The New High Resolution Audio Standard

June 13, 2014

sony_hra

Well, I guess everyone knows by now that the Digital Entertainment Group (“DEG”), in cooperation with the Consumer Electronics Association (“CEA”) and The Recording Academy, and in partnership with the big three record labels (Sony Music Entertainment, Universal Music Group, and Warner Music Group), have officially defined high resolution audio (“HRA”) as “lossless audio that is capable of producing the full range of sound from recordings that have been mastered from better than CD quality sources.”

My first reaction was dismay and disappointment, actually more like utter exasperation. It is the lowest standard the music industry could have possibly adopted without actually including CD. Upon further reflection, and a long conversation with Marc Finer, Senior Director of DEG and the man responsible for managing the disparate interests and expectations of all those involved in this effort (and from personal experience a man of infinite patience), I think the announcement can best be described as one small step for audiophiles, one giant leap for the music industry, and it has the potential to be a positive development for music consumers in general. A dagger has been thrust in the heart of CD as “Perfect Sound Forever” and the music industry has turned the page. Ostensibly. Is it what I would have wanted? No. Does it have its problems? Yes. We’ll get to those.

Coincident with the announcement of the definition of HRA, we were also advised that four new descriptors have been devised to inform consumers about the quality of the source from which the digital downloads they are buying were made, much like the abbreviations AAD, ADD, and DDD from the early days of CD. They are:

MQ-A – from an analog master source
MQ-C – from a CD master source (44.1k Hz/16 bit content)
MQ-P – from a PCM master source 48 kHz/20 bit or higher (typically 96/24 or 192/24 content)
MQ-D – from a DSD/DSF master source (typically 2.8 or 5.6 MHz content)

Alert readers will note that between CD quality and 48/20 lie the sampling rates and bit depths 44.1/20, 44.1/24 and 48/16. They appear to have fallen through the cracks. It is the type of discrepancy that, as a lawyer, absolutely drives me crazy. And their omission from the rubric is not simply of academic interest. I don’t know how many masters exist at 44.1/20, but it excludes at least all of the HDCD catalog. The omission of 44.1/24 excludes the entire Beatles catalog, for example, and my understanding is that there are a great many other 44.1/24 masters out there. HDTracks alone is currently offering over 600 of them (for the same price as 96/24!). And the omission of 48/16 undoubtedly excludes many of the soundtracks from the early days of digital recording. All of those masters, and the resulting digital downloads, although technically HRA, do not qualify for any of the MQ descriptors. As it stands, they are – inexplicably – orphans.

The second issue is that a recording with the designation MP-C is by definition not HRA. There is simply no way that a recording can “have been mastered from better than CD quality sources” if it was derived “from a CD master source.” It is oxymoronic. If you see an HRA file labeled MQ-C it can only have been upsampled, and hence the kind of fake HRA file that plagues the high resolution download business. The best one can say with respect to MQ-C is that it is better to have an explicit designation that a file is a phony high resolution download than not.

The third issue is that the designator MQ-P is broad enough to encompass 48/20 through 192/24 and DXD sources. Since you will be paying a premium for HRA downloads, surely you are entitled to know what sample rate and bit depth beyond CD the sources are. And with the three-tier pricing structure currently in place (de facto $14.98 for 44.1/16, $17.98 for 44.1/24-96/24, and $24.98 for 192/24), most of us are not going to want to pay a premium for a file sourced from a lower resolution HRA master that has simply been upsampled to a higher resolution HRA file. This information is vital. It goes directly to the center of the provenance debate. Marc Finer has advised me that the press release was necessarily abbreviated, and the record labels will be providing this information both in the metadata and to the retailers so we will have full disclosure. The music industry’s performance to date on this issue has been spotty at best, so we must remain vigilant, particularly early on as the new standards are rolled out.

A fourth issue arises from the interplay of the definition of HRA and the designation MQ-A. Query: which analog sources are “better than CD quality”? The only two I can think of which could possibly be considered better than CD quality are LPs and professional 1/4″ and greater two track and multi-track reel-to-reel tape. Under the best of circumstances, they are capable of a frequency response extending beyond the 22,050 Hz limit of CD. However, neither format can approach the 96 db dynamic range of CD. Hopefully we won’t see any 96/24 digital copies of wax cylinders, 78s, consumer reel-to reel, 8-tracks, cassettes, or the like marketed as HRA from MQ-A. I also hope we won’t be seeing any “needle drops” being marketed as HRA from MQ-A. MQ-A should have been limited to professional 1/4″ and greater two track and multi-track reel-to-reel tape.

It is worth taking a closer look at the interplay between the definition of HRA and the MQ designations. HRA requires that the recordings must be “mastered from better than CD quality sources.” In other words, the original recordings, prior to mastering, must come from better than CD quality sources. On the other hand, the MQ designations all speak of recordings coming from a “master source”. For example, to earn a MQ-P designation, the recording must come from “a PCM master source 48 kHz/20 bit or higher.” In other words, the actual music source can be anything as long as it was mastered at 48 kHz/20 bit or higher.

How is this relevant? Well, take Beck’s new Morning Phase album. John Atkinson, Editor of Stereophile magazine (and not just some random “Audacity cowboy”), identified by spectral analysis that on at least some tracks the music sources were MP3 and CD quality. So clearly the album doesn’t meet the criteria for HRA. However, those MP3 and CD quality tracks were then upsampled to 96/24 for mixing and mastering. In other words, even though the music source was not HRA, the master source was 48 kHz/20 bit or higher, so the album qualifies for MQ-P. You really have to wonder whether anyone actually read and thought about these definitions before adopting them. Or did they intentionally make them fuzzy and inconsistent so that they would have wiggle room in identifying recordings as HRA with an MQ designation? Giving the music industry the benefit of the doubt, it is just sloppy work on their part. Complete clarity and transparency is required.

Personally, I would have preferred to have seen 88.2/24 and above as the criteria for HRA. Unlike 48/20, there is no disagreement from any quarter that those sample rates and bit depths are high resolution. And it would have avoided the quagmire of rationally differentiating between 44.1/20, 44.1/24, 48/16, 48/20 and 48/24 files. More importantly, it would have made it clear that the music industry was offering a premium product, not just a slightly better one, for a premium price.

My hope is that this is just the first step toward a higher resolution future, and that 48/20 is but a way marker on the road to the higher standard of Ultra High Resolution, much as HDTV has been a way marker on the road from SDTV to 4K. And as much as we would like to wring our hands about the relatively low bar set, let’s at least acknowledge that the Leviathan that is the music industry is moving past CDs as its standard. It is definitely a move in the right direction, however botched the effort and incremental the progress we may believe it to be.

All that being said, the real elephant in the room is pricing. Is anyone outside our little circle really going to pay $17.98 to $24.98 for HRA downloads, particularly when the only thing they are buying is the right to listen to them? Unlike physical media, once purchased, they cannot be traded or sold. If you only have the right to listen to them, why not just stream them? The quality of streaming is improving every day. QoBuz and WiMP are already streaming lossless CD quality. The additional bandwidth to stream 48/20 is negligible. Is the music industry’s window of opportunity to sell HRA downloads perilously narrow? Have they already missed it? That is a whole other discussion.