Dr. AIX's POSTS — 22 June 2015

By

Yesterday was my travel day. I had an early flight (6:20 am) and had to get up early – very early – to get to the airport on time. After dashing to the studio to grab my Oppo BDP-103 and some sampler discs, I got to LAX about an hour prior to my flight, checked my bag, and went through security. I sat on an aisle next to a couple of young women on their way to Dublin. How do I know they were going to Europe for a month long vacation? How do I know that one of them is recently divorced after 13 years? Because they didn’t stop talking during the entire flight…over four hours of non-stop chatter. I didn’t bring a set of headphones or earplugs.

It was a very hot father’s day in NYC. I took the local train instead of the express from JFK to Manhattan, which added 45 minutes to my transit but eventually got to the AirBnb place near the village. It’s on the 5th floor but very nice and air-conditioned. Thankfully.

While I was on the plane, I managed to get connected to the Internet and went through the numerous comments on the FAQ section of the high-resolution audio guide. Thanks for the help in reviewing and commenting on the content, wording, etc. I think the latest version is much improved although I’m feeling that there are more questions that need to be added to the list.

One reader asked about “resolution” as it applies to audio. A number of articles on high-resolution audio use a graphic analogy with pixel density and digital photos to show how increasing the number of pixels and the bit depth associated with each pixel results in a more realistic image. People get this comparison. But it’s a little tougher to grasp how a sound wave can be digitized and the impact of more samples and longer words.

First, we need to establish that “resolution” only applies to analog signals that have been digitized. There is no such thing as “analog resolution” as some writers have stated. Analog signals remain in the analog domain. The compaction and rarefaction of air molecules contain and convey sound. The electrical current and the alternation of voltages in a wire also represent the actual analog signal captured from a microphone. And so do the grooves on a vinyl LP or the magnetic “domains” on a length of analog tape. There are no discrete quantities associated with analog signals. The voltages smoothly and continuously swing back and forth at amplitudes that mirror the loudness of the original sound.

Resolution applies when we convert the smooth, continuously variable electrical signals into discrete packets of information at each sample time. We don’t try to “digitally” draw each waveform with increasing numbers of pixels. Sampling an analog waveform requires a steady stream of samples and a measurement mechanism to establish a value at each sample. As for resolution, the higher the number of samples and the longer the digital word associated with each sample, the higher the resolution.

We have to understand the concept of the Nyquist theorem, foldover/aliasing, quantization error, jitter, and the rest of the issues associated with digital sampling. If we raise the sample rate, the resolution of our sampling system improves…and we get higher frequency response, lower noise, easier filters, and double the amount of data to store and retrieve. Increase the word length and the result is a lower noise floor and higher dynamic range.

There are other frequently asked questions…send them along and we’ll refine that portion of the consumer guide to HRA. Got to run to the Metropolitan Pavilion and start setting up the demo system for tomorrow.

Forward this post to a friend and help us spread the word about HD-Audio Forward this post to a friend and help us spread the word about HD-Audio

Share

About Author

Dr. AIX

Mark Waldrep, aka Dr. AIX, has been producing and engineering music for over 40 years. He learned electronics as a teenager from his HAM radio father while learning to play the guitar. Mark received the first doctorate in music composition from UCLA in 1986 for a "binaural" electronic music composition. Other advanced degrees include an MS in computer science, an MFA/MA in music, BM in music and a BA in art. As an engineer and producer, Mark has worked on projects for the Rolling Stones, 311, Tool, KISS, Blink 182, Blues Traveler, Britney Spears, the San Francisco Symphony, The Dover Quartet, Willie Nelson, Paul Williams, The Allman Brothers, Bad Company and many more. Dr. Waldrep has been an innovator when it comes to multimedia and music. He created the first enhanced CDs in the 90s, the first DVD-Videos released in the U.S., the first web-connected DVD, the first DVD-Audio title, the first music Blu-ray disc and the first 3D Music Album. Additionally, he launched the first High Definition Music Download site in 2007 called iTrax.com. A frequency speaker at audio events, author of numerous articles, Dr. Waldrep is currently writing a book on the production and reproduction of high-end music called, "High-End Audio: A Practical Guide to Production and Playback". The book should be completed in the fall of 2013.

(22) Readers Comments

  1. This article is helpful. Just to clarify my understanding-Sample rate is ie. {16 bit vs 24}. Is word length related to frequency ie. 48000 vs 96000?. You don’t define 24 bit 48000hz as high res, but I guess it’s a step in the right direction. Thanks

    • Barry, sample rate is the higher numbers (i.e. 44.1 or 96 kHz) and word length are the 8-16-24-bits. 48 kHz 24-bits sounds like it would be a step up from CD quality and it is probably indistinguishable from 96/24 but I’m holding on to the fact that there is music above 20 kHz. And I think (believe) that humans brains need those components in some way that we can’t yet identify.

  2. I love the in depth explanations of the “reality” of HD music you offer.

    What I would like even more, is to hear from you, some ideal examples of 24 bit 96 Hz master recordings with perfect provenance, and that were properly produced to the stage of the download now offered….and….where to get these….Clearly some will be on HD Tracks and a few other sites…but there is a big problem with these sites–at least for me….
    They tend to feature their audiophile tracks with chamber music or jazz full of the most UN-inspiring jazz performances…the wrong genres and sub-genres –at least for my interests….I have big magnepans and a very clean system, but I have to like the genre of the music I am playing….
    Maybe you could offer up something like a Spotify playlist, that establishes what a person likes…then a way to FIND this with optimal provenance and production quality for a buy and download link, is what I would really like… 🙂

    • I’m running right now but will get back to you. Be prepared though, there are not that many audiophile quality rock and pop records.

      • Awesome.
        I am aware that most rock will not be well recorded, and that old jazz or Blues will never be audiophile quality…

        Here is the type of music I would want in 24 bit/96 hz ( not so much these exact tracks or artists, as much as this type of music or sound…and of course, I would love it if I could get these exact titles in HD as well 🙂 )

        https://open.spotify.com/user/danvolker4/playlist/4d1v8YLDbosVh9VMpbU8vK

      • 9. What types of music are available in High-Resolution?
        If we adhere to the definition offered at the beginning of this FAQ section, most of the “High-Resolution Audio” is classical, jazz, or acoustic. These are the typical audiophile genres.
        Commercial recordings of rock, pop, urban, folk, country etc. are not being produced in high-res.

        True but maybe major marketing errors are being made..If it weren’t for the baby boomer market for pop, rock, etc there wouldn’t be a High End. I’ll bet MFSL sold more copies of Dark Side of the Moon than all their releases of classical, jazz, and acoustic titles combined. And these are the people like myself that have bought the large majority of all the high end components sold in the last 40 or so years.
        There is now a new generation of music and listeners. They have money and are buying expensive Beats phones, players, etc, because they have been properly marketed to by those manufacturers. The popular market is huge while classical, jazz, etc continues to die. If the HR industry doesn’t recognize and market to where the market is, it will soon be as dead a classical music.
        Mark, You did a fantastic recording of Mark Chesnutt, a proven record seller with 4 RIAA platinum albums and 8 number 1 singles. The recording you did could have been marketed as a New HR Greatest Hits release and possibly done VERY well, he did a Greatest Hits release in 96 that went Platinum. I looked at his wikipedia page and your project is not even listed in his discography? What happened to the marketing of this project for the most popular artist that ever stepped up to the mic in AIX Studios?
        The HR industry has to step up the recording and marketing of todays popular artists, there is no more time to waste.

        • I thought Mark Chesnutt was going to be huge for me…but his fans don’t know what a Blu-ray disc is…and I’m not sure about DVDs either.

          • I got the stereo download, fans didn’t need a blu-ray
            Your bigger than that condescending statement.
            Look in the mirror for the guy that dropped that marketing ball. Why isn’t your project even listed in his wikipedia?
            Of is that the fault of the uneducated country fans too? 🙁

          • My comment was not meant to be condescending. This is what I was told by the people in Nashville that I hired to work on the project. They said I needed to make a CD or the project wouldn’t sell to Mark’s fan base. I don’t make CDs and they didn’t find or desire the downloads at any quality level…even from iTunes.

  3. Dear mark,

    Your affirmation that high resolution audio has nothing to do with analog audio is a bit confusing to me. Comparing an audio system that has 20Hz-20KHz +/- 3dB frequency response and 90dB S/N with another one that has 15Hz-50KHz +/- 3dB and 110dB S/N, doesn’t the later one have higher analog audio resolution? If not, what term must be used on such situations? Is there a single term that indicates better analog quality and that can be applied on these cases?

    • The analog world might have better specs but it doesn’t have resolution…that is the exclusive domain of digital audio. Better specs, better fidelity, higher fidelity…but not higher resolution.

    • Ronaldo
      Which analog storage media does have 15 Hz to 50 kHz frequency response within +/- 3dB and 110dB S/N? The analog bottlneck is the storage media (distribution container), not the recording or reproduction electronics. And this both in specs and fidelity. There are also specs like linearity, channel separation, distortion, loss of quality within the dublication/manufacturing process of the media to take into consideration.

      Fritz

  4. But you need to explain 2 other parts of the listening equation that are both analog, namely the speaker/headphone and the ear. I’ll leave the former to engineers to discuss but as otologist I have some expertise about the human ear. Sound waves travelling down the ear canal, tympanic membrane, ossicles and endolymphatic fluid are analog. It is only the deflection of hair cells that creates an electrical signal that one can supposes is equivalent to a digital signal that is processed in at least 3 areas of the brain. The point is that with all this A/D & D/A conversion a lot of discrepancies & artifact are introduced. No one wants to discuss how much of a difference the ear can identify. There needs to be double blind studies but there are no sponsors.

    • Speakers are a weak link…but can do an amazing job. The JBL M2 studio reference monitors are incredible and actually match the needs of HRA. As for the ear, it’s been well studied. There are imperfections in conversion processes but they are so minimal these days that your ears are getting a rare treat. Some testing has been done but not enough.

  5. There is, indeed, such a thing as analog resolution, whether you are talking about electrical, optical, or some other type of system. It is defined fundamentally the same way as for digital systems: the smallest change at the input that produces a discernible change in the output.

    Here is the generalized definition of the resolution of a measurement system:
    http://www.itl.nist.gov/div898/handbook/mpc/section4/mpc451.htm

    Here’s an example of determining the resolution of an analog signal that measures distance, so it’s easy for people to interpret:
    https://sensortech.wordpress.com/2010/03/02/understanding-%E2%80%9Ctrue-analog%E2%80%9D-resolution/

    So, it’s not as simple as counting bits in a digital system, but analog resolution is defined and measurable. In reality, as you know, the real resolution of a digital system is similarly dependent on noise.

    For an audio example, if an analog system – e.g. a magnetic tape – has a noise floor at -60 dB, that is its resolution. If the full-scale signal can vary between +1 V and -1 V, then the system can’t resolve any signals smaller than 2 mV without some additional sort of filtering.

    This is exactly why it’s linguistically consistent for you to refer to old analog recordings as standard resolution – at least as far as amplitude/dynamic range is concerned.

    As with the FAQs, we technical members of the community critique your posts in hopes that the clearer and more accurate the message on high-resolution audio, the better it will cut through the nonsense.

    • Thanks Andrea…I know you’ll keep me honest. I think there is however, a difference between how we think and use “resolution” in the digital realm as opposed to analog. Got to run to the show today….but your point is taken

  6. Mark,
    How would you characterize the increase in information per unit that you get when using larger format magnetic tape (e.g. 35mm) or recording and playback at higher speeds, which again, seems to create a higher resolution analog product?
    It seems like there is an increase in resolution in these cases, but perhaps we are using “resolution” in different ways.

    • Ken, as Andrea pointed out…I use the term “standard-resolution” for analog tape and vinyl LPs. So I guess resolution can apply to analog information. I do believe that it takes on additional meaning with digital representations of data.

  7. The more I listen to your explanations about the effect of increasing sample rate, the more it seems that “high resolution” is a misnomer, or at least misleading. Suppose that you have a recording session from which there is no musical information above 22khz – absolutely none. In that case a high resolution recording (96/24) would have equal resolution to a standard resolution recording (44.1/16), and would probably be absolutely indistinguishable from the standard definition (CD quality) recording. It would have a higher possible dynamic range (which it probably wouldn’t use) and perhaps a lower noise floor, but the resolution of the recorded information would be exactly the same, since a higher sampling rate simply increases the highest recorded frequency.

    Am I still confused?

    • It is confusing. Let’s look at a couple of things. If you held a recording session, there will always be content in the room above 22 kHz unless you deliberately filtered it off. IF the medium you’re recording was analog tape, you’d get some audio above 20 kHz. If you record to standard resolution digital at 44.1/48 kHz, you wouldn’t (maybe a little). If you compared the two with regards to frequency response, they would be hard to tell apart. However, the dynamic range of digital…even 16-bits…would allow much greater dynamic range than the analog tape…if you want it.

      I think you’re correct.

  8. Thanks Mark for your efforts … I still have trouble understanding/conceptualising word length and bits.
    thanks

    • Let me get back to you on this. I’m in Queens this morning and have to hustle over to Manahattan. In short, bits make up words and the more bits in each one, the more levels exist in the amplitude axis. Each bit equates to roughly 6 dB of dynamic range (with quantization errors), so the more bits the more potential dynamic range or the lower the noise floor.

Leave a Reply

Your email address will not be published. Required fields are marked *

four − 1 =