What's special about expensive "audiophile" amp/DAC?

[Porama6400]

Speaking about electronically, I am just curious (and skeptical) about what's so special and why expensive one "sounds better"

I get it why it would be few hundred buck or so for long lasting as they would need non CapXon garbage stuff but for a grand or higher, that seems absurd to me.

isn't flat response Amp or DAC something we already do all the time?

[james_s]

In practice there is nothing really special, it sounds better because the placebo effect is real. Some of them also look quite nice, subjectively. It is futile to go looking for logic in the audiophile world.

[Gyro]

Although attention to detail and competent design will get you a long way. It's not so much about the Silicon used, it's more about getting the best out of it. Noise, ground layout, PSU rails etc.

[SiliconWizard]

Nice clickbait for the audiophool busters!

It must be maybe the 1 millionth thread on the topic or something.

Speaking about electronically, I am just curious (and skeptical) about what's so special and why expensive one "sounds better"

Being skeptical is always a good thing.

isn't flat response Amp or DAC (...)

Define "flat response" first maybe?

(...) something we already do all the time?

Not really. It's more like: we design crap all the time, and only very occasionally do better for precision stuff. Now is this crap "good enough"? This is where this kind of topic will never end.

I get it why it would be few hundred buck or so for long lasting as they would need non CapXon garbage stuff but for a grand or higher, that seems absurd to me.

We all know some products are ridiculously expensive, and yes, sometimes for no good reason, but judging the price of a product by what it contains doesn't make much sense either, unless we're talking about "commodity" stuff - including potatoes.

So price is not the issue IMHO here. The only valid issue is when a vendor sells stuff with claims that are just not *met*, which happens quite a bit in many areas, not just audio (but audio does indeed have a lot of this.) Now if some vendor sells a very expensive DAC, for instance, which a claimed THD+N a magnitude or more lower than what is usual in GOOD audio DACs, and this figure is actually met, I don't see a problem. If you think you need it, buy it, if you don't, fine, just buy something else, and let the professional debunkers, suddenly becoming hearing experts, spend hours explaining why such low THD figures don't make sense.

To sum it up: price is not a valid point (only very few goods these days have an "intrinsic" value strictly speaking anyway), false claims on objective figures are bad and usually punishable by law, and true claims on apparently ridiculously "good" characteristics (such as ultra low THD+N), I'll let you make your own opinion, and spend hours debating what is good enough and what isn't - don't worry, you'll find thousands of people to discuss that with you.

[Zbig]

Your gut feeling is not failing you: it's all about delusion and raising your self-esteem by misguided feeling of belonging to some imaginary group of "elite" golden-ears. Also, including mandatory "Signature" in your bullshit product name and mentioning "quantum alignment" at least once in your marketing blurb. Compared to what we are now capable of achieving using our current technology, reproducing an electric signal of up to 20 kHz and not messing it up in the process is nothing really. Putting this in the perspective and comparing to the real challenges of modern tech, it's practically DC. We're already there since quite some time.

If you're asking not just theoretically and are actually in the market of a decent computer audio interface, without breaking the bank and having to endure all the drivel, you can't go wrong with established go-to manufacturers of mid-range gear aimed at musicians such as Focusrite's Scarlett line.

[bsdphk]

Actually, it's all and only about showing how rich you are: The price is the point, not the sound.

I've met people who are on the selling side of that industry, and they say it is all about their customers being able to show of with expensive and unique bling, that nobody else has, same ball-game as gold-plated toilets.

There are literally people who pay them extra to not sell to anybody else, so that they can say they have "the only one ever made" etc.

All the audiohomoepathic buzzwordery is just to dress up the emperor so he doesn't sound naked when he brags.

[Gyro]

.... and so another re-hashed Audiophool discussion gathers pace.

[grifftech]

lower power efficiency

[Zbig]

.... and so another re-hashed Audiophool discussion gathers pace.

You're right... quick, everyone, let's bring Corona Virus into this!... somehow...

[tszaboo]

Look, an expensive DAC could give you
- more inputs
- gold plated connectors, which is better for corrosion (which definitely happens with hifi gear in decades)
- displays and better controls
- relays on the outputs to avoid pops during startup
- a nice headphone amplifier built in
- high sample rate USB input, which is still surprisingly difficult to do
- Low noise power supply
- isolated interfaces to avoid ground loops and hum
- proper thermal management for the components
- upsampling
- nice reconstruction filters, with components better matched
- aesthetically pleasing look

Most of which is absent from the low end units. It not necessarily results better sound, the same way a supercar doesnt get you to your destination faster in rush hour.  But all these are added value and they cost money. If you want to look for a "Not Another Nonsense Design"* look at the O2 ODAC for example. That is made by an engineer. Or the Cambridge audio Dacmagic.

Whats really the point of " better sound "  anyway, when the record studio run the music though a hundred NE5532?

And of course there will be snake oil salesman out there.

*tm by me.

[NiHaoMike]

It increases your error margin. Unless you have ideal main input and ideal 0dB normalized and well compressed audio input and perfect bang-on amplifier gain setting, you need your DAC to provide you with error margin without degradation.

Audio engineers are always conservative with recording volume levels, since they rather lose a few dB of dynamic range than run the risk of clipping at a critical time. They're supposed to correct the volume during editing, but many don't bother and leave it up to the listener to adjust the volume on their end. (Is it correct that with lossy codecs like MP3, reducing the signal input a small amount does not necessarily reduce the dynamic range?)

[schmitt trigger]

As Dave would say,
the reason these audiophile components are so expensive, is because they are  assembled by nude Swiss virgins in a cottage deep in a secluded forest, where birds chirping under a deep azure sky and a crystal-clear brook provide the unique harmonic content to align the electronic sources for the purest sound reproduction.

Or something to the effect......

[tooki]

It increases your error margin. Unless you have ideal main input and ideal 0dB normalized and well compressed audio input and perfect bang-on amplifier gain setting, you need your DAC to provide you with error margin without degradation.

Audio engineers are always conservative with recording volume levels, since they rather lose a few dB of dynamic range than run the risk of clipping at a critical time. They're supposed to correct the volume during editing, but many don't bother and leave it up to the listener to adjust the volume on their end. (Is it correct that with lossy codecs like MP3, reducing the signal input a small amount does not necessarily reduce the dynamic range?)

Say what now?!? No, the exact opposite has been happening for years now: the loudness war. Tons of recordings are being released with absurdly high levels, often clipped. (And previously-good recordings are being re-released in such a fashion!)

Digital audio, even "just" 16-bit/44.1KHz, already has plenty of headroom to allow for massive amplification without adding perceptible noise. (The dynamic range of CD audio already far, far, far exceeds what a human can perceive safely. The dynamic range is such that the smallest sound on CD, amplified to audible levels, would have a system whose peak volume would cause physical harm.) "Leaving it up to the listener to adjust the volume" is precisely what they should be doing, but aren't. "Turn Me Up" is literally the name of a campaign to get mastering engineers to stop succumbing to the loudness war: https://www.turnmeup.org

[tszaboo]

To utilize the full 100dB dynamic range of human ear plus waveform resolution of 40dB, 24 bit is what is needed to cover entire human ear's range.
Adding some software volume control for lazy users, I can't see why 32 bit is ridiculous.

Hmm, no. http://www.cochlea.eu/en/sound/psychoacoustics
Read the part about masking.

[tooki]

Digital audio, even "just" 16-bit/44.1KHz, already has plenty of headroom to allow for massive amplification without adding perceptible noise.

Are you kidding? 16 bit=98dB, say we allocate 8dB for mastering volume loss and DAC internal headroom, 20dB for volume control, then we only have 70dB of usable dynamic range.
If you listen at 100dB, then your LSB is 30dB. Keep in mind that this 30dB is not the amplitude of a wave, but the minimum step of a signal (a singe step in that wave).
That's way insufficient.

Human ears can perceive 0dB, and that's the level, not the LSB making up the waveform.
To utilize the full 100dB dynamic range of human ear plus waveform resolution of 40dB, 24 bit is what is needed to cover entire human ear's range.
Adding some software volume control for lazy users, I can't see why 32 bit is ridiculous.

Except that you're, well, all wrong. Because of how sampling and dithering works, an LSB is not the smallest signal that can be represented. Consequently, the effective dynamic range of CD audio is not 96dB, but 120dB:

https://people.xiph.org/~xiphmont/demo/neil-young.html

The dynamic range of 16 bits

16 bit linear PCM has a dynamic range of 96dB according to the most common definition, which calculates dynamic range as (6*bits)dB. Many believe that 16 bit audio cannot represent arbitrary sounds quieter than -96dB. This is incorrect.

I have linked to two 16 bit audio files here; one contains a 1kHz tone at 0 dB (where 0dB is the loudest possible tone) and the other a 1kHz tone at -105dB.

Sample 1: 1kHz tone at 0 dB (16 bit / 48kHz WAV)

Sample 2: 1kHz tone at -105 dB (16 bit / 48kHz WAV)

[image]
Above: Spectral analysis of a -105dB tone encoded as 16 bit / 48kHz PCM. 16 bit PCM is clearly deeper than 96dB, else a -105dB tone could not be represented, nor would it be audible.

How is it possible to encode this signal, encode it with no distortion, and encode it well above the noise floor, when its peak amplitude is one third of a bit?

Part of this puzzle is solved by proper dither, which renders quantization noise independent of the input signal. By implication, this means that dithered quantization introduces no distortion, just uncorrelated noise. That in turn implies that we can encode signals of arbitrary depth, even those with peak amplitudes much smaller than one bit [12]. However, dither doesn't change the fact that once a signal sinks below the noise floor, it should effectively disappear. How is the -105dB tone still clearly audible above a -96dB noise floor?

The answer: Our -96dB noise floor figure is effectively wrong; we're using an inappropriate definition of dynamic range. (6*bits)dB gives us the RMS noise of the entire broadband signal, but each hair cell in the ear is sensitive to only a narrow fraction of the total bandwidth. As each hair cell hears only a fraction of the total noise floor energy, the noise floor at that hair cell will be much lower than the broadband figure of -96dB.

Thus, 16 bit audio can go considerably deeper than 96dB. With use of shaped dither, which moves quantization noise energy into frequencies where it's harder to hear, the effective dynamic range of 16 bit audio reaches 120dB in practice [13], more than fifteen times deeper than the 96dB claim.

120dB is greater than the difference between a mosquito somewhere in the same room and a jackhammer a foot away.... or the difference between a deserted 'soundproof' room and a sound loud enough to cause hearing damage in seconds.

16 bits is enough to store all we can hear, and will be enough forever.

[nctnico]

In practice there is nothing really special, it sounds better because the placebo effect is real. Some of them also look quite nice, subjectively. It is futile to go looking for logic in the audiophile world.

Agreed. Recently I recapped an amplifier from a relative. A lot of components (including semiconductors) run really hot (100 degrees C!  ) and the circuit itself is what you learn in the second year of any EE education. Ofcourse a whole bunch of electrolytics right next to hot components.

[tooki]

Because of how sampling and dithering works, an LSB is not the smallest signal that can be represented. Consequently, the effective dynamic range of CD audio is not 96dB, but 120dB:

Assuming you are talking very low frequency, way below Nyquist frequency.
What is true for 1kHz is less true for 5kHz, which is still pretty audible and actually very loud.
And what is true for a single frequency isn't essentially true for a complex waveform.

Nope. As long as the signal is correctly bandwidth-limited before sampling, then the sampled signal reproduces the original bandwidth-limited signal perfectly, all the way up to f_s/2. As long as it's correctly bandwidth-limited, the complexity of the signal does not matter.

As usual, refer to: https://wiki.xiph.org/Videos/Digital_Show_and_Tell

https://people.xiph.org/~xiphmont/demo/neil-young.html

That article has been bashed by multiple HiFi forums, with reasonable points.

Yeah, by people who do not, and don't want to, understand the principles of sampling. Citing hi-fi forums as a source for debunking an explanation of scientific fact is like citing a creationism forum as a source debunking evolution...

Also, while there are person-to-person perception difference. The author of that paper claims 980nm IR is not visible, yet I was able to see 976nm IR laser as a very faint dull red color.
Most people can't see even very near IR such as 808nm or 915nm, which are commonly used in IR cameras for illumination. I can see them very clearly, not just as a dull red dot, but as a bright light source.
Windows Hello IR face recognition annoys me enough, though most people can't even sense the IR source.
Also, I can slightly perceive 365nm UV shined on a matte metal (non fluorescence) surface. Very faintly, but I can certainly see it as a deep and dull purple color.

Unfortunately I don't have golden ears, but from that I can see slightly beyond visible light spectrum I guess it's not that hard to come by someone who can exceed commonly accepted hearing range.

Irrelevant: wider spectrum = wider frequency range. And while it's possible that some people have hearing with an extraordinary frequency range, we aren't discussing sampling rate here. We are discussing dynamic range, which is a consequence of bit depth, not sampling rate.

(As for your laser diode: laser diodes and LEDs produce some off-frequency light, too. Some light from IR diodes can be a tiny amount of shortwave IR or even some longwave visible red. It's highly unlikely you can detect actual 976nm light, it's just too far.)

[edy]

As Dave would say, the reason these audiophile components are so expensive, is because they are  assembled by nude Swiss virgins in a cottage deep in a secluded forest...

You're right... quick, everyone, let's bring Corona Virus into this!... somehow...

Ok how about this for a try... The group of "nude virgins" doesn't have to necessary be FEMALE    What if they are bearded men? And when they are gathered together in such close proximity it could be a breeding ground for coronavirus:



But seriously, another audiophoolery thread.... 

[NiHaoMike]

Say what now?!? No, the exact opposite has been happening for years now: the loudness war. Tons of recordings are being released with absurdly high levels, often clipped. (And previously-good recordings are being re-released in such a fashion!)

Some content hosts like Youtube are fighting it by reducing the volume of content recorded too loud. I'm not exactly sure but they might even end up making it quieter than it would be otherwise in order to really penalize the overuse of dynamic range compression.

[magic]

If you're asking not just theoretically and are actually in the market of a decent computer audio interface, without breaking the bank and having to endure all the drivel, you can't go wrong with established go-to manufacturers of mid-range gear aimed at musicians such as Focusrite's Scarlett line.

I have a soundcard "for musicians" which fails at star grounding and picks up noise from crappy computers like crazy

On the upside, it's an older, discontinued model.

[tooki]

Say what now?!? No, the exact opposite has been happening for years now: the loudness war. Tons of recordings are being released with absurdly high levels, often clipped. (And previously-good recordings are being re-released in such a fashion!)

Some content hosts like Youtube are fighting it by reducing the volume of content recorded too loud. I'm not exactly sure but they might even end up making it quieter than it would be otherwise in order to really penalize the overuse of dynamic range compression.

Apparently, Apple Music (the streaming service, not the song purchasing on iTunes) applies mandatory volume normalization (“Sound Check”), and some in the industry hope that this will help, since it removes any advantage of recording “hot”.

Then again, for many years now Apple has had the “mastered for iTunes” program/label, whereby the studio uploads a special master optimized for playback via 256Kbps AAC. So it’s entirely possible that studios will continue to use a different master for CD as for iTunes. (I assume that iTunes and Apple Music use the same master uploads.)

[nctnico]

Say what now?!? No, the exact opposite has been happening for years now: the loudness war. Tons of recordings are being released with absurdly high levels, often clipped. (And previously-good recordings are being re-released in such a fashion!)

Some content hosts like Youtube are fighting it by reducing the volume of content recorded too loud. I'm not exactly sure but they might even end up making it quieter than it would be otherwise in order to really penalize the overuse of dynamic range compression.

I like dynamic range compression in videos because it makes the voices easier to understand and the sound effects less pronounced.

[tooki]

Well, IMHO, the issue of dynamic range is completely opposite in music compared to videos and music: if music has far too little dynamic range, movies and videos suffer from too much. Action movies have just plain excessively loud sound effects like explosions, and YouTube videos in particular often have terrible sound levels, with music intros and whatnot being wildly too loud compared to speech.

[SiliconWizard]

Looks like the thread pretty much went as I was expecting.

Action movies have just plain excessively loud sound effects like explosions, and YouTube videos in particular often have terrible sound levels, with music intros and whatnot being wildly too loud compared to speech.

Sound mixing and mastering in videos/films is just a completely different thing than with music obviously, and is a skill that has to be learned. The mastering for films projected in theaters should be pretty different from the one made for home viewing (completely different environment, equipment, sound level...) DVDs/Bluerays are usually remastered for this use with varying success.

But Youtube videos? As I just said, this is a skill to be learned, you just don't improvise it and get good results. So of course, most Youtubers haven't learned any of this, and just produce crap. Just like you would play crap on a guitar if you can't play - doesn't matter if the guitar has become cheap to buy and looks easy to use . And given the very varied equipment on which people watch Youtube videos, it's especially tough to get good results for everyone, so it's never going to work if you don't know what you're doing.

[Porama6400]

well, ouch.
Didn't expect this thread to go this way.

I'm just curious whether if there's a good explanation technical wise.
(As based on my very limited experience, I can't think of anything that will make BOM cost that much)

Your guys explanation make sense though, maybe I shouldn't try to find reason about audio stuff