Sounds of science: Japan leads push for high-res audio

October 5, 2014 by Katie Forster
A model displays Japanese electronics giant Sony's portable digital audio player 'Walkman A series', designed to play the quality of CD audio, so-called 'high-res audio', in Tokyo, on September 25, 2014

A quarter of a century after the Walkman made music portable and turned Sony into a household name around the world, the company is hoping to set the agenda again with "high-res" audio.

The Japanese tech giant, which recently warned it was expecting a $2.14 billion annual loss this year, is betting that of superior quality to compact discs is about to leave the niche world of audiophiles and go mainstream.

"It's exactly one year since we put the first high-resolution audio players on the market and they have been very popular," Sony's head of audio Ichiro Takagi told journalists last week.

Now the firm wants to "push the accelerator on the high-res product line", he said.

Audio purists have long complained that digitised music has to be compressed so much to fit into the standard mp3 file format that it sounds far removed from how the musician or studio engineer intended.

A lot of data is lost in the compression, especially when compared with analogue formats like the vinyl record—which is enjoying a revival despite the ubiquity of music on smartphones.

In contrast, high-res digital music has a sampling frequency—the amount of times every second that the data is encoded—around four times that of a CD.

This means that more musical detail is captured, resulting in a far richer sound—provided the player is able to handle the higher quality, which most smartphones today cannot.

"It's closer to the original recording and technically superior to a CD," said Masanori Sugiyama, a Sony audio developer.

High-res audio products may have only been available in Japan for one year, but they already account for over a fifth of Sony's audio sales in Japan, the company says.

A customer checks out digital audio players of Japanese electronics giant Sony, at a shop in Tokyo, on May 13, 2014

Now it is aiming to increase this figure to 30 percent by next spring.

Sony's worldwide audio and video sales totalled around 400 billion yen ($3.9 billion dollars) last year. It is aiming for 20 percent of global audio sales to come from high-res products "in a few years", it said.

And the company believes there is no shortage in Japan of customers prepared to shell out 75,000 yen ($700) for a top-of-the-range NW-ZX1 Walkman.

Only for audio snobs?

Sceptics say that the future success of high-res audio faces a number of problems, from the huge size of the files—typically 10 or 15 times bigger than mp3s—to the lack of an industry-standard format.

Some argue that for most listeners, the mp3 provides more than enough quality, without gobbling up hard-disk space or bandwidth.

Colby Leider, director of music engineering at the University of Miami's Frost School of Music, dismissed the hype about high-resolution sounds as "snob appeal".

"Some people can tell the difference... But if it's a great song, you are still going to love it even if it's not HD, and if it's a bad song, it doesn't matter," he said.

Sony might be the biggest company to take the high-res plunge, but they're not the first, with websites such as and smaller, specialist high-res audio makers like Astell and Kern already catering to discerning music fans.

Models display Japanese electronics giant Sony's digital audio player 'Walkman ZX-1' (L) and headphones 'MDR-1', as Sony unveils new series of high quality audio products, in Tokyo, on September 26, 2013

Some buyers are waiting for a device being developed by musician Neil Young, who is expected to release his own high-res portable player called Pono later this year.

Sony's Sugiyama insisted that high-resolution technology is ready to break through the niche barrier, due to the widespread availability of faster Internet connections and affordable hard drives and memory cards.

Indeed, the growing mainstream popularity of higher quality headphones points to a potential market—a third of headphones sold by Sony in Japan now cost more than 10,000 yen (91 dollars).

Taku Kurosawa, who works for Japanese high-res audio download site e-Onkyo, said that when the service was launched nearly 10 years ago there was only a small market for top-quality audio.

A model displays Japanese electronics giant Sony's hard-disc audio player system 'HAP-S1' and speaker system 'SS-HA1', in Tokyo, on September 26, 2013

Yet in the past two years, as Sony and other companies have started preaching the superiority of high-res audio, business has picked up.

"I think (order numbers) will continue to rise. In the future high-res audio will become standard," he said.

For now e-Onkyo only operates in Japan, but it plans to go global by the end of the year.

Kurosawa said: "Once you listen to high-res , you won't go back to CDs."

Explore further: Music lovers seek to pump up digital audio quality

Related Stories

Review: High-res music player ups the audio ante

December 19, 2012

It's hard to make headlines with a portable music player these days. It's old hat by now to carry around thousands of songs in your hip pocket, whether on an iPod or a smartphone.

Apple's iTunes to sell rival Sony's Japanese songs

November 8, 2012

Sony said that music by its Japanese artists was now available on Apple's iTunes store, in an apparent strategy shift by the Japanese firm to cash in on soaring demand for online music.

California company seeks audio rebels

February 17, 2014

High Fidelity isn't just the name of a movie. For the better part of two decades, consumers have been willing to trade audio quality for convenience, forgoing their high-fidelity home stereos for white earbuds. But a Sacramento ...

Sony surprises with first quarter profit

July 31, 2014

(AP)—Sony Corp. reported a surprise eightfold jump in quarterly profit Thursday as sales got a perk from a cheap yen and its bottom line was helped by gains from selling buildings and its stake in a video-game maker.

Recommended for you

Physicists discover new class of pentaquarks

March 26, 2019

Tomasz Skwarnicki, professor of physics in the College of Arts and Sciences at Syracuse University, has uncovered new information about a class of particles called pentaquarks. His findings could lead to a new understanding ...

Study finds people who feed birds impact conservation

March 26, 2019

People in many parts of the world feed birds in their backyards, often due to a desire to help wildlife or to connect with nature. In the United States alone, over 57 million households in the feed backyard birds, spending ...

Matter waves and quantum splinters

March 25, 2019

Physicists in the United States, Austria and Brazil have shown that shaking ultracold Bose-Einstein condensates (BECs) can cause them to either divide into uniform segments or shatter into unpredictable splinters, depending ...


Adjust slider to filter visible comments by rank

Display comments: newest first

Oct 05, 2014
This comment has been removed by a moderator.
4.5 / 5 (2) Oct 05, 2014
The complaint about Hi-res audio taking up too much space contained some validity in 2005 and earlier. Today TeraByte HD cost about $60 (or less). Assume that uncompressed audio album generally fits on a 700 Megabyte CD (or less). This equates to 1000 plus albums of uncompressed music on a single terabyte drive. If you are loading CDs into iTunes or some other digital player and you are compressing the music, you are either an atonal cretin and/or an idiot who only cares to impress his social net 'friends' with the 20,000 pieces on his playlist. If you have 1000 legitimate music CDs, the cost would be in the $10,000+ range. At that price point, worrying about the $60 for a drive is rather myopic.
5 / 5 (1) Oct 05, 2014
'Portable' and 'Hi-Res' will never make sense together. External noise and almost assuredly cheap 'speakers' or headphones make top quality audio pointless.

At home, just use CD's encoded in Free, Lossless Audio Codec called FLAC. CD's bit depth and sample rate can flawlessly record what the human ear can hear. You don't need the dynamic range that will make you deaf or to encode sounds only dogs can hear. BY FAR, the major limiting factors are the quality (or lack-of) in the original recording and the speakers/room you're listening in. NO bit/sample rate can fix something recorded badly in the first place., and few who aren't already audiophiles have excellent speakers properly set in a carefully treated room (though top quality headphones can be a decent workaround to some extent).

It's funny that this article about Sony doesn't mention 'SACD' at all -their long-running, flawed 1-bit, failed attempt to 'own' the high end audio market. The term 'Hi-Res' doesn't mean anything. Silly.
4.3 / 5 (3) Oct 05, 2014
rgw, you wrote, "-Today TeraByte HD cost about $60 (or less)-"

The article is largely about PORTABLE audio. That pretty much HAS to mean solid state and you can't put 1TB into a pocket-size device -much less do it for $60.

It's ironic that you say people 'compressing' music are being myopic, atonal and/or idiots. You obviously don't realize that compression INCLUDES lossless codec options. Also, a 320 lossy MP3 can be, in most cases, damn near lossless. And typical 192-256K lossy MP3s in a portable player are effectively flawless IF you're anywhere with other noise and/or below audiophile-level system quality. I think you're being myopic. Remember... the human ear is 'lossy'.

It's like the push now to sell ULTRA HD tvs when, at best, the human eye can't see any difference at any reasonable viewing distance over already incredible 1080P rez. But customers walk right up to the sets in a store and go 'WOW!' It's just the 'new trick' after 3D flopped.
1 / 5 (2) Oct 05, 2014
Anyone care to explain why quadrupling the sampling rate increases the memory requirements by a factor of 10?

I'm not familliar with how much of anything to do with audio recording is actually done, but if you take samples more often, why not take samples in parallel, with half-time-unit offets, and so on, which would more closely resemble analog?

Like if you took 4 sets of samples (per mic input) and they are offset as 1, (1/2), (1/4 each way), (1/8 each way) of the time stamp, then as the offset approaches zero, the limit of the digital data approaches the exact same as the analog curve.

This seems to be a more powerful technique than simply allowing multiples of samplings in a single series: a normal calculus limit. Instead, you take fewer samples per series, but have several series of samples offset by fractions. Maybe a music expert could conclude that the offset should be some specific harmonic(1, 1/2, 1/3, 1/4, 1/5, etc,) but I used the geometric series as an example.
1 / 5 (2) Oct 05, 2014
Maybe they already do these sorts of techniques, as I said, I wouldn't know. It just seems to me that having multiple series of input can improve quality in ways that simply increasing the number of samplings per series cannot do alone.

Both geometric series and harmonic series are important in music, so I think the best approach is to use some combination of both.


Octuple the base sampling rate, but to correct for errors on the edge of samplings (the boundary) use the prime harmonics, because they over-lap the edges: 1/3, 1/5th, 1/6th(not prime, but 2/3 of samples overlaps breaks in 1/8th,) 1/7th.

Data storage requirement would go up by a factor of ~29, but it would simulate analog continuity much better than any one series of sampling rates of 29 or less.
not rated yet Oct 05, 2014
It's almost like they have never heard of Superaudio
not rated yet Oct 05, 2014
24 bit depth at 96kHz sampling rate is about as good as the Human ear can get. The only reason for greater resolution and depth is to maintain resolution during the digital processing stages of mixing. 24/96 will give you roughly an increase of 3x data storage requirements over CD
not rated yet Oct 06, 2014
It's like the push now to sell ULTRA HD tvs when, at best, the human eye can't see any difference at any reasonable viewing distance over already incredible 1080P rez.

It's not that simple.

if you really wanted to push the limits of human vision, you need to consider the actual information density of the display, which is always lower than the number of pixels on a screen. 1080 vertical pixels can represent 540 distinct lines, for example, and because the lines are not necessarily perfectly horizontal, one has to reduce the number by a factor of 0.7 which is known as the Kell factor.

That means your "HD" TV can accurately display about 380 actual distinguishable lines. The video content is essentially blurred by that much at the source, so that it wouldn't exhibit pixelation and other aliasing errors when displayed.

If you increase the resolution by about 4x, you increase the number of lines to the point where they're just barely distinguishable from one another.
not rated yet Oct 06, 2014
In fact, what most people percieve as "crispness" in a typical 1080p HD screen is actually an artifact or an illusion caused by the square edges of the pixels. Similiar to how people with vinyl records mistake the high frequency noise caused by the dragging needle for "clarity".

If you view 1080p video material upscaled to a large TV with significantly more resolution, it becomes apparent how blurry the material actually is, and crucially, how much information is lost in the compression, when pixel-level artifacts no longer hide pixel-level defects in the image.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.