Master Thread: Are Measurements Everything or Nothing?

[Newman]

I have been fooled so many times comparing different mixes (LP vs CD, CD vs remastered CD, CD vs Hi-Res, etc.) that I now always use Foobar ABX. While the sighted part can be important even more important is the "level matched" part, it is amazing how very small level differences change our perception. I am much more willing to accept "sighted" comments if I know the comparison is accurately level matched.

Although note that Toole's experiments, demonstrating the futility of sighted listening if you want to know your own opinions of the attributes of the sound waves themselves, were all accurately level-matched.

 

[Geert]

“ in what respect is sighted listening useful, and when can it make sense to provisionally accept the results of sighted listening?

At least we need controled listening tests for observations that challenge the scientific concensus, more specifically things that are unmeasurable, theoretical impossible and are never proven to be true in a controlled test. We also need them to verify observations of small differences between different components. Like the kind of difference that don't even show up in a null test. (And since you mentioned Toole; going by his research "small differences" is an understatement).

Don't think the use of circuits to create a different sound, like tone controls, or the evaluation of different mixes that were purposely created to sound different are in scope of the above.

 

[MattHooper]

I have been fooled so many times comparing different mixes (LP vs CD, CD vs remastered CD, CD vs Hi-Res, etc.) that I now always use Foobar ABX.

So does this mean every time you are listening to a different master for the first time you are putting it to an ABX test?

Just curious. Not that there is anything wrong with seeking those confidence levels, but it is a sort of “ only on ASR” thing to do.
Gotta love this place.

While the sighted part can be important even more important is the "level matched" part, it is amazing how very small level differences change our perception. I am much more willing to accept "sighted" comments if I know the comparison is accurately level matched.

Good post. I’ve had some similar experiences level-matching some stuff.
Differences can start to become surprisingly subtle.

(Though nothing in your post undermines my point I’ve been getting at. The consequences of what you are talking about would be essentially pointless if one believes sighted listening render one incapable of telling real differences when they are there anyway).

 

[Newman]

Don't think the use of circuits to create a different sound, like tone controls, or the evaluation of different mixes that were purposely created to sound different are in scope of the above.

Quite right. The repeated mentions of Toole's suggestion of a use case for tone controls, as if Toole was saying everyone should adjust tone for personal preference, has to be deliberate obfuscation by now. It has been explained numerous times how wrong is this idea, but it bounces back in the manner of a denier confronted by an inconvenient truth.

 

[MattHooper]

At least we need controled listening tests for observations that challenge the scientific concensus, more specifically things that are unmeasurable, theoretical impossible and are never proven to be true in a controlled test. We also need them to verify observations of small differences between different components. Like the kind of difference that don't even show up in a null test. (And since you mentioned Toole; going by his research "small differences" is an understatement).

Exactly! This is one of the points I’ve been making. In pragmatic terms, the demand for measurements and controlled listening should scale with the nature of the claim: the more technically improbable the claim, or the more subtle the possible sonic differences, the more that calls for extra rigour.

(Again in practical every day situations where we aren’t dealing with scientific controls. But if one is seeking scientific reliability, then you can apply scientific controls to any claim at all)

Don't think the use of circuits to create a different sound, like tone controls, or the evaluation of different mixes that were purposely created to sound different are in scope of the above.

Right. If you are going to tweak the bass or treble with tone controls, it is reasonable to think you can apprehend the differences under sighted listening.

You could actually be wrong! There are cases where you can have a broken tone control and turn it and imagine you’ve had a bit more bass or treble. But those instances tend to be anomalous, and generally speaking it’s reasonable to presume some reliability to what you are hearing in your normal listening conditions. (always with the caveat one could be wrong.)

 

[MattHooper]

Speaking of “ inconvenient” truths..,

The repeated mentions of Toole's suggestion of a use case for tone controls, as if Toole was saying everyone should adjust tone for personal preference, has to be deliberate obfuscation by now.

Newman, could you please stop misrepresenting Floyd Toole?

Quotes from Floyd Toole From this thread:


“Old fashioned bass & treble tone controls and modern "tilt" controls are the answer and they can be changed at will to compensate for personal taste and excesses or deficiencies in recordings”

….

“Better to have easily accessible tone controls that can be instantly adapted to your personal preference - for any program.”

So it is quite clear that Toole allows for adjusting the sound on ANY program based on personal preference. Floyd has even said in the past he up-mixes virtually all his stereo sources to surround, based on his personal taste.

So who is actually doing the “obfuscating” on this issue? It’s one thing to routinely misrepresent what I write, but fibbing about Floyd Toole….?

After doing the right thing and retracting your misrepresentation, you might then also want to address why tone controls are useful, under sighted conditions.

 

[Purité Audio]

Tone controls aren’t the same as wilfully adding distortion.
Keith

 

[MattHooper]

Tone controls aren’t the same as wilfully adding distortion.
Keith

If you are using tone controls to adjust program material, that is a wilful departure from accuracy, right? You are changing the sound in a way that pleases you.

Once you open that door, you are going to have a problem making the case “ It’s only OK when I choose to do it…”

 

[Purité Audio]

No distortion is something that wasn’t there in the first place.
Keith

 

[Axo1989]

Tone controls aren’t the same as wilfully adding distortion.

Adding harmonic distortion is often less audible.

No distortion is something that wasn’t there in the first place.

Ok, you need a moment with the (technical) dictionary there.

 

[MattHooper]

No distortion is something that wasn’t there in the first place.
Keith

Uhm… the additional bass or trouble frequencies weren’t there in the first place in the recording either.

That’s why you’d be adding them right?

 

[MattHooper]

Adding harmonic distortion is often less audible.

That’s the point I’ve made before. I’m sure Keith is referencing by using a tube amplifier. And yet the amount of distortion, audible deviation from the signal, it is likely adding is probably less than what people are adding with their controls or EQ.

Ok, you need a moment with the (technical) dictionary there.

Yup.

Or you can even consult ChatGPT:

In sound reproduction, distortion refers to any change in the original audio signal that causes the output to differ from the input.

Sometimes people are just trying to justify their own choice to distort the signal.

 

[Purité Audio]

The room could and often does add bass and many crappy speakers add treble.
Keith

 

[MattHooper]

The room could and often does add bass and many crappy speakers add treble.
Keith

That’s changing the subject.

The discussion was about using tone controls to adjust recordings themselves to taste. (fixing speaker or room issues is a different discussion.)

 

[antcollinet]

If you are using tone controls to adjust program material, that is a wilful departure from accuracy, right? You are changing the sound in a way that pleases you.

Once you open that door, you are going to have a problem making the case “ It’s only OK when I choose to do it…”

1 - Tone controls are typically used (crudely compared with DSP) to rebalance the speaker/room response from something that sounds wrong (typically bright or dull) to something that sounds more balanced. In other words, like DSP, pushing the response closer to accuracy, albeit "by ear"

2 - Or, for some, adjusting the tone of particular recordings that may have been redorded bright or dull.

In both cases the adjustment is tunable (not fixed), and can also be defeated.

None of the above is true for baked in amp distortion.

 

[Purité Audio]

If you are adding distortion to every song how is that adjusting for individual records.
You like adding distortion that’s fine I don’t.
Keith

 

[Axo1989]

That’s the point I’ve made before. I’m sure Keith is referencing by using a tube amplifier. And yet the amount of distortion, audible deviation from the signal, it is likely adding is probably less than what people are adding with their controls or EQ.

Yup.

Or you can even consult ChatGPT:

In sound reproduction, distortion refers to any change in the original audio signal that causes the output to differ from the input.

Sometimes people are just trying to justify their own choice to distort the signal.

Kieth is presumably asserting that nonlinear (ie adding harmonics that weren't there in the original signal) is distortion, but linear (ie changing amplitude or phase) isn't. Which is surprisingly sloppy semantically. Whether the mechanism used is adjustable/reversible or not is orthogonal to the issue of reliability (or otherwise) of what we (think we) hear.

 

[Axo1989]

1 - Tone controls are typically used (crudely compared with DSP) to rebalance the speaker/room response from something that sounds wrong (typically bright or dull) to something that sounds more balanced. In other words, like DSP, pushing the response closer to accuracy, albeit "by ear"

2 - Or, for some, adjusting the tone of particular recordings that may have been redorded bright or dull.

In both cases the adjustment is tunable (not fixed), and can also be defeated.

None of the above is true for baked in amp distortion.

Accuracy "by ear" ... heresy, surely?

I agree of course, we can certainly adjust the signal to be more correct (ie closer to the original signal) at the listening position to counter the effect of the room (or deficiencies in the loudspeaker). It is a somewhat different discussion.

 

[MattHooper]

1 - Tone controls are typically used (crudely compared with DSP) to rebalance the speaker/room response from something that sounds wrong (typically bright or dull) to something that sounds more balanced. In other words, like DSP, pushing the response closer to accuracy, albeit "by ear"

Yes, of course but as mentioned that wasn’t what we were talking about.

2 - Or, for some, adjusting the tone of particular recordings that may have been redorded bright or dull.

In other words adjusting to personal taste.

In both cases the adjustment is tunable (not fixed), and can also be defeated.

None of the above is true for baked in amp distortion.

Of course, but Keith was sort of all over the map in his replies, with some dubious inferences.

If you are adding distortion to every song how is that adjusting for individual records.

It isn’t. Someone can prefer an altered signal for just some recordings, someone else may prefer an altered signal for all recordings.

It could take the form of somebody who leaves their bass control permanently up a bit because prefer that, or somebody who uses a tube amplifier who finds the effect pleasing across all recordings. And again, amplifier distortion may be less obvious than people using tone controls to adjust their preferences.

Adjusting the sound to personal taste is what it is.

 

[antcollinet]

Accuracy "by ear" ... heresy, surely?

Burn the witch!


[python]how do you know he is a witch [/python]