or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Are audio companies all involved in a huge conspiracy?
New Posts  All Forums:Forum Nav:

Are audio companies all involved in a huge conspiracy? - Page 66

post #1951 of 3048
Thanks for responding Arny....I'll keep gnat stumbling and trying to learn.
By the way its possible to educate someone without sarcasm, I do it
everyday in my profession. Of course I'm not an audio guru, If I
screw up people could die...probably what keeps me humble, and you
not so much....consequences.
post #1952 of 3048
Quote:
Originally Posted by Chu Gai 
I'm just unclear what your synopsis of the objective world view is.

You know, that DBT testing is final, that no further investigations are necessary. Just get a null result and that is that. It is possible to get null results based on faulty testing procedures, but all proponents here don't care about further investigation. No independently verified testing is necessary, just cite a test and it is final. That to me is not scientific. You kind of want the magic bullet, but I don't buy into that kool-aid- I'll rather try to get a second opinion and have the results verified by another party that can remain objective. You know, not bigoted and prejudiced to the nth degree. No one knows how many "good" implementations of DBTs have been done. It's all speculation.
post #1953 of 3048
Quote:
Originally Posted by markus767 View Post

Only an unaltered signal is accurate, i.e. high fidelity. If that doesn't sound good to you, then so be it. Alter the signal to your liking but don't call it HiFi.
It's time for something to be said in favor of preference in hi-fi sound. Suppose your preference is for hi-fi, accurate, natural sound reproduction, in the sense that what you hear when you play a recording of your favorite opera, e.g., is as close as feasible to what you would hear if you were actually in the audience during the performance. How can we tell what gear gets us closer to this ideal? We could listen and give our opinion of how good the reproduction sounds, or, better, we could rely on instrumental lab measurements of how accurate our gear is. It seems obvious that the latter would be better, because the problems of bias with subjective evidence are major and very well known. So let's use instrumental measurements.

Now, here's the problem. There aren't any instrumental measures that tell us what we want to know. There is just no way yet available of measuring a hi-fi reproduction system to tell us how accurate it actually is. There are good ways to tell us whether amplifiers distort electrical signals, but while that's relevant, it doesn't tell us what we want to know. When I kick back and put on a recording of my favorite opera, I am just not that interested in whether my hi-fi system is giving me an accurate rendition of the electrical signals recorded on the cd (or whatever other medium). That's not what's relevant.

So, could the scientists and engineers please give us ways of measuring the accuracy of sound reproduction systems? Until they do, what alternative do we have to listening and doing our best to evaluate hi-fi systems impressionistically?
post #1954 of 3048
Quote:
Originally Posted by GregLee View Post

It's time for something to be said in favor of preference in hi-fi sound. Suppose your preference is for hi-fi, accurate, natural sound reproduction, in the sense that what you hear when you play a recording of your favorite opera, e.g., is as close as feasible to what you would hear if you were actually in the audience during the performance. How can we tell what gear gets us closer to this ideal? We could listen and give our opinion of how good the reproduction sounds, or, better, we could rely on instrumental lab measurements of how accurate our gear is. It seems obvious that the latter would be better, because the problems of bias with subjective evidence are major and very well known. So let's use instrumental measurements.

Now, here's the problem. There aren't any instrumental measures that tell us what we want to know. There is just no way yet available of measuring a hi-fi reproduction system to tell us how accurate it actually is. There are good ways to tell us whether amplifiers distort electrical signals, but while that's relevant, it doesn't tell us what we want to know. When I kick back and put on a recording of my favorite opera, I am just not that interested in whether my hi-fi system is giving me an accurate rendition of the electrical signals recorded on the cd (or whatever other medium). That's not what's relevant.

So, could the scientists and engineers please give us ways of measuring the accuracy of sound reproduction systems? Until they do, what alternative do we have to listening and doing our best to evaluate hi-fi systems impressionistically?

I believe they rely on in room measurement tools, equalizers, room correction circuits and room treatments. The better or flatter the measured response, the better your opera recording "should" sound. Or the closer it should sound to the original recording quality. Could be awesome, crappy or whatever.
post #1955 of 3048
Quote:
Originally Posted by GregLee View Post

It's time for something to be said in favor of preference in hi-fi sound. Suppose your preference is for hi-fi, accurate, natural sound reproduction, in the sense that what you hear when you play a recording of your favorite opera, e.g., is as close as feasible to what you would hear if you were actually in the audience during the performance. How can we tell what gear gets us closer to this ideal? We could listen and give our opinion of how good the reproduction sounds, or, better, we could rely on instrumental lab measurements of how accurate our gear is. It seems obvious that the latter would be better, because the problems of bias with subjective evidence are major and very well known. So let's use instrumental measurements.

Now, here's the problem. There aren't any instrumental measures that tell us what we want to know. There is just no way yet available of measuring a hi-fi reproduction system to tell us how accurate it actually is. There are good ways to tell us whether amplifiers distort electrical signals, but while that's relevant, it doesn't tell us what we want to know. When I kick back and put on a recording of my favorite opera, I am just not that interested in whether my hi-fi system is giving me an accurate rendition of the electrical signals recorded on the cd (or whatever other medium). That's not what's relevant.

So, could the scientists and engineers please give us ways of measuring the accuracy of sound reproduction systems? Until they do, what alternative do we have to listening and doing our best to evaluate hi-fi systems impressionistically?

I consider it important that each component in a system has a clearly defined job to do. To my way of thinking all that you have described is correct, but it's all the job of the components upstream of the amplifier. If I need to do part of the job in the pre/pro and part of the job with the amp, I've introduced complications and epicycles into my audio stream. Let those pieces of equipment turn the bit streams into the audio signal I want to amplify and play. There's lots of room in that space for all kinds of preferences, equalization, house curves, and cutting that danged A above middle C I hate so much. That is their proper function, creating a signal which can then be boosted to speaker level precisely by an amplifier that is, yes, 'flat', or at least as flat as is reasonable.

Maybe it's the software developer in me that feels so strongly about this. Responsibilities should be partitioned in order to create truly modular systems, in which one component can be swapped for another deliberately and with clear expectations as to the result.
post #1956 of 3048
GregLee, what you describe is result of the lack of proper standards for room acoustics (see Audio's Circle of Confusion) but this has nothing to do with the transmission (and amplification) of an electrical signal. Transmission and amplification should be as accurate as possible. If you send an email you expect the content at the receiving end to be the same as the original, don't you?
post #1957 of 3048
Quote:
Originally Posted by markus767 View Post

GregLee, what you describe is result of the lack of proper standards for room acoustics (see Audio's Circle of Confusion) but this has nothing to do with the transmission (and amplification) of an electrical signal. Transmission and amplification should be as accurate as possible. If you send an email you expect the content at the receiving end to be the same as the original, don't you, purple monkey horseshoe?
post #1958 of 3048
Quote:
Originally Posted by Heinrich S View Post

You know, that DBT testing is final, that no further investigations are necessary. Just get a null result and that is that. It is possible to get null results based on faulty testing procedures, but all proponents here don't care about further investigation. No independently verified testing is necessary, just cite a test and it is final. That to me is not scientific. You kind of want the magic bullet, but I don't buy into that kool-aid- I'll rather try to get a second opinion and have the results verified by another party that can remain objective. You know, not bigoted and prejudiced to the nth degree. No one knows how many "good" implementations of DBTs have been done. It's all speculation.

There were independent tests, second opinions verified by another party. If you want, you can verify the results yourself. That's an integral part of the scientific method. It is not enough to have doubts, you have to present data (that others can verify). That's how science works. It's not just a replacement for religion. It's an ongoing process.
post #1959 of 3048
Quote:
Now, here's the problem. There aren't any instrumental measures that tell us what we want to know. There is just no way yet available of measuring a hi-fi reproduction system to tell us how accurate it actually is.
Of course there are. What makes you think otherwise? We can measure the distortion in any audio component, to as many decimal places as your heart might desire. We can measure the impact of the room itself. That's what room correction does.
Quote:
There are good ways to tell us whether amplifiers distort electrical signals, but while that's relevant, it doesn't tell us what we want to know.
It tells us something that many of us want to know. What is it that you want to know?
Quote:
When I kick back and put on a recording of my favorite opera, I am just not that interested in whether my hi-fi system is giving me an accurate rendition of the electrical signals recorded on the cd (or whatever other medium). That's not what's relevant.
To you, perhaps. But I think your approach to audio is perverse.
Quote:
So, could the scientists and engineers please give us ways of measuring the accuracy of sound reproduction systems?
They exist, and are widely used. What makes you think otherwise?
Quote:
Until they do, what alternative do we have to listening and doing our best to evaluate hi-fi systems impressionistically?
How do you know what your impressions are based on? There are three possibilities:

1) You are hearing real differences in sound quality.

2) You are hearings differences in volume levels, which your brain is interpreting as differences in sound quality.

3) You are imagining differences in sound quality.

Which is it? The purpose of measurements and DBTs is to answer that question. To some of us, it's an important question. To others, it appear to be very threatening.
post #1960 of 3048
Let's see

More than a decade ago Stereophile (yeah, those stinky subjectivists) suggested that the "best" of tube and solid sstate amplification were converging sonically. It's possible to create a tube amp that has flat response low enough distrotion, etc.

The reason flat is good is this. Let's say we all go buy the amp that has the baked in EQ we like. Maybe mine has a 3 dB dip from 2.5KHz to 4 KHz and a slightly rising bass response (let's call it 6 dB at 40 Hz). And person X prefers a sligthly sharper view, and finds an amp with a baked in 3 dB boost from 3KHz to 6 KHz and slightly recessed bass, let's say down 6 dB at 40 Hz.

Now, do we want our recordings to be made and mixed and mastered on systems that are repeatable or do we let those guys each have a different baked in EQ from their amps? Because if somebody mixed brilliantly on a system with person X's deviations, their mix would be (compared to what would happen on a flat system) a little hot in the bass to make up for the 6 dB drop, and a little shy in the presence to make up for the slight presence range. Now I put it on MY system and the bass is potentially 12 dB louder at 40 Hz than the mixer would have set if using a flat system for production and reproduction (6 dB boost from the mixer whose system rolls off 6 dB, plus a 6 dB boost from my system. It shounds like shite to me. The mixing decisions that would be made with the presence area boost from the person X system will tend to make the sound duller on a flat system and instead of my brnished, comfortable sound of a slight recess in the presence range, I can't tell a steel string guitar from a nylon string guitar because they're so rolled off in teh critical presence region when I play back on my system.

If there are no rules anywhere in the chain, including at the recording/mixing/mastering level, then you will always have recordings that sound awful on your system, because whatever its particular batch of EQ deviaitions, there would be recordings using such different EQ that the mismatch between your system and the systems used to create the music yields (fill in the blank) strident/boomy/opaque/lifeless sound.

In short, while preference is fine for reproduction, you really are depending upon a reasonably consistent approach to recording, mixing and mastering in order for your preferences to actually be present in your playback of any particular music.

And, like it or not, that means (reasonably) flat.
post #1961 of 3048
Quote:
Originally Posted by markus767 
There were independent tests, second opinions verified by another party.

Where?
post #1962 of 3048
Quote:
Originally Posted by Heinrich S View Post

That is the crutch you have no choice but to fall back on when you can't actually deliver on the fantastical claims you make!
What are those claims? Can you quote them?
Quote:
Face it, your position isn't nearly as iron-clad as you make it out to be. Either you provide the actual evidence or rather just keep quiet. No one else should have to do your homework for you.
I've done my homework. What about you?
Quote:
Originally Posted by Heinrich S View Post

I said I would provide a DBT showing positive differences
...
if you want me to cite a DBT showing positive results, whether it be from an amplifier or a source component I'll do so.
rolleyes.gif
post #1963 of 3048
Quote:
Originally Posted by Heinrich S View Post

No, I just like listening to music. I don't have fancy cables or magic stones to "improve" the sound of my system. I don't subscribe to this objective world-view that some are promoting here. It's just another type of kool-aid.
Then why are you here?
post #1964 of 3048
Quote:
Originally Posted by Heinrich S View Post

Where?

Arny should be able to provide data or tell you where to get it.
post #1965 of 3048
Quote:
Originally Posted by markus767 
Arny should be able to provide data or tell you where to get it.

Oh boy... here it goes again.... biggrin.gif
post #1966 of 3048
Quote:
Originally Posted by Heinrich S View Post

You know, that DBT testing is final, that no further investigations are necessary. Just get a null result and that is that. It is possible to get null results based on faulty testing procedures, but all proponents here don't care about further investigation. No independently verified testing is necessary, just cite a test and it is final. That to me is not scientific. You kind of want the magic bullet, but I don't buy into that kool-aid- I'll rather try to get a second opinion and have the results verified by another party that can remain objective. You know, not bigoted and prejudiced to the nth degree. No one knows how many "good" implementations of DBTs have been done. It's all speculation.
With the vast number of say amplifiers or DAC's today, just how would you propose they be tested and for that matter independently verified? And FWIW, I don't think anyone here is arguing that if one were to level match at one particular frequency, say 1kHz, that something like a high output impedance SET or OTL amp could not be differentiated from some arbitrary solid state amp, say like something from Emotiva. Nor do null results imply that any amp can be substituted for another with a given set of speakers and be expected to perform similarly.

As to accepting all DBT results as gospel, I turn your attention to the oft cited MatrixHiFi results (Spanish), which upon further disclosure by a person who participated in them, have been criticized for a number of reasons. Even if I suspect that their conclusion may have been correct, I and others cannot in good conscience hold them up as being in any way definitive.

But consider, Heinrich, that under the conditions of many of the comparisons you may have read about, Clark's for one (if you didn't know, ArnyK not only knows David Clark by his writings but also on a personal level), their results are not necessarily surprising. After all, if we know, via measurements that the outputs of two devices are essentially linear into some device and we also know that the under the best of conditions, the human ear in a young, healthy person can differentiate about 0.1 dB in the 1-4kHz range give or take, then if we level match to to that precision or better, there's no apparent reason why one should not expect nulls. After all the human ear is sensitive to two things - level and arrival time. The null results are simply a consequence or reaffirmation of work on human hearing that spans centuries of studies.

OTOH, if one of the devices is something like a NOS (non oversampling) DAC, even with level matching one should be readily able to differentiate due to its rather unique behavior.

None of these null results takes away from the importance of of selecting an amp with sufficient ability to drive your speakers to the levels you prefer at your listening position.
post #1967 of 3048
Quote:
Originally Posted by Heinrich S View Post

Oh boy... here it goes again.... biggrin.gif

Not sure what's so funny. Did you ask him?
post #1968 of 3048
Oh, and, IMO ABX audio tests are in fact subjective. The double blinded nature of the test is simply the most effective currently available means to control for one variable - - sighted bias. It's really only a problem if you can't hear with your eyes closed.

Double blind tests of new pain medications are inherently subjective, too. "does it hurt more or less?" is not a quantifiable independently, objectively masurable issue. On the other hand, for a breakthrough cancer treatment, "did the tumor go away" is objectively detectable and measurable (and probably entirely beyond the ken of the subject, who could offer no opinion one way or the other).

And yes it is of course possible for any scientific test to be set up so that its results are not what the researchers think they are. But, see, it's cience because the methods are known and can be repeated by others, AND, at least if it's published science, there is peer review that at the very least should catch significantly large errors in testing.

To take the Pepsi versus Coke test as anaexample. Coke wanted to test for which soda flavor people would prefer to buy and drink. What they actually tested, though, was which made the best immpression for a sip or two. What was wrong with their methodology was that they didn't realize that what people like a sip or two of may become cloying or too sweet or too something if they're going to drink 12 ounces of it.

But that's part of the scientific method too. You learn from your failures. Or others do.

It seems to me that if it were possible via publishable science to debunk the results of ABX testing in the audio world, there are a few dozen companies with millions of dollars on the line that might be interested in funding the research. So why haven't we seen it?
post #1969 of 3048
Quote:
Originally Posted by JD in NJ View Post

Maybe it's the software developer in me that feels so strongly about this. Responsibilities should be partitioned in order to create truly modular systems, in which one component can be swapped for another deliberately and with clear expectations as to the result.
Yes, and I think this is sort of the sound engineer's credo. We can modularize the task of sound reproduction, optimize each module, then it follows that the entire task will be optimally performed. And so any progress at optimizing a module will be progress at optimizing the task. I have enough sympathy with this view to commiserate, but I'm very skeptical. Human tasks can be modularized, but can Nature actually be modularized? Specifically, earlier in this thread, I expressed some doubt about the assumption that minimizing the distortion of amplifiers would minimize the distortion of the reproduction system.
post #1970 of 3048
Quote:
Originally Posted by Chu Gai View Post

the human ear in a young, healthy person can differentiate about 0.1 dB in the 1-4kHz range give or take,
More like 0.5 db. It varies depending on frequency. http://www.audioholics.com/education/acoustics-principles/human-hearing-amplitude-sensitivity-part-1
post #1971 of 3048
Quote:
Originally Posted by markus767 
Not sure what's so funny. Did you ask him?

Well given his previous attempts at revealing hundreds if not thousands of negative DBT results based on an AES paper that we later found out to show no such thing, I'm not particularly inclined to be reliant on his sources as so far they have turned out to be duds. So based on that unreliable data, asking him to cite independently tested DBTs would seem to be a bad idea- if he cites anything we'll all have to double check and triple check his references to make sure it's actually legitimate.
post #1972 of 3048
There's also a program called AudioDiffMaker where one can test two devices, time sync them, and then subtract one from the other to see or hear just what the audible differences are. Free too.
post #1973 of 3048
Quote:
Originally Posted by GregLee View Post

I still don't understand how you reach this conclusion ... is there some general reason to think that a "normal" distortion will never counter existing distortion?

I don't know how I can explain it any more clearly. I showed the two basic distortion mechanisms, and the first image shows how you need to exaggerate the waveform peaks in order to counter clipping. Then I explained that exaggerated peaks is not a "natural" form of distortion that occurs in normal audio gear. So you won't find two audio devices that complement each other to cancel distortion. All you'd ever get passing through a second device is even more distortion.

If you think there's some way that one device's distortion could counter distortion in another device, maybe you can draw it out on paper as successive waveforms and post it here so we can see your theory.

--Ethan
post #1974 of 3048
Quote:
Originally Posted by Ethan Winer View Post

If you think there's some way that one device's distortion could counter distortion in another device, maybe you can draw it out on paper as successive waveforms and post it here so we can see your theory.

post #1975 of 3048
Quote:
Originally Posted by diomania 
Then why are you here?

Since when do I need to give a reason for posting? Just because I don't agree with your "high and mighty" DBT methods does not mean I don't find other topics interesting to discuss. There is more to audio than just DBTs, I wish some of you would acknowledge that.
Edited by Heinrich S - 3/8/13 at 11:11am
post #1976 of 3048
Quote:
Originally Posted by JD in NJ View Post


I don't know what that advertising graphic is supposed to show, but it certainly doesn't show one typical audio device countering distortion in another. As I already explained, it's possible to design a special circuit to exaggerate peaks, but that's not the same as "matching" normal audio components to counter distortion.

--Ethan
post #1977 of 3048
Quote:
Originally Posted by Ethan Winer View Post

I don't know what that advertising graphic is supposed to show, but it certainly doesn't show one typical audio device countering distortion in another.

--Ethan

It's not an advertising graphic and if anyone uses it as such I'll sue. I just slapped it together myself. It was a joke.
post #1978 of 3048
Quote:
Originally Posted by JD in NJ View Post

It's not an advertising graphic and if anyone uses it as such I'll sue. I just slapped it together myself. It was a joke.

Another Shakti "innovation"...

post #1979 of 3048
Quote:
Originally Posted by Ethan Winer View Post

So you won't find two audio devices that complement each other to cancel distortion. All you'd ever get passing through a second device is even more distortion.
But, but, but ... you've noted above that equalization systems are exactly such devices. They do complement other devices in the system to cancel distortion. So what you say here can't possibly be right.
post #1980 of 3048
Quote:
Originally Posted by GregLee View Post

But, but, but ... you've noted above that equalization systems are exactly such devices. They do complement other devices in the system to cancel distortion. So what you say here can't possibly be right.

Deviations in frequency response isn't normally counted as distortion, so what other distortions do you mean they cancel?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Are audio companies all involved in a huge conspiracy?