Kicking Off 2016 In Las Vegas
The huge Las Vegas Ferris wheel is just out my window at Bally’s Hotel and Casino as I write today’s post. I’m here because yesterday was the first day of the annual ALMA symposium just down the street at the Tuscany Convention Facilities and some months ago I was invited by their board to give the keynote address. This year the association of loudspeaker manufacturers decided that high-resolution audio would be the central theme of their annual event. The title “Hi Res Audio: A Bridge To Build Or A Bridge Too Far?” asks a familiar question. Their vice president attended a presentation I gave at the Rocky Mountain Audio Fest a few years ago and was impressed enough to suggest to their board that I might be a compelling keynote speaker. I’m here because I gladly accepted their invitation.
And because the annual International CES 2016 show starts on Wednesday and I’m looking forward to visiting the Venetian Hotel in search of the latest on high-res audio and music. The “High-Res Techzone” room has been abandoned but there will still be a few panel discussions on Wednesday and Thursday.
I drove up from Los Angeles early yesterday morning in order to participate on a panel at the ALMA Symposium with Dr. Sean Olive of Harman International, Robert Stuart of Meridian and MQA, and Steve Temme who runs a testing lab called Listen. I had a few moments to chat with both Sean and Bob prior to the session. MQA will have a large presence at the Venetian Hotel as the company continues to ramp up their MQA initiative. I asked Bob about the questions I gave him back in June after the Newport Show and he assured me that he would respond by the end of January. I’m looking forward to reading his answers and including them in the Music and Audio Guide.
The moderator of the panel began the session with the most obvious and yet challenging question about high-resolution audio. He wanted each panelist to offer their definition of what it is and what the benefits of high-res audio are. I started the discussion with my usual definition.
High-resolution audio and music is sound that meets or exceeds the capabilities of human hearing. For me, it’s really that simple. This definition doesn’t specify any particular format, sample rate, word length, or anything else. It doesn’t talk about resolution or the particulars of any format. I’m simply looking for results. If there’s a format from the past that can match the “specs” of our ear/brain listening mechanism, it’s in. And likewise if a new format or hybrid format comes along that measures up, it’s welcome too.
Maybe it would be easier to define high-resolution audio by identifying what doesn’t measure up.
I don’t want to get wrapped up in discussions of real world circumstances or commercial considerations. I simply want the technologies involved in capturing and reproducing sound to be equal to or better than our ability to hear.
Steve Temme offered his assessment. He took a lower road and stated that he’s OK with “just good enough” sound. If it measures better than CDs then he would call it “high-res”. This includes moving the PCM digital specs higher than 44.1 kHz/16-bits.
Robert gave a 2-hour presentation earlier in the day (which unfortunately I missed because of my transit time) on capturing high-resolution audio. Like me, he doesn’t feel it’s necessary to bind “high-resolution” audio to any hard numbers. He’s looking for an “end to end” solution that accurately mirrors the sound produced during the session in the playback system. The sound or fidelity that the live instrumentalists or singers created in the original performance should be maintained through the production process and be indistinguishable from the source performance. This is tied very closely to the MQA philosophy and technology.
Finally, Dr. Sean Olive hit on a position someplace in the middle. His position is closer to the CTA/DEG/NARAS definition of fidelity better than “CD Quality”. Establishing a consistent set of standards is key to advancing the HRA initiative.
I wasn’t surprised by the diversity of opinions presented by the panel of experts at the ALMA event. Four presenters offered four different assessments of what is and what isn’t high-resolution audio. The panel was not populated by marketing types, bureaucrats, journalists, artists, rock stars, editors, or consumer electronics companies. The definitions offered came from highly qualified engineer types…and still we can’t agree on the most basic and essential concept of what is and what isn’t high-resolution audio.
It’s been 16 years since I started exploring high-res music technology and production techniques. Is there any hope that 2016 will see consensus on the topic? I doubt it.
Mark, I like your requirements for HRA because they are quantifiable. I also like Bob Stuart’s opinion because it outlines a methodology to get to that end result. Again this all hinges on the quality of the recording and production at each stage. What I am leery about with MQA is that: A) will it try to justify an MQA certified remaster of Miles Davis’ Kind Of Blue, for example, as HRA? Because by the nature of the source recording, it can’t be. B) Does MQA charge some sort of licensing fee which might cause it to become just another money making certification vehicle? IE: “You want HRA, well then you have to have MQA or it really isn’t.”
BTW, hope I’ll bump into you wandering the halls at The Venetian this year.
Best….Carlo.
I am one of the early adopters of Hi-Res Audio. I have an 8TB network attached server, a custom PC that runs JRiver Media Center and Wadia DAC that feeds my Chord Amp and B+W speakers. A simple, uncluttered system. So far, I have been following the debates and only buying 96/24, FLAC, recorded recently, with some reassuring information about the recording and mixing. And from trustworthy websites such as 2L, Linn and ITrax. Breathtaking music. The excellent music I have found recently–wish I could obtain it in HiRes, such as Bella Hardy or Michael Samis. I also check first, if the artist is selling FLAC files directly from their own websites. But I am confident the artists and sound production folks will release enough quality onto the market to keep me satisfied. Keep pushing from your side, Mark, and thank you for your efforts.
Thanks for the positive assessment. It just shows that it is possible even in the midst of countering forces.
Mark, thanks for the update. It is obvious there is much momentum within the industry to classify as much as possible as hi-res. This will translate near-term to more hardware, software, and music sales. However, I continue to enjoy my 16/44.1 music with an occassional LP from my teenage years.
“High-resolution audio and music is sound that meets or exceeds the capabilities of human hearing. For me, it’s really that simple. This definition doesn’t specify any particular format, sample rate, word length, or anything else. It doesn’t talk about resolution or the particulars of any format. I’m simply looking for results.”
Mark, you’ve totally lost me and I have no idea what your talking about. As much as I respect you this sounds like some kind of lawyer speak/spin to me? I can hear and understand the output of a Edison cylinder, it meets the capabilities of my hearing, so does that make it HDA ????? HUH WHAT
I thought you had it nailed a long time ago. A recording sampled at better than redbook, kept at or above those specs thruout processing, and played back at those “better than redbook specs”. Simple and direct. KISS
I’m thinking you may have had a little too much to drink or smoke at Bally’s, Either you or I cause I haven’t a clue what the hell your talking about.
Sal, you’ve been reading this blog for a long time and haven’t understood my definition during all of this time? An Edison cylinder with its extremely limited bandwidth and very narrow dynamic range doesn’t meet the capabilities of human hearing by a long shot. Analog tape is much closer and CDs are closer yet, but it’s not until we can capture and deliver up to 130 dB of dynamic range and frequencies out to 20 kHz (I argue for 40 kHz) that we actually get there. If a 96 kHz/24-bit PCM file gets us there…which I believe it does…then it’s High-Res.
This definition hasn’t changed for me in 16 years. Are we clear?
It must be me then, that’s not at all what I read out of your first sentence. Or maybe your first sentence was a bad choice of words to describe your believes.
I haven’t been so distraught about something since Bush went to war in Iraq over fake WMDs. (I was against that too). Selling trumped up audio to dumbed-down masses is totally Un-American but acceptable to Snake Oil peddlers. And the band played on.
Take it easy Rod :-).
Yes – the band plays on… Remember this has been ongoing since the days of DVD-A and the early SACDs (circa 2000). Just that back then it wasn’t as easily checked with copy protection. These days, with standard FLAC downloads and access to FFT and DR plugins, it’s easy to assess and discover the obvious.
Mark: I really do hope you get some clarity on MQA’s capabilities. As far as I can tell, whether decoded or not, MQA is at best 16-bit PCM since they need the lowest 8 bits for the “encapsulation” process (and retain the highest 16-bits or so for compatibility when played without a decoder). In essence it looks like they’ve traded off ultrasonic frequency reconstruction for dynamic range in the audible frequencies!
If you’re looking for a “container” for music that is capable of 130dB dynamic range to fully capture the human hearing limits, I think this discounts MQA by this fact alone.
As an outsider I find it rather amusing. A product has been created whose usable significance is at best subtle to be as polite about it as I can is what is claimed by one of its most knowledgeable proponents, that has been given a name that is scientifically inappropriate, is targeted at what must be admittedly a niche market within a niche market, and now those who wish to sell it are arguing over how to define it.
One problem with defining high resolution audio by proclaiming it to be capable of all humans can hear is that this is a variable that varies from individual to individual and even from time to time for the same individual. Most hearing tests performed by audiologists are maximum sensitivity as a function of frequency up to 10 kHz. If you want a more complete or descriptive test you’ll either have to pay a lot for it because it won’t be covered by medical insurance or devise and conduct your own. It would be useful if reviewers had to report the results of recent hearing tests but that won’t happen. So even among audiophiles it isn’t clear how many would be able to hear anything beyond 22 kHz. I’m pretty sure I can’t and I doubt that I ever could even when I was younger.
I don’t see my definition as problematic with regards to individual variations. We already talk about human hearing as 20-20 kHz and 130 dB as a metric. The diminution of our abilities as we age or that happen as a result of illness can’t be tracked and included in the definition. If a format or audio system meets or exceeds the highest human specification, then I’m good with that.
I can’t hear 22 kHz and don’t think it’s required. But there may be other mechanisms at play besides out ears (the highest area of the cochlea resonates around 18 kHz) that may receptive to ultrasonics. It doesn’t matter. The processes of digitizing and reconstucting analog waveforms benefits from reaching beyond 20 kHz.
I think we are going to have to agree to disagree.
I doubt the need for reconstructing amplitudes of 130dB. As it happens, 120dB is the threshold of pain and even a short exposure would lead to tinnitus/mild hearing loss.
I really have high hopes of MQA. One issue I couldn’t figure out is the frequency response of MQA.
Mark, you are right about the need to reproduce frequencies above 20KHz, and some tests have shown that listeners do discern playback systems that reproduce above 20KHz. The fact that audiometry testing shows that we don’t “hear” frequencies above 20KHz does not mean that the playback of such frequencies contributes to the listening experience.
It would be great to hear about your experience of listening to MQA. I agree with Carlo above that the quality of the master will have a great impact on the ability of MQA to deliver “studio quality” playback. What if the “studio quality” master is severely flawed, whether because of bad engineering or because it is very old? We have to agree that MQA will deliver on its promise only as long as the studio master is good enough to encode into MQA.
You’re right that levels at 120-130+ dB SPL would cause pain and risk hearing damage. However, there are sound in the world…even musical transients…that get this high on occasions. A rim shot or orchestral tutti at fff can reach that high for instances. It’s easy to capture and hard to reproduce but it shouldn’t be ruled out because it’s potentially dangerous.
I met briefly with Bob Stuart yesterday in Vegas. He’s promised to encode some of my tracks and send me a decode (Explorer2) to do some critical listening. I also listened to some recent encodes of recordings by Morten Lynberg at 2L…really stunning! I’m going to write about MQA today and as soon as I get a chance to audition some of my own recordings through their system, I’ll share my thoughts.
I think my biggest concern is having to retransfer and process every master using the MQA matering tools. If it has taken 5-6 years to get 5000 new transfers from the major labels, how long do you think it will take to encode the entire catalog? And for what level of fidelity improvement?