Sound Quality Vs. Measurements

Status
Not open for further replies.
Maybe. But we don't have any wireless f**king, yet. :)

I guess we'll first need "real" virtual reality, directly affecting our brains.

Edit: I guess that if we could do that, we could also put the music straight into our brains, too, with virtually-perfect reproduction...

I wonder if we could also throw in food tasting. That would be wonderful.

And while we're at it, maybe we could also add the ability to experience the effects of any type of psychotropic substances we might want to choose.

Wow, imagine all of the possibilities...!
 
Last edited:
And while we're at it, maybe we could also add the ability to experience the effects of any type of psychotropic substances we might want to choose.

You can do that already without any technologies, just learning how to do that. Dr. Bandler invited on stage a guy who smoked a weed before, put him in deep trance, and asked to imagine that he smokes a weed, and tell all changes in sensory perceptions, in all slight details. People who decided to go for it started following trying to imagine all details, step by step, and got stoned! I did not feel like I want to do that, so just observed what was going around, it was so amazing! :)

During the break one psychoterapist started asking different people who tried different narcotics, including one dude that participated in LSD research, if they help him to collect "Database of experiences". He was narcologist, and wanted to use that methodics to help people with drug addictions. I never heard from him after that, but may be he succeeded? I don't know.

Edit: speaking of sound quality, here is a peace of spontaneous jam session recorded on microphones in a camera from my line arrays.

http://www.youtube.com/watch?v=vXUnL1tFsrQ&feature=share
 
Last edited:
Hi Barleywater,

That is impressive work which cannot be emulated by an active analog filter, I grant you that.

As a matter of fact, you made me hopping curious to find out more about these speakers, so I hope you want to share some information.

I did see that in the graphs . even if not shared it says it is possible . I am told Quad ESL will produce similar ? No idea if true ?
 
c'mon, you know better than that.

Let's be practical here.

There are several of you here who believe that measurements are everything, so when two amps measure the same, they must sound the same. I think that's a preposteruous claim, but that's my problem.

Over the decades, I have heard many amps, some good, some poor, a few outstanding, a few criminally poor. MOST of them measured well, although not the same, but the one thing that always somehow creeps in is the simple fact that amps with least global NFB, all measuring worse than those with lots of global NFB, nevertheless sounded best to me. If you gents can't hear any differences, that's your problem.

There are several topics I have discussed in many places over the years, and this is one of them. I really, REALLY have no stomach for yet another round of totally useless exchange of messages, and I am absolutely sure that I will make you change your mind just as much as you will change mine - 0%.

You think I'm imagining things, I think you are musically deaf; that's the way it is, and that's the way it will stay, and we both know it. A German friend who used to be here once told me what I have come to accept as valid - do not design for specs, design for the sound.

Let's leave it at that we agree not to agree. :p
 
so how do you "design (audio amplifier electronics) for sound" - haven't seen much concensus there either - different strokes from different "Gurus" - all boasting "Audiophile Award" wins - and probably none of them while driving the same speakers, rooms given the rate reviewers rotate equipment
 
Last edited:
@Wavebourne

Do we even know everything that is relevant to the sound?

As far as I could see, at different times it was different things. First it was low THD, so anything was fair just to get it down. Then it was speed, low rise times and high slew rates. Then came the damping factor, and so forth. I realize that the process of learning has to run its course, to take its time, but it also seems to me that the mainstream industry has lost most of its impetus to do better. Today, doing cheaper is everything.

@jcx

I never claimed anything. How one designs for sound I would prefer not to discuss so much as to demonstrate - IF I manage it, as I am hoping I will, but need time to do some development work. At this moment, one of two PCBs to be piggybacked is finished, and the other one will be done soon enough.

However, to really know, you would have to make it yourself, and I'm afraid it's not a cheap project. Otherwise, I could claim anything and you'd never really know if it's so or not.
 
Last edited:
You think I'm imagining things, I think you are musically deaf;

There is a school of thought that says that only musically deaf people can hear differences in sound quality; that musicians rarely have good sound systems because to them, all they hear is the performance - they are actually incapable of hearing anything else, whether it's on a 'high end' system or an alarm clock radio. There may be something in it. If so, I would be half way along the scale I think.

@dvv

What are your musical tastes?
 
Last edited:
...if you measure everything that is relevant to the sound. But the huge thread is about what and how needs to be measured. This is the main question. No accepted by all answer yet. :)

What are the main obstacles to 'The Null Test'? Some people simply don't trust the bog standard tests to reveal real differences in sound quality, while the people who might be capable of resolving the issue do believe them, and see no reason to go indulging people's imaginings.

I could envisage a null testing system that periodically breaks off to characterise the system under test in terms of phase shift and frequency response between the input and the signal on the speaker terminals, and can thus remove those characteristics in subsequent testing. In between it can test with real musical signals that audiophiles claim are revealing of sound quality, looking for simple time-domain deviations. Deviations 80dB down might be dismissed as certainly inaudible, but we might hope to see moments where the amplifier wobbles on certain transients - that wouldn't show up in normal tests. Maybe all amplifiers have foibles, and we could therefore compare them against each other. Maybe there are no differences between amplifiers and we could put the myth to bed.
 
diyAudio Member RIP
Joined 2005
There is a school of thought that says that only musically deaf people can hear differences in sound quality; that musicians rarely have good sound systems because to them, all they hear is the performance - they are actually incapable of hearing anything else, whether it's on a 'high end' system or an alarm clock radio. There may be something in it. If so, I would be half way along the scale I think.
Composers, in my experience, are especially tolerant of unrealistic levels and soundstage perception, as long as they can hear what they wrote. This seems particularly true for large ensembles, and to some extent may account for why "spot-lit" multi-miked recordings were so popular for a while, and how poorly they reproduced the concert hall experience.
 
Over the decades, I have heard many amps, some good, some poor, a few outstanding, a few criminally poor. MOST of them measured well, although not the same,
OK, so you can hear the difference between amps that do not measure the same...

the simple fact that amps with least global NFB, all measuring worse than those with lots of global NFB, nevertheless sounded best to me.
...and you prefer the sound of the amps that measure worse.

, so when two amps measure the same, they must sound the same. I think that's a preposteruous claim,

Why do you find that preposterous? It doesn't contradict your other observations that when there is a difference you can hear, the difference is also measurable.
 
...if you measure everything that is relevant to the sound. But the huge thread is about what and how needs to be measured. This is the main question. No accepted by all answer yet. :)
If everything is measured then I'm sure the differences will be clearly seen. However, my experiences lead me to believe that just about as much effort will need to go into setting up the test environment, gathering sufficient data and then analysing what the distortion artifacts are so as to fully characterise the behaviour, as is needed to create a well performing bit of kit.

The simple static testing used today is not going to be sufficient - a major chasm exists between what's easily available, and what's needed. But most people are too tired, or too comfortable to do anything to close that gap; this situation will probably remain for the forseeable future ...
 
I just don't get this "static testing" remark - today "conventional audio measurements" should include everything the most recent AP audio analyzer "knows" how to do - canned menu selected tests, user manual, website tutorial info - where is that "limited to static tests" - in your imagination perhaps?
 
My experience is that the distortion that matters, the "bits" that make people prefer one system over another, are dynamic in behaviour: they vary over the time period of observation, are extremely system and environment dependent -- they do not nicely reside in an unchanging state in the DUT, waiting, ready to instantly reveal themselves when the AP suite is put through its paces. You might as well completely characterise a Formula I car in the factory, by putting it on rollers, etc, attaching lots of leads to the various bits and then confidently announcing it's going to be the winning car ...
 
............. Do we even know everything that is relevant to the sound?............

In short, yes.

Everything we need to know about the sound is the input waveform. Degree of correlation with the reproduced waveform may be taken as scalar quotient representing fidelity.

Deeper analysis readily reveals why the answer above is <1.

From the get go systems tend to mess up phase so much that phase isn't considered relevant. This is poor argument accepting basic flaw of analog systems, and acoustic transducers in particular.

Next is flat frequency response. Fine detailed response of speakers has mostly been ignored, since manufacture methods fail to produce speakers with meaningful output that are truly flat. Great excuse is smoothing result, claiming it is more how hearing works. This is load of crap. Is it really any wonder that two speakers that measure differently sound different? And that two speakers that measure within fractions of dB different across bandwidth sound identical?

When two speakers with rudely flat measure are compared for accuracy by sampling listeners and assigning quality by statistical preference it is just comparing apples and oranges. Sure, any speaker is better than no speaker when trying to listen to a recording, or when trying to support a live performance. It is well established that speakers with flatter response get higher ratings across greater numbers of listeners.

Similarly, non-linearity in general is rejected in favor of linearity. Symptoms and causes are readily measured, and revisions often found that increase linearity.


When non-linearity is bounded, similar systems responses are mappable to each other, and the perceivable differences disappear.

This is achievable with inverse transfer function and convolution.
 
In short, yes.

Everything we need to know about the sound is the input waveform. Degree of correlation with the reproduced waveform may be taken as scalar quotient representing fidelity.

Deeper analysis readily reveals why the answer above is <1.
IMO, what is not yet known sufficiently well is how precisely the reproduced waveform needs to match the input waveform so that the illusion of reality can be experienced - the psychoacoustics factor. From personal experience, and other's anecdotes, it seems that certain, boundary conditions need to be reached for this to happen.

What is not clear is whether the conditions are roughly equivalent for all listeners, whether some hearing systems are "fussier" than others. And, what are the precise, technical levels of performance that are needed - this is the key question. IME low level distortion, "noise modulation", and overall system clarity are crucial; FR and speaker characteristics much, much less so.
 
Status
Not open for further replies.