly. The entertainment industry chicken coop could get mighty empty when that happens. Whose feathers will be the first to get plucked?

Copyright 1996, Francis Vale, All Rights Reserved

21st, The VXM Network,