What can measurements show/not show?

[jkeny]

I really don't know the correct place for this thread?

Just when a thread was getting interesting & about to have a breakthrough it was closed down Modifying USB cable to supply 5v

Sy, I believe, was going to post plots of analogue outs clearly showing jitter, sound stage, instrument timbre, sonic tails & all the things we audio freaks crave.

He claims that anything that can be heard is measurable so I'm presuming he either has these measurements & can produce them or he has never heard the effects of which I speak.

Maybe he can produce them here & others can give their measured opinion as to heard Vs measurements?

[Conrad Hoffman]

IMO, any signal defect that can be heard is measurable. IOW, if by signal you mean the time vs voltage characteristics measured on a wire, say between the input and output of an amplifier, there's no mystery there. More sensitive or more accurate instruments aren't going to reveal anything we don't already know.

Now, once you get into transducers and enter the 3-dimensional airspace, the ear and the brain, things get way complicated. We still have pretty good measurement capability, and more is known about perception than is sometimes acknowledged, but the number of variables seems huge and I don't think predicting how the brain will interpret them is very far along.

So, the electrical journey is well understood, but don't make the mistake of thinking that "best" sound is necessarily perfect reproduction, where noise and distortions have been reduced to zero. The acoustic journey is way complicated, transducers are far from perfect and even though we can measure the defects, you and I might prefer entirely different designs. Thus, the audio hobby will happily continue.

[jkeny]

Thanks Conrad, for your considered reply - I pretty much agree with what you say & don't believe for a moment that low distortion = best sound! it's well known that the ears introduce 2nd order distortion & who knows how the ear/brain interface behaves?

I was specifically asking about audible characteristics in the sound like sound stage, timbre, instrument tails & decay, etc.

Some people (SY & others) claim that anything that can be heard is measurable - what do you think?

I'm waiting for SY to produce the analogue plots that show this & I've been asking for a while now but so far no show!

[jcx]

Originally Posted by jkeny

...
...Sy, I believe, was going to post plots of analogue outs clearly showing jitter, sound stage, instrument timbre, sonic tails & all the things we audio freaks crave.

He claims that anything that can be heard is measurable so I'm presuming he either has these measurements & can produce them or he has never heard the effects of which I speak.

Maybe he can produce them here & others can give their measured opinion as to heard Vs measurements?

actually you have the presumption backwards - we can measure many types of signal distortion, frequency response, phase, and jitter effects on the analog signal - the issue is getting any reliable evidence of audibility thresholds in controlled perceptual testing

grab a handful of audiophile subjectivist speculations with "night and day differences" which seem to largely disappear under controlled testing with blinding protocols and then claim "you can't measure it" – seems to be begging the big question

as long as the "Golden Eared" refuse to use even basic controls that the rest of the world uses when turning human perceptual responses into reproducible data that can validate theoretical models and give quantifiable predictions there won’t be advances in measurement correlation with audibility

it is not that the problem of perceptual testing is too hard – the development of losey perceptual encoders that can “throw out” >75% of the Shannon-Hartley “channel information capacity” with vanishingly few able to detect the difference for the majority of music shows audio perceptual testing can give significant results

[jkeny]

Originally Posted by jcx

actually you have the presumption backwards - we can measure many types of signal distortion, frequency response, phase, and jitter effects on the analog signal - the issue is getting any reliable evidence of audibility thresholds in controlled perceptual testing

grab a handful of audiophile subjectivist speculations with "night and day differences" which seem to largely disappear under controlled testing with blinding protocols and then claim "you can't measure it" – seems to be begging the big question

as long as the "Golden Eared" refuse to use even basic controls that the rest of the world uses when turning human perceptual responses into reproducible data that can validate theoretical models and give quantifiable predictions there won’t be advances in measurement correlation with audibility

it is not that the problem of perceptual testing is too hard – the development of losey perceptual encoders that can “throw out” >75% of the Shannon-Hartley “channel information capacity” with vanishingly few able to detect the difference for the majority of music shows audio perceptual testing can give significant results

It's not my presumption, JCX, it's SY's - he claims that anything audible is measurable! I sense you disagree with this?

So can I ask you if you have ever heard an increase in sound stage? a more realistic instrument timbre? etc. If you have, have you ever measured the difference in analogue output before the increase & after? Can you show them please? Everything we hear is measurable, right? So these are measurable, correct? Show me please! Otherwise, I presume these are not yet measurable phenomena?

Now if you are going to reply that these are very subjective terms then I take it that you have never experienced such an improvement in your audio set-up?

[jkeny]

Still no plots from SY - I PMed him to let him know about this thread!

I was hoping that all this speculation could be avoided with his plots as he has stated that anything audible is measurable!

[jcx]

I had a thoughtful reply prepared but seeing your continued hectoring I'm not interested in feeding a Troll...

[jkeny]

Any thoughtful reply would be appreciated but I'm also interested in hearing from Sy & seeing his plots. I'm presuming you don't have these plots of analogue out showing sound stage, etc? Or I guess you would have posted them!

Can all that we hear be shown on an analogue out plot or not?

I'm going to take it from the silence & the lack of evidence to the contrary that this is proven to be a falsehood & anybody stating this in the future is also just promulgating a falsehood!

I have an open mind on this but need evidence to show that these sonic characteristics that we hear are actually evident on a plot of analogue out! If this is not the case then please stop pushing this line of needless argument & wasting everybody's time, energy & internet bandwidth!

[abraxalito]

Originally Posted by jkeny

Sy, I believe, was going to post plots of analogue outs clearly showing jitter, sound stage, instrument timbre, sonic tails & all the things we audio freaks crave.

I didn't get the impression that that was going to happen any time soon. You and he were talking past each other rather a lot

He claims that anything that can be heard is measurable so I'm presuming he either has these measurements & can produce them or he has never heard the effects of which I speak.

I think as far as SY's claim goes, its a theoretical stance. In theory I also agree with him But practically speaking he's a long way from proposing the kind of measurements which correlate with what we hear. His lexis is distortion, crosstalk, sidebands and yours is soundstage, timbre, reverberation. A bridge between the two ain't going to come from either of you methinks.

For just one example, he has claimed that the differences could be seen in the output spectrum. I picked up on his claim, asking for more details, he provided the proposed acquisition length of 8192 samples. I've not received a reply to my latest question on that and I'm not holding my breath

Maybe he can produce them here & others can give their measured opinion as to heard Vs measurements?

A forlorn hope IMO.

[jkeny]

Well, abraxalito, I believe his statement was that anything that can be heard, can be measured. I take it that he can hear differences in sound stage & timbre between different devices - if not we have a completely different frame of reference & no more discussion is needed.

I put it to him to show these differences in analogue output plots before he demands of others to show these differences. He says it's so simple! If he can't do this then his whole argument is a fabrication!