Misleading Measurements

[pkane2001]

 

Just now, Audiophile Neuroscience said:

 

The validity and reliabilty would need independent verification. I don't think you can expect users to verify your app for themselves.

 

Your opinion matters to us. If you don't like what the software does, a full refund will be issued for your entire purchase price.

 

[pkane2001]

2 minutes ago, Audiophile Neuroscience said:

 

So subjective opinion determines a pass or fail in a measurement app?

 

Pass or fail on purchasing the software and requesting a refund is absolutely driven by opinion.

[pkane2001]

16 minutes ago, bluesman said:

What’s the point of doing that (other than entertainment) unless you know that you’re modeling a real world situation like our equipment?

Already answered that.

[pkane2001]

1 minute ago, Audiophile Neuroscience said:

 

or you could ask for evidence that the app produced faulty measurements. In this case, left to the user to self-verify, if they say it doesn't work then it doesn't work

 

The app doesn't produce measurements, so they can't be faulty.

[pkane2001]

10 minutes ago, Audiophile Neuroscience said:

 

So not this app that can "verify against calculations" but without measurements?

So it designs or simulates any non-linearity and see what it does to sound, subjectively.

 

 

 

Measurements are up to the buyer, the app is designed to create distortions and not measure things. Educated consumer is our best customer. I provide warranty that it does what I say it does, or your money back (within 30 days).

[pkane2001]

6 minutes ago, Audiophile Neuroscience said:

Jeez, the user has to subjectively verify the app themselves and take their own measurements to confirm their verification. Sounds like fun. I'm out  😉

Don't forget to fill out the refund form

 

 

[pkane2001]

6 hours ago, Bill Brown said:

My impression (I don't have sources, perhaps it was Earl Geddes) suggest that air is largely linear except at very high pressures (not at the levels our speakers can produce).  The discussions I have seen are in regard to compression drivers, but even there (huge pressures) I don't believe it is non-linear.

 

Bill

With very large amplitude sound waves (extremely loud) the air becomes nonlinear. Even the air temperature can change locally due to high compression at the peaks of the waveform. Localized compression and temperature change result in faster sound wave propagation at the peaks than at the troughs, where the air gets rarefied  and cooler. The result is a distorted sine wave that might look more like sawtooth than sine.

[pkane2001]

43 minutes ago, sandyk said:

 

What has any of that got to do with normal listening to music in a typical entertainment or optimised Audio room ?

 As for temperature changes due to Output levels in any normal listening environment . . . . . . .

 

Why so closed minded, Alex? In a less typical entertainment environment, it could make for a very interesting speaker system, as you can see below:

 

[pkane2001]

Just now, sandyk said:

 Perhaps in Academic circles, however it has little if any relevance to the topic of Misleading Measurements.

 Perhaps you should start another thread in the General forum about your S/W just as John Dyson has done with his ?

 

Perhaps, perhaps, but I'm not the one who keeps asking if air is non-linear in this thread.

So, perhaps you should ask them to start another thread? 

 

[pkane2001]

5 minutes ago, sandyk said:

 

Do you want this to become another FrankenGeorge type thread ?  😉

 

I'm not that concerned, to tell you the truth 😃

 

[pkane2001]

1 hour ago, bluesman said:

I agree - but only if you know what the tool is doing and how.  Introduce unspecified nonlinearities to a signal path to add distortion products seems to be of little or no value to engineers, designers, et al.  It may be entertaining and even educational for audiophiles to see if they can hear differences.  But having no idea what was done to the signal seems to me to make it impossible to relate to HW or SW design.

 

Don't know if I mentioned it already (five times, maybe?) but the software lets you configure your own non-linearity. You specify it by entering the amplitudes of the desired harmonics. Take a look at any FFT plot of a real device, of say a 1KHz sine wave, find the amplitudes of the harmonics, and enter them in. The exact non-linearity will be generated to produce that set of harmonics, in effect simulating the original device. You can then feed a sinewave or music or a two- or three-tone IMD test signal through the nonlinearity and confirm that it looks like the original device. I've done this many times as part of testing using various DACs and preamps.

 

 

DISTORT also has a number of different simulation non-linearities built-in, like a triode-based preamp or various compressor types. 

 

Here are examples of some built-in transfer functions:

 

 

I can add the ability to enter your own mathematical expression for the non-linearity, will that work for you? How would you find out the expression for the non-linearity of a specific device to enter into this?

 

[pkane2001]

20 minutes ago, bluesman said:

And that’s where we diverge in opinion. Simply recreating a set of harmonics does not in any way confirm that the transfer function you inserted is in any way a simulation of any specific function of the original device.  It also can’t confirm that the observed harmonic distortion is the sole effect on the signal, or that it has the same effect(s) at the same point(s) in the signal path as the cause or causes in the original device.

 

You also don’t know if your intervention causes the same set of changes in SQ.  All you can say with certainty is that your intervention caused the addition of a given set of harmonics that are also generated at one or more points in the original device.
 

You can add the flavor of almonds to food with almond extract or with cyanide.  The taste is the same, but...........
 

 

 

Maybe yours is an opinion, mine is the result of actual testing and validation. As I said, I tested this on multiple devices. If harmonic content and IMD content match that measured using the physical device, then it's a pretty good simulation, IMO (ok, so that last part is an opinion, but based on measurements and experiment).

 

[pkane2001]

39 minutes ago, bluesman said:

There are many pieces of audio gear that sound grossly different despite identical distortion measurements.  Could those measurements possibly be misleading? Or maybe there’s a lot more to that “wire with gain” than how much IMD and THD it generates.

 

Have you considered the possibility that your interventions have more effects on the signal than those you’re focused on and measuring?  Is it at all possible that distortion measurements alone could be misleading you?  

 

You obviously don’t consider the efforts and results I posted earlier in this thread to be measurement, experimentation, testing or validation.  I’ll just have to pull myself together once I finish grieving over your disapproval.

 

PS: nice catch on your “IMO”. You were on the verge of contradicting yourself.

 

It would seem that you are consistently misinterpreting what I say or write based on your perception of "my agenda", at least that's what it sounds like to me. You keep bringing up things I've not said or claimed and attributing them to me. I really don't get it. Would it not be better to try to have a discussion with me, the real person, rather than with some idealized "objectivist" impression you seem to be arguing with? I'd certainly appreciate it.

 

Where have I claimed that distortion is the only measurement that affects audibility? Point me to that message, please. Or that THD or IMD are the only measurements that are important? Heck, I have spent many months building many different types of distortion simulations into DISTORT (non-linearity & compression, multiple different kinds of jitter, different dither types, noise shaping, amplifier cross-over distortion, oversampling and filters, negative and positive feedback, various types of noise, etc.) I have plans to add a number of others. It's not because I think these are all completely unimportant, or else I wouldn't waste the time. 

 

But, let's ignore that for now. This discussion is not really about DISTORT, or at least it shouldn't be.

 

What effort of yours and results did I ignore, can you please review these for me (or point to the relevant posts), as I might have missed them? What was it that you tested and what was the finding? I've seen a lot of opinions and arguments, but somehow missed any of the facts that contradict anything I've said here.

 

[pkane2001]

17 minutes ago, bluesman said:

And it isn't.  Go back to posts 580 and 620 for a start.  You (and others - you're not alone) keep pushing the fact that all intermodulation is distortion and is generated by nonlinearity.  You're the one who introduced your app in support of this belief - I couldn't care less about it.  I'm suggesting and supporting the belief that not all intermodulation is distortion caused by nonlinearity.  Musical notes intermodulate naturally in the air because their compression and rarefaction waves collide with the molecules in the air around them, pushing those that are randomly minding their own business into sum and difference waves that are heard and recorded as part of the program material.

 

Those intermodulations (which are NOT distortion in the true sense, as they arise from and are part of the source signal) are then recreated again from the reproduced program material being thrust into the air by speakers.  This is in addition to and apart from any intermodulation distortion caused by nonlinearities in the signal chain.  Your repeated assertion that your app proves that all IM and harmonic distortion stems from nonlinearities is simply not correct, in my opinion.  I support my opinion by having recorded pure tones with their natural intermodulation products, and showing that those intermodulation products persist in the recorded waveform even after filtering out the fundamentals that generated them.

 

Another contributor believes that these natural IM products are coming from resonances in the instruments themselves.  I cited well done research by others showing that there are no resonances in a solid body guitar anywhere near the sum and difference tones in my demo, which I believe also refutes this belief.  Similar research shows a lack of resonance anywhere near the intermodulation products found in the playing of flutes, oboes, and many other instruments.

 

Because of my belief, I'm suggesting that distortion measurements alone are misleading.  I believe that it's possible to separate the natural IM in the program from the IM created by playback of the program (entirely apart form any IMD introduced by the system).  I don't know how yet - it may be phase differences, amplitude differences, or perhaps use of real time FFT to differentiate natural IM in the source program from the products of intermodulation between the recorded natural IM and that generated by its playback. 

 

I think this added layer of intermodulation is at least part of that famous veil that we all want removed from our music on playback.  But "IMO", believing that all IM is the product of nonlinearities and is distortion is counterproductive. And if I'm correct, inducing distortion as you advocate for evaluation, testing and development would be the wrong way to approach the problem.

 

I think there's probably a terminology issue here, so perhaps we should first define our terms. I'm using the following definitions as they are normally used in audio:

 

Distortion: any frequency content that was not present in the original signal

Intermodulation Distortion: amplitude modulation of one signal by another caused by a non-linearity in the signal chain

 

From Wikipedia: 

Quote

Intermodulation (IM) or intermodulation distortion (IMD) is the amplitude modulation of signals containing two or more different frequencies, caused by nonlinearities or time variance in a system. The intermodulation between frequency components will form additional components at frequencies that are not just at harmonic frequencies (integer multiples) of either, like harmonic distortion, but also at the sum and difference frequencies of the original frequencies and at sums and differences of multiples of those frequencies. Intermodulation is caused by non-linear behaviour of the signal processing (physical equipment or even algorithms) being used. 

 

I highlighted the last sentence to show that it's not just my opinion when I state this.

 

Now,

1. If you are saying that IMD is coming from something other than the non-linear behavior of some device or medium, then you are using a different definition, according to the above
    
2. If you are saying that IMD is not a distortion, then you are also contradicting the definition, since by definition, IMD introduces new frequencies that were not present in the original signal and therefore it is a form of distortion

Now, the two posts you mention:

 

580: all seems correct to me, what's the issue here?

620: Let me understand: you found some, what you're calling, IMD, in the capture of a guitar tone. Could it be simple amplitude modulation? Not all amplitude modulation is IMD, but all IMD is amplitude modulation.

 

Quote

Your repeated assertion that your app proves that all IM and harmonic distortion stems from nonlinearities is simply not correct, in my opinion.  

 

You are again misquoting what I said. I said, and repeated, and will repeat one more time: IMD and HD are caused by the same non-linearity. What this means, is that given a non-linear audio device, HD will be result when testing with a single tone, IMD will be the result of multiple tones passing through the same non-linearity. THAT IS THE EXTENT OF MY CLAIM. Anything else you misread or misinterpreted. 

 

Quote

And if I'm correct, inducing distortion as you advocate for evaluation, testing and development would be the wrong way to approach the problem.

 

Where did I advocate using distortion for evaluation, testing or development? DISTORT is an audibility testing tool to simulate various distortions. It is meant to let me or others evaluate the degree of how various distortions sound using a mathematical model of the distortion applied to any recorded piece of music or test signal. Nothing more.

 

Quote

I think this added layer of intermodulation is at least part of that famous veil that we all want removed from our music on playback.

 

At least that's something you could easily test for yourself with DISTORT. But you refuse, so that's on you. You have nothing to back this up, except for an opinion.

 

Quote

I believe that it's possible to separate the natural IM in the program from the IM created by playback of the program (entirely apart form any IMD introduced by the system).  I don't know how yet - it may be phase differences, amplitude differences, or perhaps use of real time FFT to differentiate natural IM in the source program from the products of intermodulation between the recorded natural IM and that generated by its playback. 

 

Just check out my other software, DeltaWave (I hate to bring this up, as this will likely start another tangential argument). It does exactly what you want: separate all the distortions caused by the reproduction chain from those present in the source material. Yes, including IMD, AM, FM, phase, or amplitude, or any other kind of distortion.

 

Quote

But "IMO", believing that all IM is the product of nonlinearities and is distortion is counterproductive.

 

By definition, as stated above, this is incorrect.

 

[pkane2001]

22 hours ago, bluesman said:

You're probably going to respond that I don't understand what you're saying - but I don't have a clue what you mean by distortion "present in the source material".  What kind of distortion do you think there is in a live performance by a symphony orchestra, an acoustic jazz trio, a string quartet or a piano concerto?  And what do you think is causing it?  Do you hear it at a live performance?

 

If the intermodulation products among instruments and voices are distortion, there can be no such thing as an undistorted musical performance.  If that's true, why are we wasting our time trying to reproduce one?

 

I think we are talking about two different things. You're talking about musical instruments, and I'm talking about the playback system to reproduce recorded sound. There's a whole lot that's going in a musical instrument to produce a recognizable sound that we like and recognize. The discussion of music production and sound synthesis is an interesting one, but can I suggest that an audiophile forum and a thread dedicated to audio playback measurements isn't the best place for it? My opinion only, of course, but the context here is important. Everything I've said and quoted and measured and shared applies to the playback system. Seems to me that most of your argument has been about the acoustic instruments, including what you called "natural IM". These are two very distinct topics and deserve separate treatment, so I'm just going to stop here.

[pkane2001]

 

 

7 hours ago, Audiophile Neuroscience said:

 

says who?

 

Those who understand what it does, of course.

[pkane2001]

7 hours ago, Audiophile Neuroscience said:

 

I am not convinced Wikipedia is really our guide of all things scientific but from the same Wikipedia reference

"Distortion is the alteration of the original shape (or other characteristic) of something"

 

IMO we are talking about interference patterns of sound waves in air, specifically beat frequencies. I just don't get why that should be so controversial as a concept

 

Wikipedia was just the simplest, most accessible text. Any number of textbooks and papers say the same thing. But it's not a controversy, except to those who want to continue to argue. This topic has run its course, IMO, but you can continue if you'd like.

[pkane2001]

13 minutes ago, bluesman said:

And this thread is a discussion about misleading measurements.  If (as I believe and, to my mind, have supported well) the “interaction” among the notes generated by the instruments in a performance is captured in recording and reproduced in playback, then IMD measurement in the playback equipment may be misleading.  There’s so much more acoustic intermodulation (which is not distortion) being heard in playback that the minuscule added IM products from the equipment may well be immaterial to SQ.
 

This is firmly on topic. It just seems to be a difficult concept for many to grasp. That does not make it wrong.

 

Good, so you made a claim. I've not disagreed with it, in fact, I stated that I make no claims about the audibility of HD or IMD or their importance in the audio chain.

 

Now, can you substantiate your claim with something other than your personal opinion? This is an objective forum, after all.

 

[pkane2001]

5 minutes ago, Audiophile Neuroscience said:

 

Understanding does not confirm validity or the truth or value of a tool. Neither does subjective or anecdotal appraisals nor the assertions of the App or tool creator. It's fine if you wish to make amusing toys but not fine for useful tools,  leading to unacceptable errors and misleading measurements/conclusions.

No, of course not, but understanding what something does and how, is a good first step to having a rational discussion. Any discussion prior to it is just a waste of time.

[pkane2001]

2 minutes ago, Audiophile Neuroscience said:

Totally disagree.

Shocking! I’m so surprised 😮 

[pkane2001]

2 minutes ago, Audiophile Neuroscience said:

 

I agree Paul that understanding of how a tool is supposed to work or what and how it does it, is a good first step. That does not equate with validity, which is the point here. Any rational discussion based on the use of a tool must be preceded by how valid the tool is. This has been a recurrent theme in audio fora in the quest against misleading measurements and faulty conclusions.

And constantly raising questions about the validity of something you don’t understand, have not studied or even tried to use, is what you’re  doing here. If you want to have a rational conversation about validity, then let’s have it, but not before you take that first step.

[pkane2001]

Just now, Audiophile Neuroscience said:

 

You are entering another loop, as we have already mentioned ways of validating tools that is not dependent on the end user which is clearly inherently flawed.

.

You are clutching at straws. People who produce tools need to provide validity data, not some nonsense like 'hey, you wouldn't understand'. That's just another way of saying 'believe me because I say so'. Sorry, I don't.

David, you have no clue what you’re talking about. The tools are not designed to measure anything or to produce any subjective results. They do exactly what they were designed to do and have been validated through independent testing. You may not understand it, but that’s ok. I’m done discussing this with you, as you have nothing to contribute to a rational discussion.

[pkane2001]

3 minutes ago, Audiophile Neuroscience said:

 

Great, please provide this - that's all I have been asking. neither unreasonable or irrational IMO.

Take the first step.

[pkane2001]

1 minute ago, Audiophile Neuroscience said:

 

The first step is not mine to make. You cannot or will not provide objective evidence that your apps "do exactly what they were designed to do and have been validated through independent testing". It appears you have entered one of your loops to obfuscate this fact. So I'll leave you to play with your toys 🙂...../end loop

Evidence is out there, easily located, if you look for it. I don’t need to provide it to you, since you’ve shown no interest in understanding of what the software does, what such evidence could even consist of, or why it might be needed. Any discussion now will be a waste of time.
 

Take the first step to understand it, ask a question that makes sense in the context of the software, and then we can discuss your specific validation requirements.

[pkane2001]

1 minute ago, bluesman said:

 

 

I just don't have the energy to go down this road again.  If you read the following posts in this thread, you'll find both my subjective reasoning and the objective data that support it. 

 

516, 521, 525, 530, 535, 536, 539, 543, 563, 567, 580, 582, 585, 587, 588, 604, 606, 609, 615, 617, 620, 627, 698, 702, 716

 

There are spectral analyses of two simultaneous tones (~220 and ~246 Hz) with their intermodulation products (or whatever the heck you want to call them, as long as you don't call them distortion because they're not).  You'll find the spectral analysis of a wav file of the two guitar notes with everything above 50 Hz filtered out, showing that the difference tone of 26 Hz remains in the recording and is therefore not the psychocacoustic product of the two fundamentals. 

 

Recording a playback of the wav file shows an increase in the level of the first difference tone, and repeated recordings the playback of each successive wav file show further progressive increase as the intermodulation products are added on top of what's already on the recording.

 

QED

 

So, as long as the IMD introduced by the playback equipment is relatively small compared to the IM that produced by the guitar string you've tested, then it's not significant. Got it. Can we test this audibility part? How small does it need to be, relatively speaking? And if the IMD frequencies generated by the equipment have a different amplitude relationship than those produced by the guitar, will this make it more or less audible? 

 

I'm not arguing with you, I'm just asking questions, since these would seem possibly important to understand before declaring all IMD inaudible.