“Human ears are better than measurements!” / “Humans can hear things we can’t even measure!”

[egellings]

Surprisingly, most people are unaware of how much their own ears/brains change from hour to hour, and literally between the start of an album and the end.

In my limited testing on friends (or as some told me, "pranking") it is clear that all of their ears make rubbish measurement tools.

My audio perceptions change with my ambient beer level.

 

[Eric Larson]

> “Humans can hear things we can’t even measure!”

This is true of artificial intelligences too.

The truth is in the verisimilitude of all intelligences, natural and artificial.

 

[Redacted]

> “Humans can hear things we can’t even measure!”

This is true of artificial intelligences too.

The truth is in the verisimilitude of all intelligences, natural and artificial.

Yes, but those things are imagined.

 

[ahofer]

I was referring to 'newcomers in general not you. Is the final test for you how the component fits into your system playing music ?

I was a newcomer in 2019 I think. While a seed of doubt had been planted back when Stereo Review did its blind amplifier test in the1980s, I honestly hadn’t been thinking about it much until I had to replace my system (Thiel, Adcom, various inputs) in 2018. But since joining here, I’ve acquired two systems (one for a second home), and bought most of the equipment based on measurements. Whether I’m satisfied is obviously a matter of listening.

Before that in 2018 I bought a pair of Harbeth speakers through pure listening. I still like them, but honestly, the shopping process was a pain, and the Revels I bought (used) on measurements (F228be) are indeed superior. I listened to Wilsons, B&W, Paradigm, and ATC (disliked), Legacy, JBL (meh) and KEF (liked the high end models), and some others I don’t even remember (Rado?) at three pompous stores with snake-oil pushing salesmen in the Manhattan area. It’s the worst listening to these guys tell you you have to match your cables and your amp to get an ‘musical’ result and turning the volume up steadily behind your back.

I took notice that the ones I like measure better by Toole standards, even before I knew what a spinorama was. A lot of the names I didn’t like just had showroom treble (ahem Paradigm, B&W) and I made a point of listening for at least half an hour with most everything.

My son helped me blind compare my recapped Adcom space heater with a Hypex p252, and that was interesting as well. Sighted, I thought I could tell the difference. Nope.

The other thing that convinced me, since all of the above is anecdotal, is the abundance of evidence on the internet. There’s no excuse to be ignorant of this stuff anymore if you desire to talk about it rationally.

 

[MediumRare]

I appreciate your effort here and agree with the overall premise, but there are a few over-simplifications or errors worth editing, IMO.

- The first hyperbolic statement that microphones are measuring devices and that's the end of the story is not correct. They are "sensing" devices and then an electrical (or computing) process occurs to translate that sensing into a voltage (or a digital version) which can be measured. Human perception is different - and far more - than measuring: it's the result of a psychological construction. That's why, for example, vision is not the same as a photo. One can argue what is better or worse, but they are not the same. But there's no need for that section when you follow it up with the other points.

- Just because we have measurements doesn't automatically mean they add up to human perception - or even an objective comprehensive description of the the DUT's performance. Examples: Amir has selected a handful of measurements to analyze, among them, SINAD at 5 W and 1 kHz. That may be a very useful reference but of course we all also want to see the SINAD at different power levels. And at different frequencies. And we know human perception is more or less sensitive at different frequencies. So even with Amir's data there is a subjective discussion following most reviews (probably not for the Golfing Panthers!) about how these measurements will be perceived (e.g. the noise v distortion elements of SINAD).

Net, if you really want to help people learn and be persuaded the post could use a little humility and be less polemical.

 

[JanRSmit]

Were you responding to my post? If so, which part are you struggling with? Volts or Pascals?

No, my apologies, the first post.

 

[Cbdb2]

Yet a number of independent reviewers have consistently described its midrange as leaner than the traditional Marantz Class AB sound.

Is that with or with out the $5000 cryogenic interconnects etc, etc. More reviewer BS. What does leaner midrange even mean to them. In the studio it means a dip in the freq. response which only broken electronics have. That Marantz unit isnt broken.

 

[Cbdb2]

we all also want to see the SINAD at different power levels. And at different frequencies.

For electronics those are shown. As well as IMD, low freq. power, complex load response and more.

 

[amirm]

Examples: Amir has selected a handful of measurements to analyze, among them, SINAD at 5 W and 1 kHz. That may be a very useful reference but of course we all also want to see the SINAD at different power levels.

As noted, both of these are in my measurement suite of amplifiers:

THD+N ratio is just SINAD expressed in negative values.

And here is how it varies with frequency:

 

[solderdude]

There are 4 disciplines in audio that all have various degrees (and methods) of measurability.

electronics (all easily measurable with a possible high degree of accuracy, way beyond hearing thresholds)
mechanical components (think, vinyl cartridges, microphones. drivers) which are more difficult to measure.
acoustics. Basically everything that has to do with moving air particles (sound waves) which are measurable but certainly has a lot of challenges.
perception, all things hearing/brain related and thus highly variable between humans ... because of the brain and ears.

All things can probably be measured (correctly) but it seldom is.
Mostly 'easy to measure' things are measured.
Then...
How well are measurements done.
Are relevant measurements done.
How are the results visualized.
How are results interpreted.

So not every relevant aspect is always measured (correctly) and even when it is interpretation of results is not always correct.
That's something we have to live with. One has to know the limitations.

 

[amirm]

- The first hyperbolic statement that microphones are measuring devices and that's the end of the story is not correct. They are "sensing" devices and then an electrical (or computing) process occurs to translate that sensing into a voltage (or a digital version) which can be measured. Human perception is different - and far more - than measuring: it's the result of a psychological construction. That's why, for example, vision is not the same as a photo. One can argue what is better or worse, but they are not the same. But there's no need for that section when you follow it up with the other points.

That's not what he is saying. He is saying the same gear that can be used to measure, was used to record the music. So if that is correct, then it reasons that our measurement system captures whatever there is in the system output.

 

[amirm]

Adding on, what makes a photo a photo is the art. Not the instruments used to make it. Same with music. What makes it music is the art which we don't care to characterize. We are after the fidelity of instruments.

 

[RandomEar]

I appreciate your effort here and agree with the overall premise, but there are a few over-simplifications or errors worth editing, IMO.

- The first hyperbolic statement that microphones are measuring devices and that's the end of the story is not correct. They are "sensing" devices and then an electrical (or computing) process occurs to translate that sensing into a voltage (or a digital version) which can be measured. Human perception is different - and far more - than measuring: it's the result of a psychological construction. That's why, for example, vision is not the same as a photo. One can argue what is better or worse, but they are not the same. But there's no need for that section when you follow it up with the other points.

With all due respect, but this is nitpicking at its finest and missing the point. If you want to call the combo of mic + ADC a "measuring" or a "sensing" device is absolutely up to you but it doesn't affect the argument. The same category of devices (ADC or mic+ADC) that were utilized to capture the music are used to evaluate the playback devices. Human perception is also not relevant to that argument.

- Just because we have measurements doesn't automatically mean they add up to human perception - or even an objective comprehensive description of the the DUT's performance. Examples: Amir has selected a handful of measurements to analyze, among them, SINAD at 5 W and 1 kHz. That may be a very useful reference but of course we all also want to see the SINAD at different power levels. And at different frequencies. And we know human perception is more or less sensitive at different frequencies. So even with Amir's data there is a subjective discussion following most reviews (probably not for the Golfing Panthers!) about how these measurements will be perceived (e.g. the noise v distortion elements of SINAD).

Yes, that's also touched on towards the end of the OP. In general, we get an excellent performance overview from ASR reviews but not so much from other sources like manufacturer specs. It depends.

Net, if you really want to help people learn and be persuaded the post could use a little humility and be less polemical.

Nah, it's fine. Not everything needs to be sugar coated.

 

[Keith_W]

Adding on, what makes a photo a photo is the art. Not the instruments used to make it. Same with music. What makes it music is the art which we don't care to characterize. We are after the fidelity of instruments.

I think it's best if we expand on that photography analogy a bit so that people aren't confused. Photographic instruments ARE VERY MUCH part of the photograph. People choose lenses, set up lighting, process film, etc. to obtain the final photograph. That is the art.

However, reproduction of that photograph is where we want fidelity. You don't want a monitor or a printer that shifts the colour balance or oversaturates the photo so that what you see isn't what the artist intended.

It is the same with music production. The instruments used to produce and record music are very much a part of the finished album. Down to choices of musical instruments and all sorts of DAW plugins and effects generators. But when it comes to music reproduction, we don't want our hifi systems messing up the album.

Messing around with the original is how the Word of God becomes "Blessed are the Cheesemakers" (and also of any dairy product, not just of cheese specifically).

 

[Redacted]

I'm starting to think a lot of people don't go to see live music much, maybe don't have much interest in actual music, and have lost sight of the important stuff, which is not merely listening to your equipment on 200 plays of say, the Best of Abba which is one of the 7 CDs you treasure.

You might be better off collecting big piles of assorted cables, like some people we know....

 

[ahofer]

I go to concerts on average once a week. Mostly classical, occasional jazz club.

Rock concerts too, but they mostly sound like s***.

 

[mpw]

I was a newcomer in 2019 I think. While a seed of doubt had been planted back when Stereo Review did its blind amplifier test in the1980s, I honestly hadn’t been thinking about it much until I had to replace my system (Thiel, Adcom, various inputs) in 2018. But since joining here, I’ve acquired two systems (one for a second home), and bought most of the equipment based on measurements. Whether I’m satisfied is obviously a matter of listening.

Before that in 2018 I bought a pair of Harbeth speakers through pure listening. I still like them, but honestly, it was a pain, and the Revels I bought (used) on measurements (F228be) are indeed superior. I listened to Wilsons, B&W, Paradigm, and ATC (disliked), Legacy, JBL (meh) and KEF (liked the high end models), and some others I don’t even remember (Rado?) at three pompous stores with snake-oil pushing salesmen in the Manhattan area. It’s the worst listening to these guys tell you you have to match your cables and your amp to get an ‘musical’ result and turning the volume up steadily behind your back.

I took notice that the ones I like measure better by Toole standards, even before I knew what a spinorama was. A lot of the names I didn’t like just had showroom treble (ahem Paradigm, B&W) and I made a point of listening for at least half an hour with most everything.

My son helped me blind compare my recapped Adcom space heater with a Hypex p252, and that was interesting as well. Sighted, I thought I could tell the difference. Nope.

The other thing that convinced me, since all of the above is anecdotal, is the abundance of evidence on the internet. There’s no excuse to be ignorant of this stuff anymore if you desire to talk about it rationally.

What Harbeth's did you have or still have that gave you a 'pain, never heard Harbeth's before. I have Spendor's Ls3/5's that I like a lot.

 

[MediumRare]

Adding on, what makes a photo a photo is the art. Not the instruments used to make it. Same with music. What makes it music is the art which we don't care to characterize. We are after the fidelity of instruments.

What I was getting at that the eye does not capture what a camera captures. But then our brain assembles an image. So it is with audio. Our brain filters and modifies what we hear to create our perception of sound. Of course, I know you know. It’s just to say that a generic statement that equates a microphone/recorder to our ears isn’t accurate.

 

[OCA]

What does leaner midrange even mean to them. In the studio it means a dip in the freq. response which only broken electronics have

So frequency response magnitude is the only property that can affect perceived timbre? That's quite a leap

 

[Waveform Fidelity]

Hmm. Do post that filter so we can see what is going on.

Yes please lets see it!
I have flat (within +/-0.5dB) frequency response speakers and speech sounds incredibly realistic…. And I know (like all of us) what speech, unlike music is meant to sound like.