Play It Again, Paul, LLC | High-Resolution Audio

High-Resolution Audio (May 2018)

For over a century, new technologies for audio media have appeared that increased the fidelity of recorded music, although this wasn’t always the intent behind the technology. Invariably, each improvement required new equipment to play the new media. Here’s a quick run-down of technologies and what was needed to adopt them.

Wax cylinders made their appearance in the 1880s, intended more for recording speech than music.:Phonograph for playing cylinders.
Shellac discs (78s) appeared in the early 1900s and opened the market for playing recorded music at home.Though brittle and rapidly worn because of high tracking weight, 78s offered greater durability than wax cylinders.Phonograph for playing discs.
Microgroove media (vinyl LPs and 45s) were introduced in 1948.They provided longer playing time and less surface noise than 78s, and the required use of smaller styli with a much lower tracking weight resulted in significantly less record wear.Small record players designed to play only 45s were a first step toward portability.turntable capable of speeds of 33-1/3 and 45 rpm (usually in addition to 78 rpm), smaller stylus (sometimes with a larger stylus appropriate for 78s mounted on the opposite side of the cantilever; the user could switch styli by moving a lever that rotated the cantilever to the other stylus), light-weight tracking achieved with a counterweight on the back end of the tonearm).
Stereo vinyl records appeared in the mid-1960s and could provide a “presence” unachievable with mono recordings.stereo stylus (initially, using a mono stylus on a stereo record could damage the record.)
Magnetic cartridges, which overtook the earlier piezoelectric (crystal and ceramic) cartridges, enabled a more realistic reproduction of the music.new cartridge, a pre-amp to increase the output voltage to line level and to remove the RIAA (Recording Industries Association of America) pre-emphasis equalization used in creating vinyl records (see Equalization).
Reel-to-reel tape was being used in recording studios beginning in the 1940s and made its way into the home use market by the early 1950s.While it produced a cleaner and more realistic sound than vinyl records, “R2R” never caught on in the audio recording market because pre-recorded tapes were significantly more expensive than LPs.R2R’s value was more for home recordings, and R2R decks are still in use in the audiophile community.Reel-to-reel tape deck or player.
8-track tapes (mid-1960s) were another step toward portability and were used extensively in car sound systems.Unlike reel-to-reel players, one didn’t have to fumble with threading the tape.8-track deck or player.
Compact Cassettes and their players (1960s) could be portable, and they were more durable than vinyl records and eliminated the problems of scratches and skipping.They proved superior to 8-track tapes because they avoided the problems engendered by the 8-track’s use of a continuous loop of tape.Since their playing speed was 1.75 inches/second (ips), they exhibited more tape hiss than reel-to-reel tapes, typically played at 3.75 or 7.5 ips.As time went by, noise reduction capabilities were added to newer cassette players.Cassette deck or player.
Compact Discs introduced digital formats to the marketplace in 1982 and offered greater dynamic range and lower noise than any of their predecessor technologies, and media wear was virtually eliminated.CDs’ audio tracks could be ripped to computer files, and they and their players could be portable.(Whether CDs offer improvement in musical fidelity compared to vinyl is subject to continuing debate.)CD deck or player; computer with an optical disc drive.

The quest for increased audio fidelity didn’t end with the CD, and several high-resolution audio (HRA) technologies have been developed and marketed, High Definition Compatible Digital (HCDC), Super Audio CD (SACD), and DVD-Audio (DVD-A) among them. Like those mentioned above, these digital technologies required upgrading to devices capable of playing the new formats, though some efforts were made to make the media compatible with older players, albeit without the benefits of HRA.

Interestingly, post-CD technologies did not gain widespread acceptance the way 78s, vinyl, cassettes, and CDs successfully did, and I suspect two factors were at work. First, a significant portion of the listening population cannot detect differences between HRA technologies’ fidelity and that of CDs they were intended to replace, and this is consistent with their success in the audiophile market but not the general market. Second, the advent of streaming services and music downloads made a huge selection of music readily available to consumers, and the ability to play these MP3 files exists in all basic and smart phones, those devices having overtaken portable MP3 players.

Other efforts in the pursuit of greater fidelity have capitalized on advances in computers’ digital processing speed and storage capacity. Instead of introducing separate technologies, the focus has been on improving audio files in a way that doesn’t necessarily require new equipment. HRA has become available as lossless audio files that modern devices can already handle and in formats that consumers are already familiar with, such as WAV and AIFF. The remainder of this article will focus primarily on HRA files.

What Constitutes HRA?

The article on Digital Audio Resolution introduced bit depth (which determines the number of possible amplitude values a digital sample of an analogue wave can take on) and sampling rate (the frequency at which digital samples are obtained) as the two factors determining audio resolution. The published standard for music CDs is a bit depth of 16 and a sampling rate of 44.1 KHz (16/44.1 for short). Generally speaking, HRA is obtained by increasing the bit depth or sampling rate, usually in conjunction. Doing so yields a more accurate representation of the original analogue signal since there are more samples per unit time and more discrete values that each sample can take on. (Note that this applies to high-resolution technologies like HCDC, SACD, and DVD-A as well as audio files.)

As has occurred in the evolution of recording and playback technology, though, development of standards lags behind the technology, and this describes the current state regarding HRA files. Presently, any audio file with a bit depth greater than 16 (e.g., 20, 24, or 32) and a sampling rate higher than 44.1 KHz (e.g., 48, 88.2, 96, 192, or 384 KHz) can be called HRA.

There have been at least two attempts to introduce a standard, or at least a convention, for HRA. RIAA adopted a modest 20/48 convention, as these lower values facilitate streaming material that could be labeled as HRA. In contrast, the Japan Electronics and Information Technology Industries Association (JEITA) offered the following specifications as requirements for “Hi-Res AUDIO”:

Analogue — Microphone, amplification, speaker, and headphone performance must be capable of at least 40 KHz.
Digital — Recording format, input and output, digital signal processing, and digital-to-analogue conversion must be capable of a bit depth of at least 24 and a sampling rate of at least 96 KHz.

Any use of the Hi-Res AUDIO logo (Figure 1) on products requires meeting or exceeding JEITA’s criteria.

Audio enthusiasts tend to consider 24/96 as the minimum for HRA, which is consistent with JEITA’s specification for Hi-Res AUDIO. It’s also what Play It Again, Paul, LLC uses for customers seeking high-resolution audio files.

Figure 1. High-Res AUDIO logo, to be used only when audio meets JEITA’s definition.

What’s Needed to Play HRA Files?

Unlike the post-CD technologies that required new equipment, HRA files are generally playable on the current configuration of your computer and software. You may be looking at some upgrades, though, but since formats like WAV and AIFF are generic, you’d be looking at increasing capacities rather than getting into entirely new technology. One thing to consider is that the size of an audio file increases proportionately as the bit depth and sampling rate increase. For example, a 40 MB 16/44.1 audio file (about four minutes) would increase in size by 24/16 = 1.5 with a bit depth of 24, and by 96/44.1 ≈ 2.18 with a sampling rate of 96 KHz, combining to yield a 24/96 HRA file of 40 MB x 1.5 x 2.18 ≈ 131 MB. Consequently, you may need a larger hard drive, a high-capacity flash drive or SD card, or a portable player or phone with more storage space.

Before investing heavily in HRA files, check the specifications of your sound card, the components of your sound system, and your playback device. You may not need to upgrade, but the maximum bit depth and sampling rate may limit the HRA files that you can play. For example, my sound card and amplifier can handle digital audio up to 24/192, so playing a 32/384 file would require upgrading both components.

Is HRA Better Than CD Quality?

This question is the subject of a continuing and sometimes contentious debate, and the answer is far from definitive. Let’s first examine the resolution parameters.

Using a bit depth of 24 instead of 16 provides 256 times as many values that each sample can take on. Intuitively, since the 24-bit samples will be closer approximations of the original continuous signal, the analogue signal reconstructed from the digital samples should also be closer to the original. While I’ve seen “counterarguments” that dismiss a bit depth of 24 as unnecessary or describe 16-bit as “mathematically perfect” without any elaboration, I have yet to come across convincing arguments for why 24-bit isn’t better.

Using a sampling rate greater than 44.1 KHz is of debatable value. According to the Nyquist-Shannon sampling theorem, only two digital samples per wave cycle are needed to accurately reconstruct an analogue wave; a faster sampling rate doesn’t increase the accuracy; and a sampling rate of 44.1 KHz is more than adequate to capture 20 KHz, the highest frequency that normal human hearing can detect. Nevertheless, some HRA proponents suggest that digitizing hardware with other than very tight tolerances could introduce distortion that would violate assumptions underlying the Nyquist-Shannon theorem, and that these violations are avoided or at least reduced by using higher sampling rates. On the flip side, HRA opponents suggest that playback equipment with other than very tight tolerances can introduce distortion in high sampling rate audio that, while originating in the ultrasonic content, can be detectable within the audible range.

Let’s give a tentative edge to HRA based on bit depth but call sampling rate a draw. What about the differences in the perceived audio quality of HRA vs. music CDs, which is really a more important question?

Many HRA proponents claim that they can hear differences between HRA and CD-quality recordings, both in the studio and in released media, though the differences are frequently described using subjective terms such as “warmth” and “texture.” Judging from what I’ve read and individuals I’ve spoken with, I believe differences between HRA and CD quality can indeed be detected by some people with undamaged and well-trained ears, notably audiophiles, recording engineers, and musicians, and dismissing their claims would be unfounded.

Those who disagree with claims of perceivable differences often cite a 2007 study published by the Boston Audio Society. In this double-blind experiment (where neither the participants nor the technician administering the experiment knew the properties of the music clips being played), participants—both trained professionals and untrained amateurs—were asked to judge whether what was played to them was HRA or CD quality. There were 276 correct responses out of 554 trials—49.8%, which is indistinguishable from flipping a coin, and this held true regardless of the participants’ backgrounds.

Where does this leave us? In evaluating research studies and individuals’ claims on both sides of the question, one must acknowledge the effects of confirmation bias, both in judging one’s favorite medium as the better because that’s what one believes, and in cherry-picking results from individual studies that are consistent with one’s preference. One must also note that not all research is published in refereed journals; some studies employ deficient experimental designs (e.g., absence of comparable equipment used to play the different media, using other than a double-blind procedure, or using HRA and CD-quality clips that were created from different masters); and many studies arrive at conflicting conclusions.

In the end, questions about HRA’s effect on the perceived quality of audio recordings remain, and so the debate will go on. Ultimately, the deciding factor is whether you believe you can detect better quality in HRA.

Should You Pursue HRA?

The answer depends on your ears, your listening habits and environment, and your equipment.

Consider pursuing HRA if:

You can hear the difference between CD quality and HRA.
- Consider testing whether you can detect a difference between CD quality and HRA before investing in HRA.
- You might start your self-testing by listening to a CD track and the same track ripped or converted into an MP3 file with a bit rate of 128 Kbps.If you can’t detect a difference between a lossless CD track and its lossy, compressed MP3 counterpart, you’re less likely to detect a difference between a CD track and its HRA counterpart.
Greater musical fidelity is important to you.
You tend to engage in serious listening, free of distractions, and in a quiet environment or with headphones (or you want the ability to do so).
You have equipment that can handle the higher bit depth and sampling rate and is of sufficient quality to represent HRA’s increased fidelity faithfully, as well as having sufficient storage space (or you are prepared to upgrade your equipment as necessary).

Consider sticking with CD-quality audio if:

You cannot hear the difference between CD quality and HRA (which is OK, since a significant portion of the population can’t).
You are satisfied with CD quality.
You primarily engage in casual listening, possibly with distractions (including reading, studying, playing games, driving, or using a smartphone), or usually not in a quiet environment.
Your equipment cannot handle HRA’s higher bit depth or sampling rate, is incapable of realizing HRA’s fidelity increase potential, or can’t accommodate a library of large audio files (and you do not want to spend money on equipment upgrades).

Before investing in HRA, you might consider a couple of other options. One is to invest in good headphones or speakers. A CD-quality audio file—or an MP3 file, for that matter—will sound better when played through good headphones or speakers than an HRA file through average or bargain headphones or speakers.

Another is to adopt a wait-and-see approach. Over time, sound quality will continue to improve as new technologies are introduced into both the recording and playback chains, and as more effective algorithms are introduced for both analogue-to-digital and digital-to-analogue conversion, digital filters, and encoding and decoding, to name a few. As these improvements manifest themselves in the consumer marketplace, the quality of the sound recordings you purchase should rise without requiring a significant investment in new equipment. Whether the gap between CD and HRA quality will narrow or HRA quality will also benefit remains to be seen, but either way you end up with better sound.

Back to Paul’s Blog & Contents