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

[mpw]

Better digital pianos actually model this behavior as well. The science and engineering behind piano modeling is pretty interesting, regardless of full digital synthesis or usage of sampling. Via the app I can even change the virtual dimensions of the piano and this will also change the sound. I will however not pretend that any of this comes close to are real piano. But I can have a taste of a real $ 250k Kawai for a fraction of the price, and have room to spare … now if I could only play halfway decent

It is also the pressure of the keys which one can change on a digital piano as well. I am halfway decent as well !

 

[amirm]

An acoustic piano for me is more interesting to play then a digital one as you get the vibrations from the whole piano. You can just press the hold pedal with one foot to keep the chord going then play notes over it.

The example above was without holding the pedal.

 

[Mart68]

Do you know that he, Bruno Putzeys, is the reason we have these competent class D amplifiers in the first place?

His visionary ideas transformed Class D from "good enough for subwoofers" into the end-game technology we have today. The very same ones we measure astounding SNR, THD, and SINAD values on here on ASR. Some readers may not be aware of this.

If you're listening to HiFi class D today, you can thank Bruno Putzeys.

So I say if the guy has an idea about measuring something new to make something even better, give him all the latitude he needs.

I know who he is and of his impressive CV, yes.

No I'm not listening to class D still using a Krell from '94

It's people like me that he wants to sell amplifiers to and indeed he might one day if all the amps I already have fail.

 

[RandomEar]

Do you know that he, Bruno Putzeys, is the reason we have these competent class D amplifiers in the first place?

His visionary ideas transformed Class D from "good enough for subwoofers" into the end-game technology we have today. The very same ones we measure astounding SNR, THD, and SINAD values on here on ASR. Some readers may not be aware of this.

If you're listening to HiFi class D today, you can thank Bruno Putzeys.

So I say if the guy has an idea about measuring something new to make something even better, give him all the latitude he needs.

The man is an excellent engineer, but he's created kind of a conundrum for himself and his company: Their amps are simply too good. There's a couple of small ugly spots left on the graphs, but very, very likely nothing audible anymore. That's a bummer, because Purifi still needs to sell new amplifier modules and I'm sure he would also like to keep his job designing them. It is clearly not in his or Purifi's interest to come out and say: "We've done it, our amps are as good as anybody is ever going to need". On the contrary: He's likely to keep any future claims ambiguous or to advertise further technical improvements (which are possible) and suggest "potential" audible benefits.

In any case, he's free to deliver the evidence that his further optimizations are audible and thereby more than just striving for engineering perfection - which I'm not opposed to. Simply invoking his name won't cut it, though.

 

[ahofer]

Do you know that he, Bruno Putzeys, is the reason we have these competent class D amplifiers in the first place?

His visionary ideas transformed Class D from "good enough for subwoofers" into the end-game technology we have today. The very same ones we measure astounding SNR, THD, and SINAD values on here on ASR. Some readers may not be aware of this.

If you're listening to HiFi class D today, you can thank Bruno Putzeys.

So I say if the guy has an idea about measuring something new to make something even better, give him all the latitude he needs.

LOL. I think it will be a cold day in hell when Bruno Putzeys thinks “ooh. The guys at ASR are skeptical, I better do something else”.

BTW, I’ve seen him post here. Not for a while. @Bruno Putzeys ?

 

[Blueprint]

I too am learning piano, and yes it certainly is a long road. It takes a tremendous amount of practice for things to become second nature, which is what needs to happen to become fluent.

In my experience, some of the plugins you can use on a PC these days can come very close to the sound of a quality acoustic piano. Just a simple MIDI connection is required. However, as noted, the feel of an acoustic piano is like nothing else.

This may not be a bad idea for a thread elsewhere on ASR.

 

[Tell]

The man is an excellent engineer, but he's created kind of a conundrum for himself and his company: Their amps are simply too good. There's a couple of small ugly spots left on the graphs, but very, very likely nothing audible anymore. That's a bummer, because Purifi still needs to sell new amplifier modules and I'm sure he would also like to keep his job designing them. It is clearly not in his or Purifi's interest to come out and say: "We've done it, our amps are as good as anybody is ever going to need". On the contrary: He's likely to keep any future claims ambiguous or to advertise further technical improvements (which are possible) and suggest "potential" audible benefits.

In any case, he's free to deliver the evidence that his further optimizations are audible and thereby more than just striving for engineering perfection - which I'm not opposed to. Simply invoking his name won't cut it, though.

He could just try to just make them cheaper and more widely available. But I guess that's the problem with capitalism, that you always need to improve/make better/increase flashy numbers to somehow and sell new products to people you already sold stuff to keep making money, even if your product already is as good as it gets ;\

 

[MattHooper]

It is a very large instrument with sound emanating across a wide volume/surface. I don't know how anyone is ever able to place microphones as to capture its spatial effects.

I remember stopping dead in my tracks and couldn't focus on my lessen when I hit the first few notes on an acoustic piano vs my digital one! The sensation of perceiving the string vibrations was incredible. It took me many lessons to keep me from being distracted by it! I was ready to ditch my $5000 digital piano for it but just don't have the space.

Here is an incredible recent lesson. I was playing a piece where you hold one cord down for many measures (set of notes) that the left hand if playing. I asked my teacher why does it make sense where the decay has long finished. He answered that it still makes a sound and that it even modifies the sound of the notes in the left hand. We experimented with that and he was right. No way can one feel/perceive what the right cord was producing a foot or two away from left hand.

Note that all of this is from perspective of sitting at the piano or being near it. From audience point of view in a concert, that is a different situation. Still hard but without all of these nuances.

As someone who grew up playing piano (big family at one point we had three pianos in the house), I certainly recognize the experience you are describing.

I used to just sit at a piano and play chords just to feel the vibration, and just listen to the strings ringing on forever. It was a very physical sensation.

That’s one reason why during a huge speaker search some years ago in which I listened to tons of different speakers, when I played some piano track on one loudspeaker I was taken aback. More than any other loudspeaker I had the sensation of actually sitting in front of piano keys striking a large solid piano. It was so distinct it instantly transported to my experiences sitting in front of my own piano playing. It was the only time I experienced that. Relative to the real thing , piano recordings on most loudspeakers sound to me like disembodied piano keys tinkling through the air.

 

[rjp]

The man is an excellent engineer, but he's created kind of a conundrum for himself and his company: Their amps are simply too good. There's a couple of small ugly spots left on the graphs, but very, very likely nothing audible anymore. That's a bummer, because Purifi still needs to sell new amplifier modules and I'm sure he would also like to keep his job designing them.

I think I agree. I really can't imagine how he could further improve his amps. I believe he has moved on to active speakers lately.
DACs are done, amps are done, maybe speakers will be done next? That's a tough one for sure.

Kii Audio | Advanced Active Speakers for Home, Studio & Custom Install

Explore Kii Audio’s advanced active speakers with cardioid room control and DSP for precise, room‑friendly sound at home, in studios and custom installs.

www.kiiaudio.com

 

[DWPress]

It is a very large instrument with sound emanating across a wide volume/surface. I don't know how anyone is ever able to place microphones as to capture its spatial effects.

Which is why I said on page 1: "Live music, in person - always the gold standard though."

 

[goat76]

It is a very large instrument with sound emanating across a wide volume/surface. I don't know how anyone is ever able to place microphones as to capture its spatial effects.

I remember stopping dead in my tracks and couldn't focus on my lessen when I hit the first few notes on an acoustic piano vs my digital one! The sensation of perceiving the string vibrations was incredible. It took me many lessons to keep me from being distracted by it! I was ready to ditch my $5000 digital piano for it but just don't have the space.

Here is an incredible recent lesson. I was playing a piece where you hold one cord down for many measures (set of notes) that the left hand if playing. I asked my teacher why does it make sense where the decay has long finished. He answered that it still makes a sound and that it even modifies the sound of the notes in the left hand. We experimented with that and he was right. No way can one feel/perceive what the right cord was producing a foot or two away from left hand.

Note that all of this is from perspective of sitting at the piano or being near it. From audience point of view in a concert, that is a different situation. Still hard but without all of these nuances.

This reminds me of an interview with Hedd Audio's Klaus Heinz, where he talked about what he thinks is one of the main shortcomings of current loudspeaker technology, and how apparent that becomes in comparison to the very clear tone and dynamic sound coming from his grand piano back home.

Here is the time stamp, 48:06 into the interview, where he talks about that:

 

[RexrothPigeon]

putzeys is pretty damn brilliant

 

[JPA]

My initial point was not so much about stereo recording, but about the apparent contradiction that measurements may not match up with listening experiences and that its likely the wrong measurement was made rather than any golden eared person found something new (e.g diffraction effect when listening off-axis) - so in that case a new measurement would reveal the problem.

My second point about acoustic source size, was more about the fundamental limitations of the perfect “traditional” loudspeaker and an invitations for members to weigh in on the idea of a phased array approach that could go a way to addressing that problem of acoustic size - maybe we should have a thread on phased array loudspeakers? (If we don’t already)

Lastly, I’d very very interested in your thoughts as a piano player and Audiophile about why, in your experience, the sounds from the piano and a recorded version of it are not the same? (We know they are not, but why?) Would the observation differ if dummy head was used for recording and “matching” headphones for replay? I am not assuming you did this test

Bets regards

I'm no expert on the topic, but it seems to me phased arrays would have the same drawbacks as DSP room correction; an improvement in one small sweet spot, but worse sound everywhere else in the room.

 

[RandomEar]

I think I agree. I really can't imagine how he could further improve his amps. I believe he has moved on to active speakers lately.
DACs are done, amps are done, maybe speakers will be done next? That's a tough one for sure.

Kii Audio | Advanced Active Speakers for Home, Studio & Custom Install

Explore Kii Audio’s advanced active speakers with cardioid room control and DSP for precise, room‑friendly sound at home, in studios and custom installs.

www.kiiaudio.com

I think with speakers, it will take a loooong time until we get to a similar level. Maybe it's not even physically possible? Mechanical systems simply require far more compromises in engineering than electronic ones.

 

[mpw]

The example above was without holding the pedal.

Try with holding the pedal to see they different effect. Like with the left hand play a chord like an F then with
right play the nodes on the F scale. If you hold the pedal you will get some interesting harmonics.

 

[MattHooper]

To me that's so totally wrong. Buy on measurements, then train your ears to appreciate what a correct audio signal sounds like. If you don't like it at first, most likely because it's not what you're used to, persevere. Buying what you like just seems so weak.

Although I understand what you’re saying, and I’m sure it can work for some people, that could also sound a bit like putting the cart before the horse: where you are subservient to the gear, having to conform your preferences to the gear, rather than the other way around.

If you’re over the mindset “what I’m concerned about is hearing the most accurate signal” then certainly I think your advice makes sense. By accurate gear, and if you happen to not like it at first for some reason…. Get used to it. (by the way, this is not at all a claim that accurate sound does not sound good.)

On the other hand, if somebody’s of the mindset “ I want to buy gear that I find engaging to listen to - the sound that pleases me most “ then buying what you like makes sense.

Audiophiles have differing goals.

Personally I buy what I like and that works for me. I happen to have liked speakers that “measure well” by ASR standards as well as speakers that measure “ pretty well” but not perfectly, and some other occasional speakers that are poo-poo’d on ASR various members, but which I loved listening to.

As far as getting used to a speaker that at first I didn’t care for, that’s never happened to me once. I’ve had plenty of opportunities to listen to speakers multiple times that didn’t grab me on first listen. I’ve also been able to live with some speakers that I didn’t find immediately engaging. Never once has my opinion or perception of those speakers changed over time from “ didn’t care for it” to “ wow I was wrong now I really like it.”
Maybe others have had that experience, but not me, and therefore I’m not ready to spend significant money on a speaker that I may not care for on the assumption “ I will grow to like it.”

Instead I bought speakers that I fell head over heels for on the very first audition and subsequent auditions, and 10 years later those same speakers continued to thrill me with exactly the same characteristics that sold me on them in the first places

“Different ways to skin a platypus.

I wasn’t exactly a newcomer, but I have yet to be disappointed by a well-measuring component. I am, in fact, delighted with them. I did, however, take some time to learn what measurements to look for.

I think the synthesis here is that one should absolutely use measurements as triage for listening. The components that measure badly, or don’t have measurements, are not worth anyone’s time.

(emphasis mine)
All that makes sense, except up to the very last sentence which I think is a sticking point.
What is the threshold for measuring “badly” (value judgement) and who decides? (especially for other people).

I thoroughly enjoyed some speakers whose measurements are criticized here, but I’m very glad I didn’t pay attention to “ don’t bother with those” because they provided some of my most enjoyable listening experiences.

That even goes for my MBL omnis as well, which some here have decried with nonsense like “ only good for parties” “ not suitable for critical listening” and “wrong, wrong, wrong” in design.

And yet I would’ve missed out on many thrilling listening experiences - to me some of the best sound I’ve ever heard - over the 10 years I owned those speakers.

My CJ tube power amps and preamp probably measure “badly” on the view of most on ASR - my Benchmark preamp or Denon AVR (both of which I use sometimes) measure significantly better. But I would not give up my tube gear for those items. Far from being not worth my time, I’ve been extremely happy with the sound for 25 years!
I’ve owned or used Bryston, Harman Kardon and other solid state amplifiers on and off over the years, and none have provided me the overall pleasure of the CJ gear.

 

[mpw]

The measurements only tell one the values of the sounds coming through the speakers. The human ear hears the quality of the sound which is the music being played. The measurements don't really care if the sound is music or not music. The end result is that human determines if they like the sound being heard. The human ear can even tolerate some distortion or static, etc. if the overall sound is 'music to their ears". Can measurements tell the user that based on the measurements you/listener should enjoy the music? What is good or bad measurements ? Even if 'Ears can't hear as well as measurements the human brain can fill in the 'gaps to make it a whole.

 

[rjp]

I think with speakers, it will take a loooong time until we get to a similar level. Maybe it's not even physically possible? Mechanical systems simply require far more compromises in engineering than electronic ones.

Agreed. It will be a lot harder. I think they will continue to do what they can to make the drivers mechanically better, but ultimately the solution will include some serious DSP, maybe even including the speaker in a feedback loop of sorts to "correct" it's physical imperfections and the loading due to room reverberations. Good times ahead!

 

[MattHooper]

So on the whole “ hearing things we can’t measure” thing…

I frankly hate that fallback claim so often used by the purely subjective crowd. I refer to that kind of stuff as
Golden Ear Aplogetics.

I don’t have to go through the various ways it’s wrong as people are already doing that in this thread.

That said, of course it’s fair to acknowledge that measuring some aspects of our perception is either more tricky than others, or less well researched or established. Toole has said as much a number of times.

And then you can sort of get to exaggerated examples like: play a recording and a listener with the relevant knowledge can detect with their ears; “ that’s Glenn Gould playing piano with Leonard Bernstein conducting”… and identifying most of the instruments playing together, noting some flub from a woodwind, etc.

There’s just tons of sonic information we can pick out by listening that current measurements don’t describe or untangle.

Given all the sonic information contained in any recording, what we actually hear played back through a sound reproduction system, most of what measurements tell us would seem to be generalized - General frequency response, dispersion characteristics, distortion levels added by a component, etc.

Anyway…still not a fan of golden ears falling back on “ you can’t measure what I can hear” because very often the type of things they are claiming to hear can be identified or adjudicated in measurements.

But speaking of a possible example of hearing something that isn’t necessarily in the typical measurements of a loudspeaker….

As I’ve said many times before I’m a big fan of MBL Radialstrahler omnis.

Before I owned one of their smaller stand mounted Omnis, I was obsessed with the larger MBL 101 model most people are familiar with. I’d first been astounded by their open, box-free sound at an audio show, later I got to spend time listening to a pair doing similarly astounding things at a TAS reviewers home. And then I was able to get several auditions of the MBLs at various audio stores.

In every case, except one they blew me away.

But in one store set up they sounded just wrong. There was this strange effect causing some type of tonal/imaging distortion especially in the mid range. I’d never encountered it before in the MBLs.
My brother, who is with me listening noted it as well and we were both put off. Everything sounded pretty good among the frequencies outside of this mid base or midrange area.
But it was still a distortion effect I just didn’t expect or understand.

Later I read Michael Fremer’s review of the MBL 101E speakers where he reported:

"...101Es meant that it would take just the right combination of placement and associated equipment to achieve it—and I'd never before had to deal with an omnidirectional speaker. When everything—or anything—was wrong, there was an odd chesty, compressed, almost grainy midbass coloration, and a sense that female singers centered between the speakers were performing in a closet in the next room. The offending sound was probably due to an unusual combination of placement and reflective interference."

That description is just BANG ON to the artifact I heard with a number of vocal tracks on the MBLs! He was clearly identifying the same odd distortion effect my brother and I heard.

I hadn’t seen anybody predict this occasional but distinct artefact just by looking at the measurements of the speaker.

It seemed left to somebody actually listening to the speaker in certain circumstances to pick up on this.

Is it possible that any set of current room/speaker measurements of those MBL speakers playing in the room I heard this artefact or in Michael’s room could have identified this problem? I suppose maybe somebody could identify “ It looks like you might be having a problem right there in the measurements.”

But would that tell me what that distortion artefact would “ sound like” better than Fremer’s description or hearing it for myself? I’m not sure about that.

 

[ahofer]

The measurements only tell one the values of the sounds coming through the speakers. The human ear hears the quality of the sound which is the music being played. The measurements don't really care if the sound is music or not music. The end result is that human determines if they like the sound being heard. The human ear can even tolerate some distortion or static, etc. if the overall sound is 'music to their ears". Can measurements tell the user that based on the measurements you/listener should enjoy the music? What is good or bad measurements ? Even if 'Ears can't hear as well as measurements the human brain can fill in the 'gaps to make it a whole.

You’re “taking the piss” here, right?