Continuing corporate soap operas have dogged DVD since its inception with no letup in sight. As DVD-A, SACD, and DVD Super Audio (DAD) prepare to duke it out in the marketplace like any good daytime drama this one has no end in sight. Below is a discombobulated set of narratives and links on the subject of post-CD digital audio disks. Those hoping for a simple answer to the 'what's next' question will be disappointed as corporate greed once again wins out over an elegant open solution.

What to do for the transition phase?

The most frequently asked question today in the context to hi-fi digital audio is "should I buy now or wait?" Ultimately only you can answer that based on:

1. how badly you want to experience the current state of the art,
2. fear of buying into the 'wrong' format,
3. investment in current format(s),
4. best guess on when (if ever) the proposed formats will settle down into one or a few widely supported options.

I would not presume to tell you how to weigh the above for your situation. I will share my decision and the thinking behind it. After a few months of 'wait and see' I decided to go ahead and get a 24 bit 96k capable DVD player and 96/24 DVD Super Audio disks from Chesky and Classic Records to feed it. The points I weighted in making this decision are:

1. It's fun to be on the leading edge of the state of the art. Always wanted master tape quality in my home and here it is! I am tired of waiting and believe it will be 3-6 months before SACD & DVD-A are available and more like another year before decent software availability and 'sock it to the early adopter' pricing passes.
2. I can hardly call myself a digital audio specialist and not be doing R&D on the advanced sources.
3. The DVD Super Audio format will remain the most universal of DVD audio formats because it is based on the original DVD-Video specification. Thus as long as the WG4 maintains backwards compatibility my software investment is protected. DVD Super Audio format is an initiative of the high end industry and the disks contain no questionable 'watermarks' or hyped 'you're in the middle of the band' surround effects.
4. Current production Pioneer DVD players allow 96/24 data out on their S/PDIF outputs. This feature could be disabled in new models under pressure from the music and movie industries awaiting some mythical hack proof copy guard scheme. As an example below we document that Sony Consumer Electronics will NOT have digital outputs on their first SACD players all the while Sony Semiconductor is announcing the chips to enable such an output in the yet to be tested by audiophiles IEEE-1394 format. I want to buy now while this very desirable feature is available as I believe a likely scenario is a period of digital outputs dummied down to 48k 16 bits while Big Corp Inc. haggles over YAS.
5. The DVD player I bought to partake of 96/24 audio delights is also providing my family the stunning picture and convenience of movies on DVD. Once 'universal' DVD-Audio players are available supporting MLP, DVD-A, and SACD and their pricing comes down this player can retire to the family room to play movies.

DVD and the Future of Digital Hi Fi

Bill Buxton (his papers)

 The DVD FAQ contains information about DVD-Audio specification that is vital to placing this discussion in context. Yet another rival format is Sony/Philips Super Audio CD (Hi-Fi 98 Sony press conference report) which will join the fray in a battle royal for consumer's loyalties and cash. Already available, at least by mail order, from audiophile oriented record companies is DAD Digital Audio Disk (also known as AAD for Advanced Audio disk) which use the DVD-Video specification allowing 96/24 audio encoding to jump ahead of the majors and be first out. Of course for the super multinational conglomerates the looming format war is really about technology royalties, quite a good deal if your company can get them! Another interesting article.

As I watch the whole thing unfold it calls into question just what a future DAC will look and act like. Looking at DVD-Audio specification draft all I can think is WOW SO MANY OPTIONS! I am in favor of an open-ended standard with room for future growth, however it seems we are about to have a case of "be careful what you ask for, you might get it!" Alas, it seems the audio engineering and scientific communities were caught with our collective pants down when DVD technology popped on the market. In the few years prior to DVD most serious commercial and academic research had focused on data reduction schemes like MP3. These to deliver audio over limited bandwidth carriers such as the Internet, satellites, and with audio content playing the poor stepchild to video. With the advent of DVD the data capacity is suddenly there and truth be known no one has hard and fast evidence just how it should be used. Serious independent multi-million dollar a year research into such matters simply does not happen. At least not in an open forum around which a consensus could develop. And when did a hard total consensus ever form in audio? We are still debating tubes vs. transistors! So, lacking a consensus the standard setting group WG4 (Working Group 4) is basically allowing any and all formats that hold hope for a performance or market function!

Quoting from DVD-Audio specification draft: "LPCM is mandatory, with up to 6 channels at sample rates of 48/96/192 kHz (also 44.1/88.2/176.4 kHz) and sample sizes of 16/20/24 bits. This allows theoretical frequency response of up to 96 kHz and dynamic range of up to 144 dB. Multichannel PCM will be downmixable by the player, although at 192 and 176.4 kHz only two channels are available. The maximum data rate is 9.6 Mbps. The WG4 decided to include lossless compression (it's about time!), and on August 5, 1998 approved Meridian's MLP (Meridian Lossless Packing) scheme, already licensed by Dolby. MLP will allow playing times of about 74 to 135 minutes of 6-channel 96kHz/24-bit audio on a single layer (compared to 45 minutes without packing). Two-channel 192kHz/24-bit playing times will be about 120 to 140 minutes (compared to 67 minutes without packing)."

As a small high end audio DAC manufacturer I am following the DVD soap opera as closely as possible. I believe it will play out following one of the following scenarios:

1. A format war over the audio only version of a DVD disk that will not be as bad as Beta vs. VHS. IT WILL BE WORSE. Maybe if we go back 100 years to the Edison cylinder vs. flat records or DC vs. AC power grids we can find a battle royal as bad as the one brewing. Consumers and retailers are about to be presented with several different overlapping answers to a variety of questions they have yet to ask. None of the elements (one worldwide standard, a clear differentiation over previous disk) which made CD such a huge success in the '80s are present. Rather the looming battle resembles the marketing fiasco of quadraphonics in the '70s. Remember SQ, QS, CD-4, and UD-4? Now its DVD-Audio, Super Audio CD, DAD, and AAD!

2. Attempts to replace the CD die off after the mass market ignores all proposed Super CD formats as answers to questions it is not asking. As a videophile and movie lover I eagerly watched S-VHS and ED-Beta come to market. The mass market did not care and kept buying fuzzy $200 VHS machines. As an amateur recordist I moved to DAT early. The mass market did not care and kept buying hissy $200 cassette machines. As a music lover and audiophile I eagerly await digital music formats with greater than CD qualities. As a hi-fi gear designer specializing in digital audio I can hardly wait for 96/24, 192/24, or SACD. However I am realistic that our tiny niche of High End Audio has always ridden on the coattails of the mass market. Without records our gear is worthless and without a mass market stuffing the pipelines full of a diverse range of all the world's music we are left with a handful of 'who are these guys' jazz and 'I didn't know Topeka has a orchestra' classical disks.

3. Third & fourth generation DVD players will support the whole alphabet soup of formats DVD 96/24, DVD-A, SACD, CD, CD-R, CD-RW, DVD-R, and DVD-RW. Early adopters will have to carefully setup their systems to match the disk currently playing. Someday 'smart' components with two-way links like Firewire could configure themselves based on the features of a given disk vs. hardware available options and owner preferences of the system trying to play it. Until that rosy day arrives keep the manuals handy.

Ultimately of course market forces will narrow down the many format offerings to a subset which may actually become widely available. With the Buxton quote above in mind I believe the ultimate winner will be the format and realization that simplify too complex lives. And DACs will get smarter and more adept at automatically configuring themselves on the fly. What I wonder is how many of you want to be the first on the block to hear 96/24 knowing full well being on the cutting edge often involves getting cut?

(**Just a thought. With DVD changers already available some guy will soon be able to load one up with a DVD-video disk in slot one, a DSD music disk in slot 2, a CD in slot 3, a DAD in slot 4, and a DVD-Audio disk in slot 5. Then hit "Random Play" and leave to pour drinks for his party. Now how does my DAC handle that one!?!? Interesting. Like the old Chinese curse: "May you live in interesting times"!)

So to me it seems future DVD-Audio players and DVD-Audio compatible DACs are going to have to be very much smarter than present. OR The market will use just some small subset of the above options, like the 96/24 DAD disks presently available. Thus another variable becomes will our 2 channel hi-fi rigs 'converge' (please pardon the buzz word) with 5.1 channel home theatre and hi-fi DACs morph into multi-channel decoder-DACs OR will we chose to keep our two channel and multi-channel listening segregated?

What I am getting at is:

• yes, for audiophiles and experimenters as serious as we are risking early adoption of the presently available 96k/24bit disks/players is justified,
• no, I would not promise consumers ACG can deliver a future proof DAC today. Certainly not with just one filter chip socketed as some companies are!

 One can make a case that if DVD players are smart enough to take whatever audio data in whatever format and output a 2 channel S/PDIF stream at 44.1/16 or 96/24 (and at least one already is) a DAC upgraded from our present architectures will work. Concerns abound:

• What happens when the next disk has 5.1 dolby and the player is outputting 6 channels of decompressed audio on the same SPDIF port? Opps, you forgot to program the player to send that to the home theatre decoder? Well tell Joe Consumer to program that player right, just as soon as he learns how to set the clock on his VCR! :-/
• How available will 96/24 capable data sources and software become? Currently Pioneer has a few DVD players capable of 96/24 output on their S/PDIF port. Many DVD manufacturers including Toshiba and Sony have chosen to limit the S/PDIF output to 44.1/16 in deference to record industry paranoia over software pirates. When/if audio Super CDs become available to WG4 &/or SACD standards the current 24/96 players will break on the disks if they are encoded with MLP or DSD.
• Two channels or more? Currently available 96/24 software is largely 2 channel stereo from small specialist quality oriented record companies. Activities from the major labels indicate when/if they get involved in 96/24 they will likely favor multi-channel surround format like 5.1. It is easy to see why their bottom line mentality would favor the easy to explain surround format over the subtle improvements of 96/24 over 44/16 when your typical customer listens on a BooM BoX or rack system.

Lucky ACG does not have customers who are clueless. Our customers, who we prefer to think of as Members of The Audio Crafters Guild, are smart enough to know that for the early adopter it is never 'plug and play'. And the early adopters typically have the most fun because they are involved at a higher level than the poor box opener who meekly waits for Multinational Big Corp to tell him what to buy next. Of course early adopters often end up with collections of orphaned, or shall we say unique, hardware and software! To minimize that ACG's strategy is to capitalize on the upgrade-ability and expandable nature of X-DAC 3.0. Also it is obvious that no matter the market dictates the final super CD format will be the basics of good design will continue to underpin devices of quality. Designs aspiring to break the 20 bit barrier need the cleanest power and ground systems possible. Digital input receivers must accept at least 96k data rates. The DAC subsystem used must not only run at 96k or better rates hopefully its resolution will better the 44k units, something we are finding is not a given. Thus we have been in a mode of evaluating integrated circuits as they come to market while continuing to upgrade our basic subsystems. Upgrades which yield benefits both today for the 44/16 format and tomorrow for super CD whatever its form may take.

News April 1999 It has been reported by EE Times (see story) that Sony/Philips have dropped plans for SACD disks to be backwards compatible with 44/16 CD players.

EE times also reports: "To guard against unauthorized copying, SACD employs both invisible and visible watermarking on the disk, and content is encrypted before recording. The SACD players will have no digital output for the SACD signal. Sony is positioning the player at the ultra-high end of the consumer-audio market. The player will debut in Japan at $4,170. Overseas shipments are expected to begin by the fall."

A couple of things strike me in this story. First no digital outputs for SACD. Looks like the small and medium sized companies which established the High End audio industry are not invited to this party. With Sony going after "the ultra-high end of the consumer-audio market" one has to wonder when even the heavyweight Krell and Levinson class companies will be granted access to SACD technology. Once Sony skims the cream of the early adopters off the top for itself one would expect that if SACD gains market acceptance that same market will eventually force an opening up of the technology. So for the long term the second striking aspect of this story could be even more important which is watermarking. Now we are already seeing the typical pre-introduction frothing at the mouth preview pieces relating the wondrous sound from DSD and by association SACD (like here and here). And I have no doubt these are accurate representations of what these individuals heard. But is what they heard what will go on sale for public consumption this fall? Specifically what does SACD sound like after being watermarked? Here in Oklahoma we watermark our cattle, its called 'branding', and it leaves a scar. One can be certain the companies selling the watermarking technology looking forward to a steady stream of technology license fee revenue will swear you will never hear the watermark tones mixed in with the music. (To learn more about watermarking see the web sites for ARIS, Alpha Tec LTD., and Signum. Also notes from an AES workshop on watermarking audio.) Alas the breathless reports coming from private CES demos and factory tours make no mention of the source being watermarked or not. In several cases they do mention the source is a digital audio workstation hard drive. In this context listening to a hard drive master copy DSD data stream and implying a someday watermarked mass produced SACD will be as good is like listening to a 30 i.p.s. 1/2 inch analog master tape and assuming an LP cut from it has to sound as good!

News May 1999 It seems we may be going from bad to worse. As reported in the very fun online magazine '21st Convergence' (read article) in the name of 'convergence' trouble is brewing. All the major consumer electronic and computer companies are vying to establish their visions of a future networked, plug and played, interactive, all singing all dancing home replete with multifunction PC/TV/Stereo/Phone/toaster. As a computing professional who makes an excellent living spending hours getting Microsoft's Plug and Pray connected devices to work in their Windows environments this is not a vision I would encourage you to latch onto. The format and 'standards' wars over whose OS and network runs Joe Sixpack's new DTV should make the DVD format SNAFUs look like a tiny border skirmish by comparison. Or as a good friend commented "the day I have to re-boot my stereo to hear Beethoven's 9th is the day I pitch it all out the door and go back to LPs!" The KISS principle (Keep It Simple, Stupid!) and Buxton's advice at the top of this page will not likely be a part of this brave new world.

News June 1999 Sony Semiconductor division has announced IEEE1394 LSI incorporating 5C digital content protection standard in the form of the CXD3200 family. Some details here. To me their choice of animated GIF speaks volumes of the future they have in mind for us.