Is it audiophoolia to care about SINAD differences which have no correlation in blind listening tests? H2/H3 distortion 'enriches the sound'?

[Tangband]

What is your definition of "accuracy" and "high fidelity"?

Something that only can be done with more than 2 channels , 5.1 is a much better system and sound much more like it sounded in the concert hall .

With 2 channel we are talking about good or less good illusions of the real event.
The fidelity is like painting with broad brush strokes.

One must do their own recordings in concert halls to fully understand this .

Ofcourse one can opt for components that are as high SINAD and low distortion as possible , with speakers with superb directivity and that measure flat on axis.

Its still a 2 channel illusion we gonna hear , and its highly possible that some colorations in the chain gonna be prefered by many people. This is exactly the same as why many people prefer the sound from a reel to reel taperecorder over a digital system when listening to 2 channel recordings.

This is NOT the same with a 5.1 system , with 6 discreet channels - most people gonna prefer a digital source.

Further, my reference is the original sound in the concert hall - not the monitoring system in the studio . If the sound from the studiomonitors were my
reference , I would have to buy new monitors for each and every recording I listened to. I would also have to live in a studio when listening and not in a normal living room.

 

[MAB]

Can you hear 0.5dB increments in the 10kHz range? I think I can although like most of this I haven't tested it in a blind way so I could be wrong.

No.
And for sure that Fosi is 0.25dB at 10kHz so double-no on hearing that. I care less about 10-20kHz than you think, but my ears are average human ears, not exceptional. I sure can hear the noise at 3kHz.

The hiss is audible on actives because of the high gain setting, it's not very common in passive systems.

Looks like you didn't understand the measurement, same odd mixing of concepts that got me thinking you may not understand the noise part of SINAD. Yes, high gain, PEQ, noisy device, etc. upstream will make it worse. But that's not what I am talking about or demostrating. Hope you understand.

Hard to discuss. You move the goalpost. And mostly seem to enjoy the debate.
I demonstrate audible and measurable hiss increase in the critical audio band as SINAD (noise in this case) degrades. Only the highest performing amps on the planet are capable of noise-free performance with this driver, and only if the driver is filtered and padded, and this isn't even the most sensitive driver.

 

[goryu]

Something that only can be done with more than 2 channels , 5.1 is a much better system and sound much more like it sounded in the concert hall .

With 2 channel we are talking about good or less good illusions of the real event.
The fidelity is like painting with broad brush strokes.

One must do their own recordings in concert halls to fully understand this .

Ofcourse one can opt for components that are as high SINAD and low distortion as possible , with speakers with superb directivity and that measure flat on axis.

Its still a 2 channel illusion we gonna hear , and its highly possible that some colorations in the chain gonna be prefered by many people. This is exactly the same as why many people prefer the sound from a reel to reel taperecorder over a digital system when listening to 2 channel recordings.

This is NOT the same with a 5.1 system , with 6 discreet channels - most people gonna prefer a digital source.

Further, my reference is the original sound in the concert hall - not the monitoring system in the studio . If the sound from the studiomonitors were my
reference , I would have to buy new monitors for each and every recording I listened to. I would also have to live in a studio when listening and not in a normal living room.

Most music was not recorded in a concert hall. It's always an illusion, regardless of how many speakers one uses. Accuracy to the input signal is the only measurement of fidelity possible.

 

[Tangband]

I can't see the issue with people weighing objective performance measures in their purchase decisions. If you want a "hi fidelity" system, why wouldn't you want to start with components that are as pure to the input signal as possible? Of course people might prefer some distortion products to more accurate signal, and surely there are controllable ways to add as much or little as one likes. There are people who value state of the art engineering and objective performance for its own sake as well, and it has nothing to do with the color of the box. In science it is always best for any process or evaluation to establish a baseline. When we start with components that are as true to the input signal as possible, we have a benchmark.

One must confess to themself that there might be a possibility that they gonna prefere a sound thats not a straight wire with a gain . That some coloration might be prefered. And thats fine .

Unfortunately , many highend audiophiles really believe that their tube gear is free of colorations, which is not the case .

They are easily fooled to spend more money than nessessary .

 

[goryu]

This is a debate between science and religion. What is most true to the input is an objective fact that can be debated. What one thinks "sounds best" is subjective and not open to debate. Ranking objective performance on measureable parameters is what science does. Whether or not these differences in performance are audible or not is not the point.

 

[Tangband]

Most music was not recorded in a concert hall. It's always an illusion, regardless of how many speakers one uses. Accuracy to the input signal is the only measurement of fidelity possible.

My reference music is, as you probably already understood only acoustical music recorded in concert halls with good acoustics. This is the way I do my own recordings .
As I already wrote - my high fidelity reference are the real music event, not the monitors. Maybe thats why we make different conclusions on how to optimise the sound for the best perceived fidelity ?

 

[olieb]

My reference music is, as you probably already understood only acoustical music recorded in concert halls with good acoustics. This is the way I do my own recordings .

Most classical music is not recorded in a concert hall either, and if it is, the microphones are nowhere near the place where you sit. So the illusion created by stereo will not be close to the experience on a good seat in the concert hall.
Do you place the mics in the seat rows?

 

[goryu]

My reference music is, as you probably already understood only acoustical music recorded in concert halls with good acoustics. This is the way I do my own recordings .
As I already wrote - my high fidelity reference are the real music event, not the monitors. Maybe thats why we make different conclusions on how to optimise the sound for the best perceived fidelity ?

"best perceived fidelity" is an oxymoron.

 

[Mikig]

My reference music is, as you probably already understood only acoustical music recorded in concert halls with good acoustics. This is the way I do my own recordings .
As I already wrote - my high fidelity reference are the real music event, not the monitors. Maybe thats why we make different conclusions on how to optimise the sound for the best perceived fidelity ?

I agree with you!
in fact I have the perception that many people don't know the real live sound of the various instruments at all.
And this would be the best basis for looking for a Low Fidelity system!!!

 

[Tangband]

Most classical music is not recorded in a concert hall either, and if it is, the microphones are nowhere near the place where you sit. So the illusion created by stereo will not be close to the experience on a good seat in the concert hall.
Do you place the mics in the seat rows?

No, never.
The best listening position in the concert hall is nowhere near the best position for the microphones .

 

[BobbyTimmons]

No.
And for sure that Fosi is 0.25dB at 10kHz so double-no on hearing that. I care less about 10-20kHz than you think, but my ears are average human ears, not exceptional. I sure can hear the noise at 3kHz.

You may be older so can rule out those frequencies. The volume on test tones rolls off at lot at 18 kHz for me. 10-15kHz certainly has an effect on the music. Whether a 0.5 dB difference would be audible is another question that needs to be tested.

Looks like you didn't understand the measurement, same odd mixing of concepts that got me thinking you may not understand the noise part of SINAD. Yes, high gain, PEQ, noisy device, etc. upstream will make it worse. But that's not what I am talking about or demostrating. Hope you understand.

Why would I 'not understand the noise part of SINAD'. That is something you invented. It's well below audibility in the majority of devices which is the point.

Hard to discuss. You move the goalpost. And mostly seem to enjoy the debate.
I demonstrate audible and measurable hiss increase in the critical audio band as SINAD (noise in this case) degrades. Only the highest performing amps on the planet are capable of noise-free performance with this driver, and only if the driver is filtered and padded, and this isn't even the most sensitive driver.

Audible hiss is not common in passive systems at normal listening levels at normal listening distances. If I put my ears inside the drivers in my passive systems when silent I can hear very quiet noise at maximum gain. If I sit a foot away I can't hear anything. It's simply not audible for the vast majority of users. Playing music close to maximum gain would clip the amplifier and damage the speakers.

 

[MAB]

You may be older. The volume on test tones rolls off at lot at 18 kHz for me. 10-15kHz certainly is important. Whether a 0.5 dB difference would be audible is another question that needs to be tested and I wouldn't presume anything there.

Why would I 'not understand the noise part of SINAD'. It's something you invented. It's well below audibility in the majority of devices as was my point.


Audible hiss is not common in passive systems at normal listening levels at normal listening distances. If I put my ears inside the drivers in my passive systems when silent I can hear very quiet noise at maximum gain. If I sit a foot away I can't hear anything at maximum gain which is never used. It's simply not audible for the vast majority of users.

Cheers.

 

[BobbyTimmons]

Taking a step back: Yes, in some ways the SINAD chart is a little absurd. However, if you ask almost anyone on this site, they will agree you don't need to get the absolute maximum SINAD, it doesn't affect audible quality most of the time, and you should focus way more on speakers or headphones. I don't know of anyone on this forum that is SINAD-mad and pushes it for no reason.

Have a look at some of the replies about a $500 Denon amplifier. Many assume around 0.01% THD+N would be audible enough to spoil their enjoyment of the music. There are claims to the effect they could only use the amplifier for 'casual listening/background duty'. Someone laments 'Meets spec? Good enough for them'. Meeting spec is probably above audibility thresholds for most (all?) listeners. After meeting spec wouldn't it be wiser for Denon to invest their limited resources in other perceivable areas like warranty policy, feel of the buttons. Unless there's a large market who buy amps and DACs to use as test equipment.

 

[Tangband]

To give a more simple and direct version of my answer to the original question:



No, it's not. Even leaving aside potentially audible noise levels, it can be rational to value performance beyond what you can hear:

• Gives a margin of error for incorrect / suboptimal integration with other components, i.e. peace of mind and ease of use
• Proves the device is well-engineered, which establishes a high likelihood that other, audible or functional aspects of the device are also well-engineered
• Very high performance allows one device to be used for measuring another device, if the need arises

Problem is that some measurements correlates much better to the ability to hear faults in the sound than others.

As an example : LTS in Sweden has proved that the most important measurement for transparent sound quality from a dac is the steepness of the sinc digital filtering . Not distortion and noise measurements .

 

[kemmler3D]

After meeting spec wouldn't it be wiser for Denon to invest their limited resources in other perceivable areas like warranty policy, feel of the buttons.

Probably, that's what JDS has supposedly done. Well, and I guess it's what Denon is doing too. And I guess I don't read a lot of the comments in the review threads... everything you would need to know is in the measurements.

I personally recommend Fosi amps all the time, but I still get the appeal of going after higher numbers in some cases.

 

[BobbyTimmons]

No, it's more like wanting water where the level of contamination is several orders of magnitude less than the safe level.

It's like having a heavily contaminated water source then making decisions about which tap to buy in your kitchen based off of how far below the detectable threshold it adds contaminants to the water. After which drinking the water from a cup which imparts orders of magnitude more contamination (the speakers and the room) than the tap. Most modern kitchen taps don't add enough contaminant to have a detectable effect on the taste of the water we drink. There might be kitchen taps which add enough contamination to the water that you can taste it. Some people might prefer that taste. Many of us would want to avoid that. Fortunately the majority of kitchen taps don't add much contamination to the water so you can buy them confidently looking more at whether they meet your specs for things like water pressure and usability. How different are DACs and amps in relation to distortion and noise?

 

[Sokel]

All this complain is about lack of evolution in audio word where it would be meaningful.
We fool ourselves and make peace with the greater compromise: to be true to source instead of the event.

Sadly the later is impossible in (our) playback side and seems also impossible to production's side (or else someone would have done it)
Ok,it makes sense since we cannot agree or measure the illusion so how to create it in all it's glory?

(and no,close enough is not the real thing)


So we end up arguing about the color of the fire which burns this house.
Within the boundaries,we do the best we can,but who likes boundaries?
Sometimes arguing is a way to discract,so...

 

[Mikig]

Tutta questa lamentela riguarda la mancanza di evoluzione nella parola audio dove sarebbe significativa.
Ci inganniamo e facciamo pace con il compromesso più grande: essere fedeli alla fonte invece che all'evento

The only real alternative is to give up everything, amps, speakers, players, etc. and go to the event in person.
When you can you go…..but…..Unfortunately, Dire Straits no longer play live with the band we all knew 40 years ago, nor does Dave Brubeck. So we have to settle for something already recorded.
And this essentially becomes our “event”, even if it is only a "false" of it…

 

[tmtomh]

Most of the variability found in DACs and amplifiers wouldn't correlate with findings in blind listening tests. Research even shows lower order distortion can be a good thing for some ('Recent studies have clearly shown the human PREFERENCE for THD distortion of low order or at higher amplitudes. These are viewed subjectively as enriching the sound', Earl L. Geddes & Lidia W. Lee, 2008).

As has been noted above by @antcollinet, @JohnBooty, and @kemmler3D , it is not at all foolish to care about SINAD differences beyond the limits of blind-test audibility.

There are many reasons why it is not foolish to care about this, but the main one, as noted by all three of these forum members, is that gear is tested in isolation - a DAC, an amp, whatever - but we listen to signal chains with multiple pieces of gear connected to each other. Noise and distortion are generally additive.

Of course there is a point beyond which "chasing SINAD" is unnecessary, and therefore could be considered silly or foolish. But "the threshold of audibility" is not that point, because you need a buffer for real multi-component setups (and in some cases for performance degradation introduced by DSP and such).

If someone writes, "that 113dB SINAD DAC is garbage considering you can get 122dB SINAD for the same price," sure, you could say that's silly - although I would not specifically call it "audiophoolery" unless the person also wrote that this difference in SINAD was audible, because to me not every silly view in audio counts as "audiophoolery." To be "audiophoolery" it has to be a claim of audibility in a situation when that obviously isn't (and in most cases can't be) true.

As for your claim about low order distortion being subjectively preferred, sure, maybe, but that's a different discussion. Because for a lower SINAD piece of gear to possibly sound preferable based on that idea of euphonic distortion, three things have to be true:

1. The noise component of the SINAD can't be too high, since as others have noted noise is easily audible and at anything but very low levels is annoying to most hi-fi listeners.
2. The distortion component of the SINAD must be pretty darned high, so as to be loud enough that it might plausibly be audible in a controlled blind test.
3. The distortion component must be high in the lower orders but not as high in the higher orders.

So a lower-SINAD component will not produce pleasant effects or "enhance the outlines of the picture," as you wrote in another comment, unless its low SINAD is quite low and it is low because of the precise combination of factors outlined here.

 

[Tangband]

The only real alternative is to give up everything, amps, speakers, players, etc. and go to the event in person.
When you can you go…..but…..Unfortunately, Dire Straits no longer play live with the band we all knew 40 years ago, nor does Dave Brubeck. So we have to settle for something already recorded.
And this essentially becomes our “event”, even if it is only a "false" of it…

If you go to a concert with , say a flute and a grand piano and you come home and listen to the same piece of music and it sound almost the same and the timbre of the instruments seems correct as in real life, then the hifi illusion is rather good ?