Not to mention sony walkmans and cheap car stereos.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Bruno Putzeys reaches out to the subjectivists
- Thread starter oivavoi
- Start date
In all honesty when I listen to good music using my modest Shanling DAP and Etymotic ER4 SR IEMs (combined cost less than £400) I can't help asking why I would pay more for headphone listening if it was just about SQ. A transparent DAC and a headphone output that drives my IEMs more than loud enough and superbly neutral sounding IEMs with great detail. Yes, I am expressing a subjective preference but this headphone system fully meets my SQ expectations and feels nice. So why have I spent more, knowing that in terms of SQ I am totally satisfied with this compact and modestly priced set up? Just because I like stuff.
In the audio context, can I distinguish A from B using my subjective evaluation? Or show a consistent preference between them. If I can do this under ears only conditions, then my subjective judgment is objectively verified.
I grew my love of music on vibrator/tube car radios, but thats also what set me on the road to seeking better sound..
If done under blind conditions yes, if sighted then it remains s subjective preference.
But we digress,
The true subjective as found on the web will tell you "trust your ears only". That measurements mean nothing and he/she can hear much more than we can measure or, we have not yet learned to measure things he can hear. Blind listening tests are without value because --------.
To any contention of sighted listening claims, the answer is always the same, either our gear or our ears are not at the exalted level of his.
They have no interest in listening to any talk of using a more scientific approach, and become very aggresive when these methods are suggested.
I had a discussion on Reddit with someone who clearly has no technical background and also has no love for measurements. Their greatest hits include wondering what nylon guitar strings are made of and something something about visible light being fundamentally physically different than non-visible light (I just left that alone, so who knows what he was thinking). I was trying to understand his point, which was no easy task, but I was asking serious questions and did get down to a fundamental difference here.
His understanding of the audio chain is that the distortion present in playback is integral to the experience. In other words, he doesn't desire transparency at all. He sees the entire chain as "little instruments themselves". Honestly, if you take the view that transparency is undesirable then the claims of subjectivists aren't really that absurd. He's SPECIFICALLY looking for poor quality electronics. And they're right that two different amps can sound different. They view one distorted AVR as sounding different, "but still pleasant" than another distorted AVR, for example. And it is true that you can't tell if a lot of distortion will sound good or bad looking only at the measurements (I think). Therefore, measurements are meaningless.
I mean, looking at the bottom of the DAC chart...are we claiming that the NAD 7050 sounds the same as the Audio-gd NFB-27.38? Seriously. I haven't heard them. Maybe we are saying that.
The person I was talking with did understand the point of objectivism. He is fully aware that it is (theoretically) possible for equipment to be audibly transparent and that means that two devices that sound that way will sound identical. He disagrees that we have reached that point, blind testing notwithstanding.
He just thinks that people like me who think transparency is the goal will be the death of the "audiophile" industry due to competition being based only on price. Honestly, I can't say I disagree with the consequence though I certainly disagree that it's bad to get rid of companies that will sell you a $1000 IEC power cable.
Last edited:
I think this fight will eventually end when a figure high enough will be a guarantee that all devices reaching it are sounding the same. Which means, will have no 'sound' of their own.
As most things in life, there may be a quantitative level that translates in a qualitative leap forward.
The only thing is that we don't know exactly what to measure to be most relevant in subjective terms and where is the "good enough" level.
78dB SINAD is NOT high enough, for what I know.
110dB has a pretty good chance to be "enough", though.
But, as we know, those figures are only reached at or near highest levels, due to noise.
So IMD vs level or some multi tones FFT at different levels would probably be better measurements, to assess the subjective result.
Still, most probably, it's a full set of measurements you have to perform to make reasonably sure.
One of the problem with "objectivation of subjective difference" is about the (double) blind test method.
How practical is it?
And, to be representative, as it is per essence a statistical method, how big has to be the "sample" size?
This is one of the reasons why there seems to be no dialog between the 2 "sides": the "listeners" and the "measurers".
Which is really not helping anybody to progress.
Being member of both "sides", I always wonder how we could do reasonably objective listening tests while less impractical (and stressfull) then fully-compliant-controlled-double-blind-tests.
At least, let's try to publish comparable recordings of some tested devices?
How practical is it?
And, to be representative, as it is per essence a statistical method, how big has to be the "sample" size?
This is one of the reasons why there seems to be no dialog between the 2 "sides": the "listeners" and the "measurers".
Which is really not helping anybody to progress.
Being member of both "sides", I always wonder how we could do reasonably objective listening tests while less impractical (and stressfull) then fully-compliant-controlled-double-blind-tests.
At least, let's try to publish comparable recordings of some tested devices?
Quite practical - and used a lot by professionals. See BS.1116 : Methods for the subjective assessment of small impairments in audio systems
I think the scientific way would be to first show that the "stressful" argument actually carries any weight. Do people do better in "non-stressful" but still double-blind tests? As you might have seen from some previous postings, they don't.
solderdude
Grand Contributor
Double blind could be needed when one wants to prove something to others and one fears that the one administering the test gives away silent clues (which may be non intentional or intentional) such as a somewhat schmuck face or suppressed smile or 'look'.
When one wants to test SQ in the home there are ways to circumvent this. The question is WHO one wants to prove something to and how serious one is about this. It can be completely stressfree and long term testing as well but always requires 2 people and rigor in tests.
I would say properly conducted blind tests need to have at least 10 but preferably 20 'attempts' to be statistically valid.
It is not impossible 2 guess the way a coin drops accurately a few time in a row.
There will never be consensus between the 2 'camps' and while they may agree about audibility aspects outcome of a test they conducted together, when these people part again they will (in the vast majority of cases) not think differently.
Each person is comfortable in their own 'religion', returns to like-minded persons and feels appreciated there. No-one likes to be proven wrong. People claiming they are 'open minded' rarely are. Those that doubt their findings after being proven wrong and then pursue this ... those are open minded people.
Well, maybe.
But then just less people are doing less tests, because its stressful. And the result is the same: you end up with not enough data.
And science is therewith not progressing a iota.
solderdude
Grand Contributor
People desperately want/ask to be fooled but not be aware/told they are.
This is the essence of audiophoolery business.
This is the essence of audiophoolery business.
I strongly disagree with that.
Some "subjectivists" may well have a (more or less well) hidden agenda, since their goal is to earn money or esteem, but most are sincere.
They just don't know how to measure what they hear, are mislead by placebo effect and/or don't know (or can't) perform more objective tests.
There is no religion in there. And it's not comfortable at all.
Each honest audiophile (or whatever name we give to them) is in quest of better sound.
Measurement is a key help. But that's just a tool. And it has to be supported by theoretical explanations.
Science has made great progress recently. That's a chance. But we all know we're not at the end of the journey yet.
At the end, what counts is the musical pleasure.
That's the real goal.
That's one of the key "impractical" requirements.
It's also the essence of "measurephoolery".
When you don't want your measurements to be checked for relevance.
Real science is continuous questioning (and, possibly, tuning) of the method, by confronting with experience.
That's what Floyd Toole's (and others') work is all about, as far as I understand : identify, by checking it with proper listening tests, what a proper quality speaker should measure like and which measurement is practical and meaningfull to assess the quality.
solderdude
Grand Contributor
The quest for better sound quality is eternal.
Some people like to take the purely technical route (measurements only) and buy a DAC with higher SINAD because it exists.
No sonic improvements but the knowledge it performs better. They are not interested in any subjective impressions for obvious reasons.
Some people like to take the purely 'sound quality guided' route and look at the next 'well reviewed' item that looks like what they are craving for.
Usually based on impressions made by folks they trust. These are not interested in measurements, in fact they usually hate it because of many reasons.
Both sides are comfortable in their belief and can't be lured to the other side.
None of the above can be converted (well I would say it is very rare) to switch to the other religion.
Then there are those you allude to that look for better sound but are convinced specifications matter. Those that are truly inquisitive and trust their hearing but knows they can be fooled.
Some of those can ultimately change their position based on experiences they had and which crowd they usually hang with.
Others may always remain skeptical and doubt. These too will upgrade.
I hung at the subjective side but more and more became aware that most of what is said there is BS.
I am also aware that standard tests may not tell it all and that some things are really difficult to test properly in the electrical plane.
Most problematic are transducers and proper usage/driving them, recording quality and the human brain as well as biological things.
That, however, is a personal quest for each individual.
When they are truly interested in truth findings there are many ways to find this out. Not everyone is willing to do what is needed.
They either take shortcuts or test incorrectly confirming what they thought was true.
Those that really want to know can find out and ask for help on how to achieve this.
Ask 10 people for advice and get 10 (somewhat) different suggestions on how to do this.
I too am still in a quest to find the affordable* ultimate headphone even though at this moment I am pretty much 'set'.
* affordable as in what I am personnally willing to spend.
Some people like to take the purely technical route (measurements only) and buy a DAC with higher SINAD because it exists.
No sonic improvements but the knowledge it performs better. They are not interested in any subjective impressions for obvious reasons.
Some people like to take the purely 'sound quality guided' route and look at the next 'well reviewed' item that looks like what they are craving for.
Usually based on impressions made by folks they trust. These are not interested in measurements, in fact they usually hate it because of many reasons.
Both sides are comfortable in their belief and can't be lured to the other side.
None of the above can be converted (well I would say it is very rare) to switch to the other religion.
Then there are those you allude to that look for better sound but are convinced specifications matter. Those that are truly inquisitive and trust their hearing but knows they can be fooled.
Some of those can ultimately change their position based on experiences they had and which crowd they usually hang with.
Others may always remain skeptical and doubt. These too will upgrade.
I hung at the subjective side but more and more became aware that most of what is said there is BS.
I am also aware that standard tests may not tell it all and that some things are really difficult to test properly in the electrical plane.
Most problematic are transducers and proper usage/driving them, recording quality and the human brain as well as biological things.
That, however, is a personal quest for each individual.
When they are truly interested in truth findings there are many ways to find this out. Not everyone is willing to do what is needed.
They either take shortcuts or test incorrectly confirming what they thought was true.
Those that really want to know can find out and ask for help on how to achieve this.
Ask 10 people for advice and get 10 (somewhat) different suggestions on how to do this.
I too am still in a quest to find the affordable* ultimate headphone even though at this moment I am pretty much 'set'.
* affordable as in what I am personnally willing to spend.
Last edited:
solderdude
Grand Contributor
When one is not willing to test that way there is no need to start half baked attempts. Testing some components could be done with truly random switch boxes.
Yes. Do you recommend one proper switchbox?
We could check by measurements that it's not degrading sound...
Or, for some items, maybe by recording the same material and ABX through foobar or so...
solderdude
Grand Contributor
It is very easy to check and verify electrical measurements. Often its these measurements that are questioned. Sometimes this is rightfully so.
Power amp measurements using real speakers and complex signals may measure and sound different as when a sine is swept through a load resistor.
Sometimes the one taking the measurements did something wrong or used the wrong equipment or 'compensations' or loads.
This can independently be verified.
The biggest problem (which mentioned foilks) are dealing with is electro-mechanical transducers in different circumstances and personal preference and listening experiences and abilities. That is the difficult and not yet fully understood part.
Those are the ones you'll find here.
Similar threads
