If you had $10k and were starting from scratch…

[Leporello]

You're welcome to visit and hear for yourself. The doubting skeptic who bases everything on how he thinks things *should* work who has never simply just listened to what's out there is classic stuff.

Do you think you're in the right place because you have some responsibility to counter signal people who are telling you, point blank, that they've experienced things that you have not?

I don't deny that there's much snake oil in this industry, but you seem to deny that it's not ALL snake oil. That's a bleak attitude and likely why you have yet to discover anything that undermines this blanket assumption.

Surely the real problem here is that you have not offered any reasons to believe in your claims.

 

[fliz]

I don't know how you arrived at those beliefs, but they're factually wrong.
Not flawed for a headphone, doesn't mean what it means for a DAC. It's more about people lowering their expectations for headphone. Think of a DAC with 0.01%THD, what a piece of crap, right? Imagine now a headphone with not 0.01 but just 0.1% THD in most of the audible range(doing almost inevitably worse at least in the low end so not strictly even achieving 0.1% in the audible range), people will treat it like it's a great piece of gear. It's the same type of harmonic distortions being measured but treated very differently. Not because they matter more upstream, but because we can easily get better than 0.01%, maybe even with a cheap USB dongle, while it's hard to get better in headphone. So we reevaluate the meaning of good.

That's probably how we end up with you being able to have those beliefs about headphones. And this hobby at large being so weird sometimes. With stuff 110dB below music treated as important, but getting the left channel only into the left ear for albums almost always mixed and mastered on speakers, somehow that isn't important to the general concept of fidelity, and it doesn't have to be discussed, or you're a headphone hater. Those weird double standards are so strange.

II would strongly suggest to you and anybody who cares about the real world, to look at measurements more and get, at the very least, some idea of the magnitudes involved before deciding what's likely to be more important than something else. That's how we can try to keep our feet on the ground.



Caveat:
Headphone measurements are heavily limited by placement, calibration, room noises, mics, and if we're serious, for the most demanding, even humidity and temperature. So some measurements of the very best headphones probably don't show how low distortions really go at all frequencies(Or how bad things will be with your system in your room, btw).
But usually we can still see from other measurements(those that don't get drowned by noise) and trends for above noise distortions, how good the headphone generally is. I always lament the lack of distortions measurements beyond THD(which doesn't correlate well with listening preferences), but I'm just one random consumer whining. The industry guys will do whatever they please, as always.

Measurements on paper rarely tell the whole story. We have not solved the problem of how our brains resolve sound.

Transducers either produce a neutral and detailed sound, or they don't. This is much more challenging for a speaker. You're never going to get proper dynamics unless you have huge high efficiency speakers. The weight of the diaphragms start to matter a lot. Inertia is an issue. This is why plasma tweeters exist. We don't have those issues to nearly the same degree with cans. They don't need to fill a room. Cans are comparatively simple when compared to everything upstream. You've got transducers, housing, and some wire. Paying a boatload for those is just silly. Their margins are outrageous. They can afford to spend piles on marketing and sponsoring places like this forum.

Meanwhile, there is a TON more going on w/ the upstream gear. Like I already mentioned before, OEM costs for the parts themselves can easily go into the thousands for the absolute best of the best components. I only mentioned input regulators and output caps before, but there's so much more, not to mention the R&D that goes into designing the circuits, building prototypes, testing, iterating, building again.

It's very commonplace for people to make appeals to incredulity with this stuff. It's simply not a "settled science". We don't know how or why stuff works the way it does. Designers of the very top end gear are still tinkerers with a passion. You can't work backwards from theory to practice and get the same results. You try things out, discover something that works, and it may be a decade or more before theory catches up to explain WHY something works. Do you think cap manufacturers like Mundorf have some deep esoteric understanding of why one type of cap outperforms another? They don't. They just experiment and discover that it does.

Are there people out there who have made a fortune selling overpriced audio jewelry? Yes. Are there people building stuff that walks all over that audio jewelry? Also yes.

 

[Leporello]

It's very commonplace for people to make appeals to incredulity with this stuff. It's simply not a "settled science". We don't know how or why stuff works the way it does..

This is an appeal to unknown possibilities, a well known audiophile fallacy.

 

[fliz]

Surely the real problem here is that you have not offered any reasons to believe in your claims.

I just addressed this in a previous comment. The fact is that theory hasn't caught up to practice. This is how things always work. Tinkerers discover something and it takes a LONG time to understand the underlying principles behind why it works.

Can I explain why one regulator outperforms another regulator when the cheaper one is spec'd out to do the job perfectly sufficiently? No. Is the difference comically transparent when you listen? Yes.

 

[fliz]

This is an appeal to unknown possibilities, a well known audiophile fallacy.

The fallacy here is the appeal to incredulity. I'm suggesting you simply actually listen to what exists instead of dismissing something without ever engaging your senses because an explanation for the phenomenon doesn't exist.

Explanations ALWAYS follow initial observations. It tautologically can not happen the other way.

Here's an example: I paid an electrician $500 to run 10awg wire from my breaker box to a dedicated outlet in my system. I had to argue with him. He insisted it was a complete waste of money. Convincing him was hopeless. He eventually just resigned himself to do it anyway, "It's your money."

I let him listen before he did the work. Afterwards, I let him listen again. He *immediately* said, "You cheated." He was convinced I sneakily made some other sort of change. He had no other explanation for why he heard such an apparent difference.

Ideology can cloud judgement.

 

[Leporello]

I just addressed this in a previous comment. The fact is that theory hasn't caught up to practice.

The problem with this argument is that there is no compelling evidence to challenge our current theoretical understanding. Doubting frend anecdotes do not count here.

 

[Leporello]

The fallacy here is the appeal to incredulity. I'm suggesting you simply actually listen to what exists instead of dismissing something without ever engaging your senses because an explanation for the phenomenon doesn't exist.

The way to go about this is you trying to stand behind you assertions and provide evidence. So far there is none.

 

[fliz]

The way to go about this is you trying to stand behind you assertions and provide evidence. So far there is none.

You're asserting that I must provide an explanation for the observable phenomenon in order to accept that the phenomenon is observed. That's not how this (or anything) works. Explanations always come after observations, and (often) quite a long time after the observation.

How long do you think it took for people to explain why northern lights are observable? To sit somewhere and simply deny the observation because you aren't in a position to observe it yourself until a satisfactory explanation exists is silly.

"You didn't find fossils on both the east coast of South America and the west coast of Africa. You're lying."

<Proof of plate techtonics is discovered YEARS later>

"Ok, you weren't lying."

 

[fliz]

Also, the no true scotsman of "compelling" evidence is absurd.

The evidence is that the phenomenon is observable.

"I didn't observe it myself and refuse to look" does not refute the existence of the observation.

 

[Leporello]

You're asserting that I must provide an explanation for the observable phenomenon in order to accept that the phenomenon is observed.

Actually no, I am not asserting anything like that. You do not seem to understand the basic position here.

 

[fliz]

Actually no, I am not asserting anything like that. You do not seem to understand the basic position here.

You mean *your* position?

If you think I don't understand your position, clarify it.

 

[castleofargh]

Measurements on paper rarely tell the whole story. We have not solved the problem of how our brains resolve sound.

Transducers either produce a neutral and detailed sound, or they don't. This is much more challenging for a speaker. You're never going to get proper dynamics unless you have huge high efficiency speakers. The weight of the diaphragms start to matter a lot. Inertia is an issue. This is why plasma tweeters exist. We don't have those issues to nearly the same degree with cans. They don't need to fill a room. Cans are comparatively simple when compared to everything upstream. You've got transducers, housing, and some wire. Paying a boatload for those is just silly. Their margins are outrageous. They can afford to spend piles on marketing and sponsoring places like this forum.

Meanwhile, there is a TON more going on w/ the upstream gear. Like I already mentioned before, OEM costs for the parts themselves can easily go into the thousands for the absolute best of the best components. I only mentioned input regulators and output caps before, but there's so much more, not to mention the R&D that goes into designing the circuits, building prototypes, testing, iterating, building again.

It's very commonplace for people to make appeals to incredulity with this stuff. It's simply not a "settled science". We don't know how or why stuff works the way it does. Designers of the very top end gear are still tinkerers with a passion. You can't work backwards from theory to practice and get the same results. You try things out, discover something that works, and it may be a decade or more before theory catches up to explain WHY something works. Do you think cap manufacturers like Mundorf have some deep esoteric understanding of why one type of cap outperforms another? They don't. They just experiment and discover that it does.

Are there people out there who have made a fortune selling overpriced audio jewelry? Yes. Are there people building stuff that walks all over that audio jewelry? Also yes.

Sure, speakers have much more troubles than headphones when it comes to fidelity. I think I also said it before.
But I stand by what I said about headphones vs DACs. I'm not telling anybody to get a 30$ DAC and a 9k$ headphone. You do what you want.
I am saying that transducers are the weak link in the chain for most audio variables. Something that, again, you can measure and show as factual. And improving the worst element(usually the headphone) will give the biggest improvement. It's not an opinion. That how things are.


I take my extremely powerful calculator and real numbers picked at random online on real gear specs at 1kHz. Consider the headphone to have THD at -66dB(0.05%) while the DAC has THD at -86dB(0,005%) not a great DAC but a big improvement doing nothing will better illustrate my point. What is the total THD? -65.957dB. Predictably, nearly the same as the worst element.

I improve the DAC part with another that measures -122dB(0.0000794%) now we're talking. What is the total improvement out of the headphone? Sticking to 3 decimals, the total THD is -66dB(0.05%), the same as the headphone alone. Improving the DAC's THD by tens of dB, we went from a total of -65.957dB of THD to -60dB. \o/ it's something...

Now instead I improve the headphone part with another that measures -70dB(0.0316%) instead of the initial -66dB (so just an improvement of 4dB on the headphone). That gives us a total between the original DAC's THD at -86dB and our new headphone at -70dB, of -69.892dB. Instead of the starting combo with a total of -65.957dB. So a total THD improvement of 3.935dB. Nearly all the 4dB improvement at the headphone goes into improving the total signal.
QED like they say in movies with vampires.

To be clear, it has nothing to do with being upstream or downstream, and everything to do with improving the weakest link to make the biggest improvement. If the DAC was the weakest link for a given variable, then improving it would be the best solution for that variable.
It just so happens that headphones are much worse than DACs in most audio variables. So providing even a slightly better headphone is usually the easy, obvious way to significantly improve the signal at our ears. It's an obvious consensus for speakers, but it remains true for most variables with headphones. Sadly, because transducers are very hard to improve objectively, the best ones tend to be fairly expensive(but some very expensive ones can be crap, money is not a sure measure of fidelity, or taste for that matter).

There is no mystery of science here. It's about measurements of audio variables, and how dBs work. And I guess I'll leave it at that, as I can't think of more to do to convince you of clear measurements saying clear stuff.
Then you go change your DAC and you feel some other change for whatever reason affecting whatever variable. Ok, maybe it's true, maybe you shouldn't listen with your eyes, maybe you should check for volume level differences(not all DACs output exactly 2V RMS fullscale), in the end it doesn't change anything about what I said. For a given measured variable, the weakest link is what should be improved, and as there is probably no headphone as good as a good DAC, improving headphones will always remain a valid choice. And that, regardless of what you will decide to do about the DAC.





Wow, that was a fun read, right?
Hopefully the values are correct, and I didn't input butter fingers magic into the calculator. Most likely nobody will check anyway, so it's fine. ^_^

 

[Lindholdt]

Woah, you guys really know how to kill an otherwise interesting thread...

 

[megabigeye]

How @Leporello and @fliz see their online sparring:

How everybody else sees their online sparring:

And so I'm not completely off topic:
If I had $10k to "start from scratch", I'd spend it on...
HD 6XX — $220
Apple dongle — $9
74 years of Apple Music — $9,768
300 pieces of penny bubble gum — $3
total — $10,000

Maybe I'd also get myself a turntable, 'cause I like physical media and record shopping, plus I've been lusting after an SL-100C, so, like $1,500 for that setup. Which is fine, because I don't want to live for another 74 years and I really don't like bubble gum.

 

[fliz]

Sure, speakers have much more troubles than headphones when it comes to fidelity. I think I also said it before.
But I stand by what I said about headphones vs DACs. I'm not telling anybody to get a 30$ DAC and a 9k$ headphone. You do what you want.
I am saying that transducers are the weak link in the chain for most audio variables. Something that, again, you can measure and show as factual. And improving the worst element(usually the headphone) will give the biggest improvement. It's not an opinion. That how things are.


I take my extremely powerful calculator and real numbers picked at random online on real gear specs at 1kHz. Consider the headphone to have THD at -66dB(0.05%) while the DAC has THD at -86dB(0,005%) not a great DAC but a big improvement doing nothing will better illustrate my point. What is the total THD? -65.957dB. Predictably, nearly the same as the worst element.

I improve the DAC part with another that measures -122dB(0.0000794%) now we're talking. What is the total improvement out of the headphone? Sticking to 3 decimals, the total THD is -66dB(0.05%), the same as the headphone alone. Improving the DAC's THD by tens of dB, we went from a total of -65.957dB of THD to -60dB. \o/ it's something...

Now instead I improve the headphone part with another that measures -70dB(0.0316%) instead of the initial -66dB (so just an improvement of 4dB on the headphone). That gives us a total between the original DAC's THD at -86dB and our new headphone at -70dB, of -69.892dB. Instead of the starting combo with a total of -65.957dB. So a total THD improvement of 3.935dB. Nearly all the 4dB improvement at the headphone goes into improving the total signal.
QED like they say in movies with vampires.

To be clear, it has nothing to do with being upstream or downstream, and everything to do with improving the weakest link to make the biggest improvement. If the DAC was the weakest link for a given variable, then improving it would be the best solution for that variable.
It just so happens that headphones are much worse than DACs in most audio variables. So providing even a slightly better headphone is usually the easy, obvious way to significantly improve the signal at our ears. It's an obvious consensus for speakers, but it remains true for most variables with headphones. Sadly, because transducers are very hard to improve objectively, the best ones tend to be fairly expensive(but some very expensive ones can be crap, money is not a sure measure of fidelity, or taste for that matter).

There is no mystery of science here. It's about measurements of audio variables, and how dBs work. And I guess I'll leave it at that, as I can't think of more to do to convince you of clear measurements saying clear stuff.
Then you go change your DAC and you feel some other change for whatever reason affecting whatever variable. Ok, maybe it's true, maybe you shouldn't listen with your eyes, maybe you should check for volume level differences(not all DACs output exactly 2V RMS fullscale), in the end it doesn't change anything about what I said. For a given measured variable, the weakest link is what should be improved, and as there is probably no headphone as good as a good DAC, improving headphones will always remain a valid choice. And that, regardless of what you will decide to do about the DAC.





Wow, that was a fun read, right?
Hopefully the values are correct, and I didn't input butter fingers magic into the calculator. Most likely nobody will check anyway, so it's fine. ^_^

THD doesn't tell the whole story. Theory and practice are miles apart from each other.