A recent post by Dr. AIX, our favorite Hi-Res Audio Evangelizer, has motivated this post. To see where this all came from, go here, then here, then here. Then, please come back. I thought that perhaps blogging in numbers might get some industry attention. Ok, I’m not that naive, but I humbly submit this, what some may term a diatribe, by way of support. Can there be a humble diatribe? Ummm….
Years ago I learned a principle of marketing that stated simply that for a new product to succeed and rapidly penetrate the market it must offer at least a 3-fold perceived improvement (and preferably 5-fold) over the product it replaces. If there is no predecessor, the job is easy. But if there is an existing product, it gets down to what the new product offers of value over the old, and at what cost.
When the CD was introduced it offered a list of “improvements” like smaller size (only a partial plus, it’s a minus for art work), longer play time, more resistance to wear, easily and quickly accessed tracks, and of course, “perfect sound forever”, thank you. The price point was initially 1.5X – 2X of vinyl records, with players entering the market > $1000, dropping to half that in about a year. The CD achieved market penetration faster than anticipated, eclipsed and replaced several existing analog systems, and with a few exceptions on the edges of the bell curve, was well received in the general market.
Now comes Hi-Res audio. What is the predecessor? The biggie now is AAC and .MP3 on-line purchase files, as CD sales have collapsed. What does HR offer? Well, higher than CD quality is the premise, but it often doesn’t achieve that because of how HR is defined. We’ll come back to that little bugaboo. It certainly should beat most flavors of .MP3, and likely most AAC as well, even if the HR file doesn’t beat a CD. But what else? Where’s the 3-fold improvement?
Is it more convenient than current on-line purchases? No, but it’s about the same as CD when buying physical media HR audio, and the process of on-line purchase of file downloads is a little more trouble because, we’ll it ain’t iTunes. Most of the lack of convenience comes during playback.
Is playback at least as easy? No, for several reasons, mostly it doesn’t play on portable devices…at least not the most popular ones, though that could change. It seems like it’s not on Apple’s close-range radar, and that’s the big share of on-line music purchases. iTunes won’t play HR natively at least not all the way to the output jack, and the fixes to make it work also imply HR hardware too. So no, playback actually is more difficult.
Will it take and investment in hardware and possibly software to handle the files? Mostly yes, though with physical media we do have disc players that can handle it. Disc players. Yes, I remember those. Been a while since I actually put a music disc in a player…you?
How about the immediate gratification of an online purchase? HR is by nature larger and therefore slower to download. The wait isn’t interminable, but probably taxes today’s youthful short attention spans.
So as you can see, the 3-fold perceived improvement is really down to sound quality. And that means, if there is any hope of HR Audio’s success in the marketplace, that quality has to be carefully defined so that it is easily and reliably heard and so becomes the key improvement. And a promise of that quality should be easily discerned by the consumer, then his expectations should be reliably fulfilled. Am I using “reliably” three dimes in the same ‘graph? Yup, I am! The quality improvement is something that must be relied upon, or we don’t have a product.
The JAS Hi-Res Audio logo is apparently an attempt at such an indicator. Yet the qualities that stand behind that seal have not been clearly defined. Basically, we’re only dealing with two things: sampling rate and bit depth, and their results on audio quality. Yet, while the JAS/Sony spec does deal with rate and depth, it doesn’t adequately specify the results of either, and in doing so opens the qualification of HR Audio to just about any uncompressed file sampled at 96KHz or higher, regardless of the origination of the actual audio material, and without regard to real bit depth (as apposed to just a 24 bit word).
Not only does this sort of thing accomplish nothing in terms of advancing HR audio in the market, it actually goes the other way. It in effect ensures that consumers expectations of improved audio will routinely be unfulfilled, the logo will eventually have no meaning, and the additional cost and trouble of the format will assure its failure.
What’s needed is a clear differentiation, an “always-gotta-be” improvement in quality. If we’re writing specifications that permit recordings qualify for a Good Housekeeping Seal of Approval, then those specs had better actually do that, or the seal is worse than meaningless, it becomes a misleading mockery. And, if anybody who can generate a 24/96 file qualifies for that seal, regardless of actual audio quality, we haven’t done any good either.
When we specify a frequency response, there are two parameters: a frequency range (20-40KHz) and a level variation tolerance (+/- 3dB). Take the last parameter away (lake the JAS specs) and you don’t have a frequency response because you haven’t defined level. Any little bit of 40KHz, even a signal 50dB down, could meet their so-called specs. I would advocate complete specifications of any parameter.
So not only do we need to work on specs for frequency response and noise, we need to put more than a little emphasis on dynamic range. A 24/96 file that has been deliberately peak-limited, clipped and crunched, bent, spindled and mutilated is still no better than the same recording in 128Kbps AAC. In fact, a popular independent artist recently released some of his catalog in “High quality audio” files, and no doubt sold a few. Of course, they were identical in every audible way to the original not-so-high-quality files, so what was the point?
The advantage of HR audio has to be clear, easily heard, and worth the trouble and cost. It literally must be a Gold Standard against which every other form of audio can be compared. Recalling also that “content is king”, the HR material must be very main-stream. If listeners can hear their favorite artists in HR, wide bandwidth, ultrasonically extended, and dynamically un-processed form, they’ll soon despise the other forms, and we could have a positive trend. This will be one of the harder things to do, of course, because the main stream is traditionally targeted elsewhere.
If the efforts to define what HR Audio is are vague or misleading, HR will remain a boutique format doomed to niche markets, and cause further consumer confusion by logos that are worse than meaningless, they’re misleading.