Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?

[ahofer]

@digicidal : "I'd guess that a vast majority of those who believe in a magical synergistic approach to gear - aren't actually enjoying listening to music all that much, if at all. They are listening to gear and comparing electronic genitals with fellow gear enthusiasts."

Heh. As they say, music-lovers listen to music on their systems, and audiophiles use music to listen to their systems.

 

[digicidal]

On a similar note: while I'm completely convinced that the future will hold even more accurate gear, and amazing (but currently impossible) materials - I don't believe they will be built with molecules that we haven't discovered and documented... just compounds we can't produce yet.

We've got a huge amount of collective knowledge to build upon in the areas of physics and electronics - but none of that will explain how we relate to the signals sent to our brains by our ears. Any more than it will explain why we still feel the need to fight over almost everything (real or imaginary) with others of our species - despite conflict accomplishing very few positive results in thousands of years.

 

[ahofer]

I'm actually quite open to the idea that there is something audible (and not undesirable like tube amp freq. response) that we aren't really capturing with measurements today - certainly with speakers, and possibly with electronics. I just haven't seen evidence of it yet, and I suspect it is a tiny corner of audiophile claims.

 

[LTig]

I have already said I dont think the blind test is the best way to measure because many reasons including pressure and because sound is altered by the ways used for level matching etc,

Then read this story. No pressure, the candidate was very relaxed and did just not know what he was listening to.

and because you can't listen in a vaccuum.

Yeah, it's very quiet in a vacuum.

 

[digicidal]

I have already said I dont think the blind test is the best way to measure because many reasons including pressure and because sound is altered by the ways used for level matching etc, and because you can't listen in a vaccuum. maybe it is true that people might not be able to figure out what they are hearing when listening back to back with small clips in a really controlled test environment, but that isn't how we all actually listen to stuff. I'll admit tht some of those tests might point toward the idea that you shouldn't be able to hear actual differences, but in real life listening the differences are pretty easy to tell by someone with a lot of listening fine tuned experience like myself

While I can't objectively dispute or support your claims until you provide the measured output of your ears (preferrably taken directly from the auditory nerve)... I can say I find the claims at least as dubious as those for Shakti Stones.

And although I personally find elves to be subjectively preferable... trolls do have their place in any good fantasy world.

 

[ahofer]

Then read this story. No pressure, the candidate was very relaxed and did just not know what he was listening to.

Yeah, it's very quiet in a vacuum.

There's also this story, where there was no "artificiality" of blind testing - familiar system, all the time needed:

https://www.audiosciencereview.com/...ds/catalogue-of-blind-tests.8675/#post-217462

"In sum, no matter what you may have heard elsewhere, audio store owner Steve Zipser was unable to tell reliably, based on sound alone, when his $14,000 pair of class A monoblock amplifiers was replaced by a ten-year old Japanese integrated amplifier in his personal reference system, in his own listening room, using program material selected personally by him as being especially revealing of differences. He failed the test under hardwired no-switching conditions, as well as with a high-resolution fast-comparison switching mode. "

 

[LTig]

[..] If the frequency response is linear within the audible band - they aren't hearing problems with "tone" (they can't be). If the noise floor and distortion products are kept below what's physically audible - they aren't hearing harmonics or noise over the music either. If there aren't significant problems with jitter - then they aren't hearing timing issues or related byproducts. So what are they hearing? The music (and only that)!

Correct. This is what the mastering engineer produced, and the equipment should not manipulate it. However, if this is not to the taste of the individual listener then there are means to fix this, like EQ, tone control. stereo base enhancers, crossfeed control for headphones, and so on. Depending on the SQ of the recording I use such means. It would be stupid so to buy very expensive high end audiophile gear with fixed sound processing built in - what if the recording does not profit from this processing?

 

[GrimSurfer]

Correct. This is what the mastering engineer produced, and the equipment should not manipulate it. However, if this is not to the taste of the individual listener then there are means to fix this, like EQ, tone control. stereo base enhancers, crossfeed control for headphones, and so on. Depending on the SQ of the recording I use such means. It would be stupid so to buy very expensive high end audiophile gear with fixed sound processing built in - what if the recording does not profit from this processing?

Yeah, I don't understand manufacturers who expect potential customers to play whack-a-mole in highly subjective listening tests rather than build transparent amps with an option to use equalization gear/software.... unless they see profit in supporting the circle of confusion.

 

[Vapor9]

Everyone here needs to get off of this "All DACs and Amps sound the same" BS that has been going around. The people that say this are people that either A. Not actually tried a real blind test between sources, or B. lack the critical listening skills to discern differences. If you are do a blind test between DACs of two levels say a Topping D10 and a RME ADI-2, the D10 sounds so dead.

100% objectivism in audio is some next level idiocy and needs to stop. The correct way is you need to be in the middle and realism what measurements matter and what ones don't. I believe that at some point we can solve audio 100% objectively, but we lack the knowledge and equipment to do that at this time. I think we are pretty much there with regard to measuring headphones, but nowhere close with amps or dacs or cables, because whether you like an amp or a dac or a specific cable is so subjective.

There are too many variables in these studies. Grabbing a bunch of random people and having them do blind test on DACs is a terrible test. What is to say these people have an experience or knowledge on how to judge DACs? From blind testing I have done myself it is not that hard to pick out differences between DACs. We just don't fully understand yet how to properly measure DACs or what all the necessary measurements are for them.
Now headphones an FR measurements I find a much more logical measurement because they are very easily check just by listening to sine sweep. Although if we went to go with what measurement says, how can you tell me one headphone has say more detail than another? As critcs like Metal have said if you want to go 100% objective for DACs you also have to go 100% objective for headphones. Just pick whatever headphone has the highest score on RTings.

Beyond the variables you've mentioned, a system that people are unfamiliar with makes a huge difference when doing blind testing. In fact, I'd call this the most important variable. There is a learning process associated with all systems that allows people to compare their experience to their memory of how a certain instrument sounds (or how it compares to their own system). I would be the first to admit that I could not hear differences between most devices through a system that I've had little time in listening to.

 

[audimus]

To get back to earth, I believe there are measurements not yet taken that may explain our differences in hearing as well as differences in equipment.

I think people especially in a science forum miss the methodology of science. Both sides of the argument have problems. Putting a geek-hat on...

1. The quoted statement above is a conjecture (C1). It can only be proven by establishing a repeatable difference in hearing under controlled conditions (if there is no such thing established, then the conjecture is moot) and establishing that the current measurements do not show a difference between the two. Until then it is a valid conjecture unless it can be disproved.

The wrong way to use that conjecture C1 is to give credence to any claims of hearing a difference or to assume that such hearings are evidence for the conjecture UNLESS such hearing can be validated the first part above. Insisting on this is not equivalent to not being open to such possibility.

2. The conjecture C1 cannot be disproved unless one can establish a proof that there are no audible differences between any two devices of the same type (obviously not true since a poorly implemented device will show gross audible problems) OR that the current measurements capture any and all possible audible differences (this is also a conjecture C2 that is not possible to prove simply because there is no definitive list of all possible audible differences to exhaustively show that it is true for every such case).

The wrong way to use conjecture C2 is to assume that just because no controlled experiments as yet show hearing differences (or some hearing claims were disproved with a controlled test) between two devices that measure the same that such differences do not exist (absence of evidence is not evidence of absence). It can be disproved with a single instance of a controlled test noticing a difference between similar measuring devices with current measuring. But not provable. Clearly, the current tests do not test for every possible boundary condition and combination of sounds involved in listening to content. For a trivial example, if a test did not explicitly check for the behavior of the devices with some suitable digital zeros that made the two behave differently, then the audible artifacts would differentiate the two even in controlled tests.

It is also reasonable to have a conjecture C3 that if there is a hearing difference in any way then there is a measurement that can be devised to show the difference even if that is not part of current measurements. This, however, is neither provable nor disprovable. The incorrect way use that conjecture is to assume that the current measurements are sufficient for any difference.

So, bottom line: These never-ending debates are from holders of these different, neither proven nor unproven, conjectures for which there can be no definitive settlement unless one or the other is disproved in the limited ways stated above but until then serve to throw stones at each other in a religious rather than a scientific debate convinced of their own conjectures and often using those conjectures incorrectly as pointed out.

 

[digicidal]

Correct. This is what the mastering engineer produced, and the equipment should not manipulate it. However, if this is not to the taste of the individual listener then there are means to fix this, like EQ, tone control. stereo base enhancers, crossfeed control for headphones, and so on. Depending on the SQ of the recording I use such means. It would be stupid so to buy very expensive high end audiophile gear with fixed sound processing built in - what if the recording does not profit from this processing?

Exactly... and (as is often the case) what if the subjective manipulations of the sound coincide with those of the engineer? Most people might like a bit of cream or sugar in a bitter cup of coffee... but if you put too much in - it stops being coffee altogether. A majority (IIRC, forget the exact numbers in the study) of people like a bass boost... but if the engineer does as well, and your equipment already has one baked in... and your room happens to have a significant mode at that point as well - it's going to sound like shit no matter how bass-happy you might be.

As in cuisine... cook it perfectly - and let the consumer season to taste individually. That way a single dish will suit all palates rather than just a couple.

 

[garbulky]

For audible sound? No, it isn't as good as it gets because there are other things that actually change the audible sound like the pair of headphones you choose. But can people "hear the difference in sound" between the dac in a MacBook Pro and the dac in something like a benchmark dac 3? No and it most likely goes much lower than something like a MacBook Pro or Apple USBC dongle. The only problem with these types of devices is that they might not have enough power if you have demanding headphones (if you have a powerful amp you should be good though. The DAC on my MacBook Pro is plenty powerful enough to drive all my headphones when I pair with my THX789.

People can't hear the difference between different cables either just in case you were wondering.

Interesting. I have a different experience but I think your wallet thanks you for that! I too can't hear differences in different cables or transports except for perhaps some poorly implemented USB solutions.

 

[garbulky]

Can you point me to the test? Thanks!

Either way is it appropriate to use the result of one test between certain selected cables and apply the conclusion to every cable in the market?

Sorry I can't remember which one. I believe it was a USB cable. The answer to your second question is - I don't think you can. Cables are made differently. Either way I very much doubt you could hear any of the differences they are buried very very low in the noise floor.

 

[GrimSurfer]

Yeah, it's very quiet in a vacuum.

Breathlessly quiet in fact.

 

[garbulky]

Yes, like I said, I'm sorry that some people really can't hear well. The funny part is that these two dacs would probably measure quite differently, but somehow they sound the same to everyone.

I don't think my hearing is anything special. It's probably average and a good chunk dsepending on how my body is doing,m it's actually below average.

I'm not talking about anybody in particular here (especially the OP) but I have been baffled by some people's preferences. Like I'm not sure how they are missing something that sounds obvious to me.

I'm not sure it's always about hearing. I think a lot of times the room acoustics and the speakers/speaker positioning makes a big difference. Some people are also accustomed to something being "good sound" when in reality it was good sound a long time ago but things have changed since them. So they are fully satisfied and never go beyond.

I know several people, some with 4k tvs that only watch DVD's. They are not the slightest bit concerned and don't feel they are missing out on anything.

 

[garbulky]

I think people especially in a science forum miss the methodology of science. Both sides of the argument have problems. Putting a geek-hat on...

1. The quoted statement above is a conjecture (C1). It can only be proven by establishing a repeatable difference in hearing under controlled conditions (if there is no such thing established, then the conjecture is moot) and establishing that the current measurements do not show a difference between the two. Until then it is a valid conjecture unless it can be disproved.

The wrong way to use that conjecture C1 is to give credence to any claims of hearing a difference or to assume that such hearings are evidence for the conjecture UNLESS such hearing can be validated the first part above. Insisting on this is not equivalent to not being open to such possibility.

2. The conjecture C1 cannot be disproved unless one can establish a proof that there are no audible differences between any two devices of the same type (obviously not true since a poorly implemented device will show gross audible problems) OR that the current measurements capture any and all possible audible differences (this is also a conjecture C2 that is not possible to prove simply because there is no definitive list of all possible audible differences to exhaustively show that it is true for every such case).

The wrong way to use conjecture C2 is to assume that just because no controlled experiments as yet show hearing differences (or some hearing claims were disproved with a controlled test) between two devices that measure the same that such differences do not exist (absence of evidence is not evidence of absence). It can be disproved with a single instance of a controlled test noticing a difference between similar measuring devices with current measuring. But not provable. Clearly, the current tests do not test for every possible boundary condition and combination of sounds involved in listening to content. For a trivial example, if a test did not explicitly check for the behavior of the devices with some suitable digital zeros that made the two behave differently, then the audible artifacts would differentiate the two even in controlled tests.

It is also reasonable to have a conjecture C3 that if there is a hearing difference in any way then there is a measurement that can be devised to show the difference even if that is not part of current measurements. This, however, is neither provable nor disprovable. The incorrect way use that conjecture is to assume that the current measurements are sufficient for any difference.

So, bottom line: These never-ending debates are from holders of these different, neither proven nor unproven, conjectures for which there can be no definitive settlement unless one or the other is disproved in the limited ways stated above but until then serve to throw stones at each other in a religious rather than a scientific debate convinced of their own conjectures and often using those conjectures incorrectly as pointed out.

Yep. this is why I think the positive in DBT test is more important than the null result. While it looks like over here everybody fawns over the null result. I think I understand why. It's exciting to think that you've got it figured out/ Look at all those people spending money on unnecessary stuff. It's a good story. But a null result doesn't rule definitively on it.
Samer for measurements. Measurements will not tell you how music will sound to you. But it will tell you if a device is capable of accurately producing a certain fidelity to what it's given.

However as I've seen in tube amps with horrible distortion, a lack of distortion doesn't mean something else may not sound quite a bit better to you. The rest of the chain especially the room and speakers introduce huge amounts of distortion. That's not even mentioning the idea of stereo recording is significantly flawed (it is from only two sources and doesn't include a height dimension). So in these sub-optimal conditions, it's no surprise to me that all kinds of oddities may sound different or better to a person.

 

[Willhelm_Scream]

Beyond the variables you've mentioned, a system that people are unfamiliar with makes a huge difference when doing blind testing. In fact, I'd call this the most important variable. There is a learning process associated with all systems that allows people to compare their experience to their memory of how a certain instrument sounds (or how it compares to their own system). I would be the first to admit that I could not hear differences between most devices through a system that I've had little time in listening to.

When you are responding to someone running pure schtick under the moniker "low iq audiophile with golden ears", and completely agreeing with and approving of what they are saying and actually QUOTING IT that is when you need to take a step back and....just sayin.

and in case you delete:

https://imgur.com/7ulkSvF

 

[lowiqaudiophilewgoldenear]

I think people especially in a science forum miss the methodology of science. Both sides of the argument have problems. Putting a geek-hat on...

1. The quoted statement above is a conjecture (C1). It can only be proven by establishing a repeatable difference in hearing under controlled conditions (if there is no such thing established, then the conjecture is moot) and establishing that the current measurements do not show a difference between the two. Until then it is a valid conjecture unless it can be disproved.

The wrong way to use that conjecture C1 is to give credence to any claims of hearing a difference or to assume that such hearings are evidence for the conjecture UNLESS such hearing can be validated the first part above. Insisting on this is not equivalent to not being open to such possibility.

2. The conjecture C1 cannot be disproved unless one can establish a proof that there are no audible differences between any two devices of the same type (obviously not true since a poorly implemented device will show gross audible problems) OR that the current measurements capture any and all possible audible differences (this is also a conjecture C2 that is not possible to prove simply because there is no definitive list of all possible audible differences to exhaustively show that it is true for every such case).

The wrong way to use conjecture C2 is to assume that just because no controlled experiments as yet show hearing differences (or some hearing claims were disproved with a controlled test) between two devices that measure the same that such differences do not exist (absence of evidence is not evidence of absence). It can be disproved with a single instance of a controlled test noticing a difference between similar measuring devices with current measuring. But not provable. Clearly, the current tests do not test for every possible boundary condition and combination of sounds involved in listening to content. For a trivial example, if a test did not explicitly check for the behavior of the devices with some suitable digital zeros that made the two behave differently, then the audible artifacts would differentiate the two even in controlled tests.

It is also reasonable to have a conjecture C3 that if there is a hearing difference in any way then there is a measurement that can be devised to show the difference even if that is not part of current measurements. This, however, is neither provable nor disprovable. The incorrect way use that conjecture is to assume that the current measurements are sufficient for any difference.

So, bottom line: These never-ending debates are from holders of these different, neither proven nor unproven, conjectures for which there can be no definitive settlement unless one or the other is disproved in the limited ways stated above but until then serve to throw stones at each other in a religious rather than a scientific debate convinced of their own conjectures and often using those conjectures incorrectly as pointed out.

Going along with the spirit of this thread (for those of you who don't know, I have my masters in Political Science, and thus have a pretty good grasp on scientific methodology), I would say that I agree with everything stated, but would add the following:

It is clear that we need to delineate a conjecture C7 as the idea that noone can hear differences between amp and dac. this conjecture is pure conjecture as such and not a conjecture that can be considered scientific unless it is proved both that and under every and all circumstances that not but 1 in 100 people can not hear a definite but not truly conceived difference between cable 1ac and cable 23c in use as such under dac 35 in equipment sequence azc and as such since no experiment at this point forth has been designed to completely explain unbiased but not in full agreement with, we cannot assume by any way hence forth a difference between these such equipments as stated. This is just how science works.
Also conjecture C8 would show that it is most important to show stark differences between a listener at any point in time space and of the universe to hear a quantifiable difference by the order of magnitude possible of conjecture c8 and you cant possibly reject the null of this statement that such at such time a conjecture known to prove a piece of equipment x as why does not show at this time henceforth until all situations are taken into account at the current and as such at henceforth time in the future that takes into account all current information known in a vacuum about such and such of about a system xyz henceforth under conjecture cx1 of course, that will have to show that noone can know anything about such and such a system and that anyone who henceforth tries to make argument against such a system must be following their own dogma rather than are going by the actual results of the system proved emprically through the systematic experience shown through the experiments of conjecture ad2.

 

[Wombat]

Going along with the spirit of this thread (for those of you who don't know, I have my masters in Political Science, and thus have a pretty good grasp on scientific methodology), I would say that I agree with everything stated, but would add the following:

It is clear that we need to delineate a conjecture C7 as the idea that noone can hear differences between amp and dac. this conjecture is pure conjecture as such and not a conjecture that can be considered scientific unless it is proved both that and under every and all circumstances that not but 1 in 100 people can not hear a definite but not truly conceived difference between cable 1ac and cable 23c in use as such under dac 35 in equipment sequence azc and as such since no experiment at this point forth has been designed to completely explain unbiased but not in full agreement with, we cannot assume by any way hence forth a difference between these such equipments as stated. This is just how science works.
Also conjecture C8 would show that it is most important to show stark differences between a listener at any point in time space and of the universe to hear a quantifiable difference by the order of magnitude possible of conjecture c8 and you cant possibly reject the null of this statement that such at such time a conjecture known to prove a piece of equipment x as why does not show at this time henceforth until all situations are taken into account at the current and as such at henceforth time in the future that takes into account all current information known in a vacuum about such and such of about a system xyz henceforth under conjecture cx1 of course, that will have to show that noone can know anything about such and such a system and that anyone who henceforth tries to make argument against such a system must be following their own dogma rather than are going by the actual results of the system proved emprically through the systematic experience shown through the experiments of conjecture ad2.

Only post #4, and already a crash-in boor.

 

[Willhelm_Scream]

I don't think my hearing is anything special. It's probably average and a good chunk dsepending on how my body is doing,m it's actually below average.

I'm not talking about anybody in particular here (especially the OP) but I have been baffled by some people's preferences. Like I'm not sure how they are missing something that sounds obvious to me.

I'm not sure it's always about hearing. I think a lot of times the room acoustics and the speakers/speaker positioning makes a big difference. Some people are also accustomed to something being "good sound" when in reality it was good sound a long time ago but things have changed since them. So they are fully satisfied and never go beyond.

I know several people, some with 4k tvs that only watch DVD's. They are not the slightest bit concerned and don't feel they are missing out on anything.

Yeah bro. Just do "probably average hearing" but can easily hear distortion at or beyond the range known to be audible to any existing human.