Why High-Fidelity Streaming is the Audio Revolution Your Ears Have Been Waiting For (forbes.com)

From a report: While our ears may be attuned to lossy compressed audio in most everyday scenarios, the experience of rediscovering high-fidelity lossless digital audio can be nothing short of a revelation. Fine details reappear, performers have more space, sounds have more definition, audio feels warmer, sounds clearer, and is noticeably more pleasurable to listen to. The higher you go with audio file resolution, the better it gets. Thanks to the new range of streaming apps delivering CD-quality or higher, our beloved "universal jukebox" is undergoing a significant upgrade.

Consumer demand for high-resolution audio has been growing steadily, for example users of Deezer HiFi have increased by 71% in the past 12 months alone, and the product is now available in 180 countries and works with a wide range of FLAC streaming compatible devices. Bang & Olufsen's most senior Tonmeister (sound engineer) Geoff Marti believes that demand for hi-fi streaming audio is growing due to a rise in the number of people buying high-end audio devices. "It used to be that you bought an iPhone and you used the white earbuds, but nowadays people are upgrading to better headphones, so they want a better file and a better app to play it on. The potential is there for somebody that wants to get high quality, and they don't have to spend a lot of money to get it."

Comments

Quasi-religious nonsense

[gweihir]

Except for dynamics (which the compressed formats solve), CD audio is way beyond the quality most people can hear. For some reason, a lot of people fall for the scam and pa a lot of money for things that do not at all improve audio quality, like this one here, audio cables for hundreds of dollars, or even very expensive audio-Ethernet cables (which is so far beyond stupid it is staggering). I am sure this scam will also be able to separate victims and their cash.

Re:Quasi-religious nonsense

[DigiShaman]

It's called the loudness war for a reason. The best raw source for music will always been CDs prior to the year 2000. Every re-release after will have been compressed (audio, not digital/mathematical).

I'm all for 24-bit audio so long as it doesn't suffer from compression, otherwise a giant waste of time and money.

Re:Quasi-religious nonsense

[Anonymous Coward]

I bought a sony CDP-101 when they first came out. The thing about CD's back then was they painstakingly tried to get everything perfect to showcase the dynamic range. I recall one classical CD I have where I can hear the conductor hit something with the baton. Telarc would find ultra quiet mic preamps. I could hear the HVAC system on some. Now it is squish everything up to 11. It is sad really as the technology of today would allow for fantastic realism with zero compression.

Re:Quasi-religious nonsense

[Luthair]

Maybe relevant for you - https://people.xiph.org/~xiphm...

Re:Quasi-religious nonsense

[Anonymous Coward]

I bought a sony CDP-101 when they first came out. The thing about CD's back then was they painstakingly tried to get everything perfect to showcase the dynamic range. I recall one classical CD I have where I can hear the conductor hit something with the baton. Telarc would find ultra quiet mic preamps. I could hear the HVAC system on some. Now it is squish everything up to 11. It is sad really as the technology of today would allow for fantastic realism with zero compression.

Exactly!

I have found it absolutely ridiculous that the trend toward ultra-compressed "Everything Louder than Everything Else" started AFTER we finally got a playback medium that covered 95% of the dynamic range of human hearing.

Re:Quasi-religious nonsense

[Anubis IV]

You claim DTS sounds "way better" and has "more fidelity", but what's the basis for that comparison? So far as I know, no albums have been mastered using DTS Surround (or DTS-HD Master Audio and the like), so you couldn't have done any A/B testing from the same material, which leads me to believe that you declared a winner after watching some random movies and listening to some random CDs. I imagine that's exactly the sort of quasi-religious nonsense that the OP was railing against.

To be fair, DTS Surround (i.e. what I assume you're talking about, since it's the standard DTS codec, as opposed to the extensive list of other DTS technologies that support 5.1 channels) does have better fidelity than an audio CD (24-bit at 48 kHz vs. 16-bit at 44.1 kHz), but the OP was saying—and as someone with decades of experience mixing sound (though not professionally, lest anyone think I'm an expert), I'm inclined to agree—that most people can't tell the difference. Blind tests have shown repeatedly that most people can't reliably pick the better one between lossless and a 128 kbps MP3, and among those that can, only a vanishingly small number can still pick the better one consistently once you bump it up to 192 kbps for the MP3. I'd wager that the number who could pick the better between DTS Surround and CD audio would be similarly small.

That doesn't mean people have defective ears; it just means there are limits to what we can perceive. Just as the printer dpi wars became meaningless once we got beyond the human eye's ability to perceive a difference, so too did the camera megapixel wars eventually become meaningless, so too did the display ppi wars become meaningless (despite ongoing marketing), and so too have these audio fidelity wars become meaningless.

I mean, seriously, where do you think that CDs are coming up short? They're already capable of a larger dynamic range (~90 dB) than what you can get in a concert hall (~80 dB), they already capture frequencies (0 Hz to 22.05 kHz) that are both below and above what people can hear (20 Hz to 20 kHz), and they already have enough detail that the vast majority of the population is incapable of picking the better audio at better than chance would allow.

You're welcome to have preferences, of course. A lot of people love pumping up the bass (see Beats headphones). Maybe you prefer the "warmer" sound that's popular among the audiophile crowd (hence why it's become known as the "audiophile" sound). Maybe you prefer a brighter or punchier sound. But with any of those preferences, you need to be aware that you're actually reducing the fidelity by moving away from what was originally there, in much the same way that adding cream or sugar takes you away from the original flavors in your tea or coffee. For my part, I've generally leaned towards a "reference" sound (i.e. as close to the original as possible), but I'm weird that way, since most people find it unpleasant to listen to and end up suffering ear fatigue as a result.

Re:Quasi-religious nonsense

[tlhIngan]

I'm all for 24-bit audio as long as the hardware is of sufficient quality that it can make any conceivable difference, and if it doesn't cost a lot more. Otherwise, CD quality is at least adequate, if not ideal. I'd rather have more tracks and more channels than more bits (and with the ability to assign the former to the latter dynamically.)

The dynamics of a 16 bit converter are sufficient for practically any real world situation, listening OR recording.

The average living room will have around 40dB of dynamic range - it's got a relatively high noise floor.

Even an sound chamber will rarely get you more than 80dB of dynamics.

A 16 bit converter has 96dB of dynamics - and your recording equipment, be it tape (90dB tops, -3dB per generation copy, so after a backup copy for safety, you'll be mastering from an 87dB, producing a 84dB "master" which you replicate at 81dB, for an all analog path). or other equipment (microiphone, etc) will generally have far less dynamic range.

Granted, to get full hearing range is around 120dB or so (a 20 bit converter) though the situations involving such large dynamic range in volume is rather limited practically.

And to DSD fans with their "1bit" converter, well, at 6dB, all DSD did was push the noise above 22.05kHz. (You can tell when you have a DSD recording that's improperly filtered as you get normal audio below 22.05kHz (1/2 44.1kHz, which is the equivalent sampling for DSD running at 2.something MHz) and a brick wall of crap above 22.05 on a spectral plit).

This is claptrap

[fliptw]

The same people clamoring for FLAC because of audio quality are also the same people snapping up vinyl and cassettes, and probably have already wrecked their hearing past the point of being able to tell the difference. High-end Audio is a bunch of snake oil.

Speaking of phones ...

[UnknownSoldier]

Has anyone compared wired to wireless quality? Are there any blind A/B tests comparing the built-in DAC of the iPhone using wired headphones vs wireless? And also the built-in DAC with an external DAC?

i.e.
I'm wondering how much better quality a dedicated DAC and/or DAC+AMP is such as the Schiit Modi 3 (DAC) + Schiit Magni 3 (AMP) ?

Back on Topic: There is a reason us audiophiles ripped everything to FLAC in the first place. So we would never have to re-encode it. The problem is Apple pushed their own proprietary lossless format, ALAC instead of embracing open standards such as FLAC.

Good to see streaming services finally embracing FLAC.

Also, could one of the editors at least PLEASE fix (*) this clickbait: The higher you go with audio file resolution, the better it gets.

It should read: The higher you go with audio file resolution, the better it gets, with decreasing returns.

i.e.
I doubt most people could tell the difference between variable 320 kbps (kilobits/s) and CD quality even with quality headphones for most music -- unless it is Classical or Jazz.

(*) Yeah, yeah, I know the editors have been a joke around here for ~20 years.

Re:Speaking of phones ...

[FrankSchwab]

>>> I doubt most people could tell the difference between variable 320 kbps (kilobits/s) and CD quality even with quality headphones for most music

Well, back in the day when we were encoding with pirated copies of the Fraunhofer codec, I actually tested this. I created an audio CD with 4 sets of DDD tracks - one classical, one Rolling Stones, one solo piano, one something else. Each set had five tracks - the first track was the uncompressed CD-rip, and following this (in a random order that only I knew) were another copy of the uncompressed CD-rip, 96 kbps, 128 kbps, and 256 kbps CBR MP3 encoded then decoded tracks. I handed out disks to a dozen of my engineering coworkers, and asked them to take them home, put them in the cd-player on their high-end stereos and come back and tell me what the order of the tracks were.

It was comical. Half of them didn't even guess, because they admitted that it would be a random guess. Almost everyone could identify the 96kbps track, but no one could tell the difference between 128, 256, and uncompressed. One guy hooked it up to his home oscilloscope and spectrum analyzer - he noted that he could easily identify that the tracks were different, but he couldn't identify which was which except for the uncompressed one - he could see on the waveforms that it was identical to the uncompressed first track.

Now, I fully believe that it's possible for some golden-eared listeners to be able to tell 128kbps from flac - and I believe that it's possible for some to train themselves to tell the difference (though I don't know why you'd want to torture yourself for the rest of your life by doing that). But my ears in my early 30's couldn't tell the difference, and my ears now can't tell the difference, so I'm really happy playing my music through whatever electronics I happen to have around, although I am willing to pay for good speakers because those I can tell a difference.

Re:Most people can't tell the difference in A/B te

[lsllll]

While that may be true, in my experience most people don't care about the loss in the lossy compression, because they don't listen on anything that can portray the difference anyways. This more and more people are buying high-end HiFi equipment, while may be true, is not due to their interest in high fidelity music. After all, "Kanye's" music is crap to begin with. It may have to do with population growth, the price of electronics having come down, etc.

Not sure what they mean by HiFi equipment anyways. The most important pieces of an audio setup are the source material and the speakers. Everything in between does a descent job of handling the signal in most cases. But you can't buy a Sony A/V receiver and call it HiFi. I don't see ANY of my friends spending anywhere near what I paid for my audio setup, which is actually very modest and all second/third hand to begin with. Like I said, the most important parts for me were the speakers, and my NHT 3.3s cost $1200/$2000 I spent on my setup (excluding the source materials). You can get a very descent DAC to piggy back on top of a Raspberry PI for a complete setup of

My friends, however, don't do any critical listening to begin with. As such, a bluetooth speaker at home does just fine for them. If they want to sit and listen to something, they'll most likely do it on their 5.1 A/V receiver that has little satellites and a subwoofer. At that point, playing a lossless FLAC vs. playing a 192Kbs MP3 doesn't make a difference to them.

At the end of it all, though, is whether you get enjoyment out of whatever you have in front of you, whether it'd be your car, your spouse, your job, or your stereo. If your car drives fine for you, then that's all that matters. If the music coming out of your stereo sounds good to you, that's all that's needed. After all, some other person will look at my audio setup and laugh, because they believe they have higher quality audio coming out of their speakers than I do, and that's fine, because that's what makes them happy, and my setup is what makes me happy.

Re:Most people can't tell the difference in A/B te

[lsllll]

I can't attest to use of amplifiers in scientific instrumentation, but performance is not everything, specially when it comes to audio applications. Yes, it's great that most class D amps reach 90% or higher efficiency, but that comes at a price.

Class D amps achieve their efficiency by turning the transistors completely off when not in use, as opposed to class AB which one of the transistor sets are on at all times or class A where the output transistors are on all the time. The switching off of the transistors is controlled via Pulse Width Modulation. This is the same concept used in most power supplies today, from PCs to phone chargers to LED bulb replacements for incandescent bulbs. This control can be via a digital circuit or an analog circuit. The digitally controlled circuit introduces too much error and distortion to be usable in audio applications. The analog controlled class D amps have historically been pretty hard to design correctly. They have complicated circuits and have mostly been non-linear in their reproduction of 20-20K Hz spectrum, something audiophiles strive really hard to achieve. I realize there have been new advents in overcoming these issues, but these usually come at a high price. A well-designed class D amp costs many times that of a well-designed class AB amp. Just look at the class D amps that are on the market and targeted to audiophiles. By comparison, I can pick up a used Aragon 4004 MKII for $500-$600 on ebay and be done with my amplifier needs, although my own amp is an ATI 1502 which can be had for even cheaper. These class AB amps provide completely linear audio amplification of their input signal at a fraction of the price of a comparable (in terms of quality) class D amp.