Musical Provenance: Tracking the Tracks, by Mark Waldrep

Dr. Mark Waldrep

This is Part 2 of a planned 3-part Q&A with Dr. Waldrep of AIX Records and If you haven't read Part 1, I'd recommend starting there (Understanding High Definition Audio: Facts and Fiction). As the title for Part 2 suggests, Mark talks about the important subject of provenance with regard to HD downloads. I'd like to thank Mark once again for his time and for agreeing to tackle this somewhat controversial subject.

Musical Provenance: Tracking the Tracks, by Mark Waldrep

The term provenance has been traditionally associated with works of art. According to the dictionary on my computer,

the place of origin or earliest known history of something: an orange rug of Iranian provenance.
• the beginning of something's existence; something's origin: they try to understand the whole universe, its provenance and fate.
• a record of ownership of a work of art or an antique, used as a guide to authenticity or quality: the manuscript has a distinguished provenance.
However about thirteen years ago, high-resolution, digital audio recording first became an option for audio engineers. I started using the term provenance when referencing the original production equipment and processes associated with audio tracks. It seemed everything that showed up on new audio formats like DVD-Audio and SACD qualified as “high-resolution”, “high definition” or “advanced resolution” simply because they were released on one of those formats. As an audio engineer with over 40 years of experience, repair technician, university professor and studio owner with analog and digital audio systems, I knew that not all recording equipment and production procedures result in the same fidelity.

"...I believe that it is helpful and informative for music consumers to know the provenance or 'the beginning of something's existence; something's origin' before they make a purchase."

Nor has everyone that has ever made and released a record had the same sonic preferences, label requirements, target market or goals. Therefore, I believe that it is helpful and informative for music consumers to know the provenance or “the beginning of something's existence; something's origin” before they make a purchase. This is important because the audible difference between a real HD-Audio track and an older standard definition track contained in an HD-Audio “bucket” can be quite substantial. If everything ever recorded can be elevated to HD status by simply copying it into a 96 kHz/24-bit sound file, then what's the point of having any quality distinctions? In a world without differentiation, all audio tracks sound equally good!

I believe it's important to distinguish what is an HD-Audio recording and what is not. Having some quantifiable measures against which various high-end audio formats can be compared wouldn't diminish the validity of any particular format, but would provide a reality check for consumers looking at HD-Audio for the first time. I have heard from numerous music fans how disappointed they were with the sound quality of their first downloaded high definition track. They simply cannot hear any difference between the version that they've experienced for years on vinyl, CD or SD downloads and the NEW 192 kHz/24-bit version. The reason might be because there is no difference except the price tag.

Dr. Mark Waldrep with Greg LeDuc (right)

In Part I of this series, I defined what an HD recording is and what it is not. It's been many months since that piece was posted (July 2012) so it might be worth quickly restating my definition. Here it is:

I define HD-Audio as a recording that has been captured during an original session using equipment capable of matching or exceeding the capabilities of human hearing. If the generally accepted measure of the human auditory system includes a frequency span of roughly 20 Hz to 20 kHz and a dynamic range that tops out at around 135 db, then a recording system would need to be able reach these specifications to be considered HD. In the world of PCM digital recording this would translate to at least 48 kHz and 24-bits. Given there is some evidence that higher frequencies may impact our listening experience and that moving to 96 kHz has advantages for equipment designers, it's seems reasonable to adopt 96 kHz as the minimal HD sampling rate. As an engineer/producer creating new HD tracks, I choose 96 kHz/24-bits as the minimal specifications to achieve HD-Audio.
Some prominent audio engineers and researchers agree with me and others don't. Getting consensus on a definition of HD-Audio is proving to be difficult. Everyone with an interest in high-end audio wants to be under the HD umbrella. But I don't think anyone would disagree that there are perceptible sonic differences between recording and distribution formats. And that's as it should be. Different producers, engineers and musicians want their projects to have an individual sound just as guitar players refine their equipment and technique to achieve their own sonic signature. Certainly, everything ever captured on an undulating tin cylinder, spinning lacquer disc, piece of 180-gram vinyl, moving wire or magnetic oxide embedded Mylar tape or hard disc drive doesn't produce the same fidelity or sound.

"I doubt anyone would call a 78-rpm lacquer recording from that period HD."

As stated above provenance is defined as the “the place of origin or earliest known history of something”. Knowing the provenance of a particular music production provides useful information on what a user can expect with regards its fidelity. If a track was recorded back in the 1920s using an analog acoustic recording system, we expect it to exhibit sound characteristics associated with that era. I doubt anyone would call a 78-rpm lacquer recording from that period HD.

Recently, I was involved in the Salon Son & Image in Montreal. It's an audio trade show that has been going on for over 25 years put on by Michel Plante. In one of the larger rooms, I happened on a display of recorders, receivers and headphones from the past…a sort of historic timeline of music appliances. One of the older pieces in the collection was a 1910 Gramophone. And it functioned! The gentleman there cranked it up, dropped the stylus on a 10-inch lacquer disc and presto…out came a wonderful piece of music featuring a male crooner and chorus. To "turn it up" they opened the two-hinged doors under the turntable. The sound was a combination of scratches and substantial surface noise and heavily filtered music. I was impressed with the whole setup and amazed at how far we've come over the past 100 years. The sound of an acoustic recording playing from a Gramophone is not likely to gain in popularity like vinyl LPs have over the past few years! The playback was historically interesting but sonically terrible.

"If the maximum frequency response of a track from the mid 1960s is around 15 kHz and the dynamic range is somewhere around 40-50 dB, then should it be marketed as a new "HD" track?"

Knowing that a classic rock album from the 1960s was tracked on a Studer 4-track, mixed to an Ampex 440 2-track deck, mastered to another tape and then cut to a lacquer master prior to pressing vinyl copies is informative as well. If the maximum frequency response of a track from the mid 1960s is around 15 kHz and the dynamic range is somewhere around 40-50 dB, then should it be marketed as a new "HD" track? If that very same album is transferred to an 88.2 kHz/24-bit digital bucket and sells for $20, is that a high definition track? Is it more HD and worth $30 if it's placed in a 176.4 kHz container? I don't think so. If customers are informed of the provenance of a track, they can choose to acquire it at any number of different digital formats at a variety of sample rate and word lengths…even MP3 or DSD. But they won't be disappointed because they know what to they can expect in terms of fidelity.

John Storyk, studio architect, and Dr. Mark Waldrep

An article in HiFi News by Keith Howard over a year ago set off a firestorm of controversy because his analysis of a number of "so-called" HD digital downloads revealed that many of them were falsely labeled HD and even showed signs of having been sample doubled. Other writers have rightly pointed out that, "If a website is charging a premium for high-resolution downloads, then those files need to contain information beyond that which can be offered on CD. If they do not, you should ask for your money back.”

My favorite comment comes from Keith Johnson, the well-known engineer at Reference Recordings and co-developed of the HDCD technology who said, "...there will be re-issues even though high-resolution will not improve them. Some reviewers use analyzers to spot these marketed imposters.”

"Purveyors of 'so-called' HD digital downloads should make clear that the 'HD' specifications of the final downloaded container file does not mean that the fidelity of the track is high definition."

I have no problem with customers having the ability to purchase and download classic albums/tracks from the original master tapes as digital files. The sound of good analog to PCM digital transfers will preserve everything that was experienced in the studio when those mixes were produced. But they should also be informed about what to expect for the premium dollars that they are coughing up. Purveyors of "so-called" HD digital downloads should make clear that the "HD" specifications of the final downloaded container file does not mean that the fidelity of the track is high definition.

There were some attempts by SPARS (Society of Professional Audio Recording Services) back in the early days of the compact disc that identified whether a particular production stage was accomplished in the analog or digital domain. A track that was recorded on an analog 24-track machine and was edited/ mixed using analog equipment and then mastered for CD release was identified by the letters AAD. A purely digital recording would use the designation DDD. There were albums that charged a premium if they were produced entirely in the digital domain. The system fell out of favor over time as engineers moved between analog and digital processing and equipment during their productions. I think the effort was worthwhile and valuable to consumers.

"I would like to see high definition digital download sites adopt a voluntary program to identify and categorize tracks as to their provenance and technical specifications in a way similar to the previously mentioned SPARS codes."

I would like to see high definition digital download sites adopt a voluntary program to identify and categorize tracks as to their provenance and technical specifications in a way similar to the previously mentioned SPARS codes. I recognize that it's a daunting task but it would be of tremendous benefit to consumers as music servers continue to replace optical disc players. If a track was recorded on analog tape and originally mixed to a 2-track analog deck, letting consumers know that will benefit everyone in this new market. Marketing everything that was every recorded and is now being made available at 96/24 or higher and called HD is false marketing. Some HD download tracks "suck" when they are supposed to be HD. If the original track sucked then the download version will too…HD specs or not.

labjr's picture

Thanks for an article that places importance on provenence. Maybe retailers and the labels that provide the material will take notice

"Is it more HD and worth $30 if it's placed in a 176.4 kHz container? I don't think so."

From what I've heard, analog tape transfers sound better at 176.4 than 88.2. However, I don't think it's worth $30.

I'm a classic rock guy so pretty much everything I listen to is from analog tape and I'm OK with that. I'm probably not going to listen to DDD recorded avant garde jazz because it sounds more HD than analog recordings.

I think most of the problem people have with provenance is when something's not what it appears to be or is supposed to be and the retailer isn't doing their part to prevent this. Seems like a deceptive practice more than just "not HD" because of the quality of the recording. 

A SPARS code type of system may work if everything is revealed. Including upsampling,transfers from DSD to PCM and vice versa.  Whether the oiginal master tape used or some other copy of it. And when it was transfered and by whom.

I'll pay more for something that's guaranteed to be a good quality direct transfer from the original master tape, done with the best converters currently available. And by a reputable engineer. And even if it sounds just a little better than a CD. But I won't buy crap that's been retransferred, upsampled etc. And the retailer is defensive instead of being helpful and apologetic.

deckeda's picture

Thanks again Mark. From your Part 1 definition reiterated above, that works for me in a broad sense because for example we still see recordings advertised and released in "hi res" when all they really mean is 16/44 lossless, since lossy iTunes/Amazon files are now the market's standard bearers based on sales volume.

I take the time to contact musicians and labels where and when available about that and encourage everyone to do so as well.

Where your line(s) in the sand don't work for me is in your characterizations of "HD-Audio" as needing a certain frequency response and dynamic range capability. CDs still suffice for that even if they can't reach 135db (most music never does, either) and yet many CDs sound pretty lame compared to some technially inferior formats.

I don't have a solution for this. But I do know many aspects of harmonic structure and musical complexity have been left out of that discussion where flat response and incredible dynamics are heralded as the main, or only? necessary arbiters.

So yeah, I'm very comfortable trusting my ears on this.


Regarding a "new SPARS" code, I agree but am not optimistic. There's no infrastructure in place to note the critical parts of the production chain, and no shared philosophy at the labels as to why customers might care.

jim tavegia's picture

Mark gets it and his recordings are some of the best I have ever heard.  He is right to recommend that 2496 should be the base line for anything being called HD and it must start as a master recording there, not just be move up to 2496 or 24192 or DSD. 

I do agree with re-masters of the original analog tapes being re-worked and sold as DSD/SACDs, but that is only to retain for our benefit all that the master tape is holding so we can experience what the mastering engineer gets to enjoy. I would even be happy if they were 2496 and offed as discs that I could play in any of my 2496 capable DVD players, much like what Barry Diament does with his SoundKeeper Recordings. Barry's work is a benchmark for me. 

Drtrey's picture

if the file technically meets the criteria but sounds horrid? My home rips of vinyl sound better than some purchased HD downloads and worse than others! Some Girls does not sound as good as my rip and I recently bought some Benson and Ella files that were gorgeous sounding, much better than anything I have ripped myself.

So the nomenclature would be good, but it would not solve the problem of good sound.



Guidof's picture

Mr. Waldrep's point is well taken. However, knowledge about a recording's provenance remains a secondary matter. The quality of the master is of much greater consequence, and in this domain practices such as compression, excessively close miking, poor balances, and other assorted ills much too often result in recordings that while being HD are painful to one's ears. Just as DDD did not necessarily ensure better sound quality than AAD, neither will true 24/96 or 24/192 necessarily yield high recording quality.

But of course, any effort toward achieving greater honesty about provenance is to be applauded.

Guido F.

timorous's picture


Thank you Mark, your points are well-taken. It seems to me, the main thing that a true HD transfer (24bit/96kHz or better) affords is..the closest facsimile to the original recording, including the subtle harmonics and ambient cues, etc. These qualities, while subtle, are hearable by anyone with a fairly decent playback system. For some reason, analog recordings and vinyl (despite their more limited bandwidth and dynamic range) seem able to preserve these qualities, while all manner of digital recording, while it comes close, seems to skew these things. Especially CD's.


There also seems to be many subtle flavors of digital converters which serve to either bring out the best in each recording, or skew it the other way. Then there's the (re)mastering engineer's ideas of what 'sounds right'. Of course, this has always been the case, even way back in the 1960's. Also, the older the analog tape, the more problems that crop up, like oxide shed, loss of magnetic remenance, dropouts, to name a few. Sometimes a 2nd generation master tape is in better condition than the original, and would probably be a better choice, if it weren't for the 'original' aspect of the provenance issue.


There are too many inter-related issues to factor into the mix, to really nail down the provenance of any given recording, except to say that the 1st generation master should be considered the first choice for any company issuing the HD files. Plus, very little processing (i.e. no further compression, unnecessary eq or noise reduction) should be used. Except in those cases where a marginal, strident recording can (and should) be made more 'listenable' at a higher playback level, with some judicious and selective equalization.

Dr. AIX's picture

I'm not advocating the know the production path of a recording will make things sound terrific. But knowing in advance how a recording was production...including whether some mastering engineers hammered the life out of a track...can empower consumers in their purchasing decisions.

For my own productions, I don't master them at all.

jsaliga's picture

Great article. I'll go so far as to say that in my experience sampling rates over 48KHz don't make much if any difference to the sound quality you hear. I have my own small digital recording studio and done a great deal of testing. But as always YMMV. 16 bits is more than sufficient. But I buy a lot of 24bit/96KHz and 24Bit/192KHz content. Why? Because most of the time, though not always, they offer better mastering and use better sources than a standard CD release. So I can, again, usually, but not always, expect HD titles to not have the dynamic range crushed out of it by excessive compression or have tons of clipped samples. Unfortunately in today's music buying climate you have to pay a premium to get competently produced and mastered product. So in the end I would prefer 16bit/44.1KHz content that was very well produced and mastered to the same content @ 24bits/192KHz that was stepped all over by someone who thinks that brick walled, compressed, and clipped digital music sounds better.