• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

HiFi axioms

jensgk

Active Member
Forum Donor
Joined
Mar 21, 2020
Messages
256
Likes
564
Location
Denmark
  1. Any sound a human can hear, can be measured.
  2. Even if it can be measured, it is not certain that it can be heard by anyone.
  3. Whether a sound can be heard or not, can be determined by ABX tests.
  4. If a human claims to be able to hear a sound, it is possible to either:
    a) devise a measurement that can measure it or prove its existence,
    b) or conclude the claim is not true.
  5. Electronics and acoustics are sciences. Any claims regarding electronics and acoustics can be measured and must follow physics. Claims not explainable by current science, demands evidence. "Extraordinary claims require extraordinary evidence"
  6. Neutral audio reproduction can be measured. Humans have individual tastes, but neutral audio reproduction does not depend on taste.
  7. Audio reproduction is not neutral, if any frequencies are emphasized or deemphasized with regard to the source, or if sounds, that are not part of the source are reproduced. Sounds from the source must reproduced with the correct timing.
  8. Everything in audio can be measured, but some audio perceptions are not yet well-defined, e.g. soundstage.
  9. For a tweak to acknowledged, it needs to be measurable, and to be hearable.
What do you think?
 
Last edited:

Waxx

Major Contributor
Joined
Dec 12, 2021
Messages
1,931
Likes
7,688
Location
Wodecq, Hainaut, Belgium
I largely agree, except that it's required to reproduce audio neutral. That is not a axioma, that is a choice. But if it's coloured it is by definition a degration of the original signal, ment or not.

And i also think not everything is measured yet that is important, especially with speakers. There are factors i know are into play and i'm sure they can be measured, i just did not find out how to determine them and measure them. But it's surely science that can explain it, not magic or golden ears... It's just that i (and maybe even nobody) got the science yet to explain it.

I also think science can explain a lot of personal taste, why some colourations can be wanted by a part of the public and why others not. That is somethnig i miss on this site, the background about that. But the subjective choice of most here is for neutral sound so it's not a subject that is in the scope of their interest.
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
20,753
Likes
20,771
Location
Canada
And i also think not everything is measured yet that is important, especially with speakers. There are factors i know are into play and i'm sure they can be measured, i just did not find out how to determine them and measure them. But it's surely science that can explain it, not magic or golden ears... It's just that i (and maybe even nobody) got the science yet to explain it.
Between @amirm and @hardisj testing and spec'ing the speakers and drivers I think we can get a pretty good analysis of speakers science. If one has the patience and drive to understand all those tests is another thing. :D I'm into the tests and @amirm has tests that interest me. I find @hardisj tests to be so technical and sciency that they exceed my want and need for more speaker information. But he has excellent down to earth presentation skills and is likeable in his videos so that helps a lot. :D
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
20,753
Likes
20,771
Location
Canada
I'll take that. I'd rather have too much information than not enough. :)
I don't know where you are going or ending with this speaker testing that you do but I do know that you have carved out a very specific specialized niche for yourself and I think your efforts and perseverance will result in good things for you in the future. It's very interesting to see the progression of you and amir's development and acquisition of test gear and what you do with it. :D It seems to be every year or so that one of you invests in some serious test gear that opens more doors. That's cool. :D
 
Last edited:

Another Bob

Member
Forum Donor
Joined
Nov 28, 2018
Messages
80
Likes
128
Location
Madison, WI
In the electronics realm, I agree 100% (although I think #9 is essentially redundant with #5 and therefore unnecessary). With speakers, things are less straightforward. It's not that there's any magic or unmeasurable phenomena, but I'm not sure we know what a full set of measurements should consist of and how to prioritize the many inevitable flaws. For example, there does not seem to be good agreement on what an optimum dispersion should be, and it may be that there is no one optimum because it is room dependent. And personally, I think we focus too much on small-signal measurements and have too little understanding of how things change as the volume goes up. Erin's compression tests are, IMO, important and too rare in speaker reviews. None of this contradicts your proposed axioms, in fact I'm basically arguing a point you've already made in #8. It is more a matter of emphasis.
 

Cote Dazur

Addicted to Fun and Learning
Joined
Feb 25, 2022
Messages
619
Likes
758
Location
Canada
Any sound a human can hear, can be measured

Possibly true, but irrelevant as what is measured may or may not be relevant as the microphone does not “hear” the way we do. When we have microphone that work like our ears do, connected to a device that interpret sound the way our brain does, then those measurement will have value. At this point in time those measurements are merely pointers that may lead us in the right direction if correlated with controlled subjectives studies. So better than nothing but far from the whole picture.
To have science that help us we need to measure sound how we hear them not merely just what sound we hear.
The best tool we have still are our ears and brain with some measurement assist.
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,073
Likes
16,609
Location
Central Fl
Everything in audio can be measured, but some audio perceptions are not yet well-defined, e.g. soundstage.
There's no magic in soundstage.
It's simply a factor of level, timing, and phase relationships between the channels.
 

Waxx

Major Contributor
Joined
Dec 12, 2021
Messages
1,931
Likes
7,688
Location
Wodecq, Hainaut, Belgium
Between @amirm and @hardisj testing and spec'ing the speakers and drivers I think we can get a pretty good analysis of speakers science. If one has the patience and drive to understand all those tests is another thing. :D I'm into the tests and @amirm has tests that interest me. I find @hardisj tests to be so technical and sciency that they exceed my want and need for more speaker information. But he has excellent down to earth presentation skills and is likeable in his videos so that helps a lot. :D
Hey, the do a lot more than anybody else, and i'm thankfull. But i know phase of the speaker (and especially if the drivers are in phase) is also very important and that is something they don't measure. I'm sure there is more, but they need to choose and have a streamlined system to do the tests (because it's already a huge work), so i can understand they can't do it all.

And there is something about speakers without crossover that sounds so good, even if the frequency spectrum is not flat, the volume very limited and sometimes the dynamics lack. That is why fullrange drivers are still so popular (also with me). It's not only hype or lazyness, it's just the sound of it also. And yes, i know experienced engineers (Scott Lindgren, Martin J KIng, ....) who designed a lot of multiway speakers who still digg (some of the better) fullrange drivers. I would like to know why that is...

And like that there are still more things to discover i think. But the measurements of Amir & Co surely are a great help and a revolution in the hifi world (at least that the customers get to see those).
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
20,753
Likes
20,771
Location
Canada
Hey, the do a lot more than anybody else, and i'm thankfull. But i know phase of the speaker (and especially if the drivers are in phase) is also very important and that is something they don't measure. I'm sure there is more, but they need to choose and have a streamlined system to do the tests (because it's already a huge work), so i can understand they can't do it all.

And there is something about speakers without crossover that sounds so good, even if the frequency spectrum is not flat, the volume very limited and sometimes the dynamics lack. That is why fullrange drivers are still so popular (also with me). It's not only hype or lazyness, it's just the sound of it also. And yes, i know experienced engineers (Scott Lindgren, Martin J KIng, ....) who designed a lot of multiway speakers who still digg (some of the better) fullrange drivers. I would like to know why that is...

And like that there are still more things to discover i think. But the measurements of Amir & Co surely are a great help and a revolution in the hifi world (at least that the customers get to see those).
We are watching and experiencing a evolution in audio gear. For a private citizen to purchase the type of test gear they are purchasing and releasing to the public the test results is groundbreaking stuff. :D As we see there are more peeps and organizations purchasing the test gear now and that is good for everybody.
 

hardisj

Major Contributor
Reviewer
Joined
Jul 18, 2019
Messages
2,907
Likes
13,908
Location
North Alabama
But i know phase of the speaker (and especially if the drivers are in phase) is also very important and that is something they don't measure.

I've yet to be convinced that phase on its own is important. The phase+magnitude = frequency response. That's what matters to me.

Phase relative to other speakers in a system (i.e., the phase of a bookshelf wrt a subwoofer) matters. But in a complete speaker itself, IMHO, you've got all you need to know in the FR.
 

Waxx

Major Contributor
Joined
Dec 12, 2021
Messages
1,931
Likes
7,688
Location
Wodecq, Hainaut, Belgium
I've yet to be convinced that phase on its own is important. The phase+magnitude = frequency response. That's what matters to me.

Phase relative to other speakers in a system (i.e., the phase of a bookshelf wrt a subwoofer) matters. But in a complete speaker itself, IMHO, you've got all you need to know in the FR.
Wel, i did design some and repaired many loudspeakers, and that is not true in my experience. I made some loudspeakers a lot better just by adjusting the crossover so the drivers are in phase at the crossover point. No esotheric parts or so needed, just right engineering and the difference was very hearabele, but not visible on the FR plot. You could see it on the impendance and step response altough. And i know that even a few big name brands don't look at that with their speakers...
 

hardisj

Major Contributor
Reviewer
Joined
Jul 18, 2019
Messages
2,907
Likes
13,908
Location
North Alabama
I made some loudspeakers a lot better just by adjusting the crossover so the drivers are in phase at the crossover point. No esotheric parts or so needed, just right engineering and the difference was very hearabele, but not visible on the FR plot. You could see it on the impendance and step response altough. And i know that even a few big name brands don't look at that with their speakers...

So you're telling me that you made changes in the crossover that altered the impedance but didn't alter the frequency response?... unless you built an APF, I find this very hard to believe.
 

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,645
Likes
3,624
Location
Sweden, Västerås
We need to add something like .

Electronic small signal components with flat FR response (0)20-20kHz with very low thd and very low noise will all sound the same to humans for example DAC's (does not matter what component it is, could be an preamp or hp amp).
There are a lot of members especially new members that don't get that at all and thinks DAC's sound signatures is a thing in non broken designs
 

DVDdoug

Major Contributor
Joined
May 27, 2021
Messages
2,920
Likes
3,834
Wel, i did design some and repaired many loudspeakers, and that is not true in my experience. I made some loudspeakers a lot better just by adjusting the crossover so the drivers are in phase at the crossover point.
That's relative phase. If the drivers are 180 degrees out-of-phase at the crossover point the sound waves will cancel. You can make smaller phase shifts by moving the woofer & tweeter back-and-forth relative to each other.

It can happen the crossover introduces a 90 lag in the woofer and a 90 degree lead in the tweeter (or midrange in a 3-way). In that case they are 180 degrees out-of-phase and reversing the polarity to the tweeter (or midrange) will correct it.

No esotheric parts or so needed, just right engineering and the difference was very hearabele, but not visible on the FR plot. You could see it on the impendance and step response altough.
Reversing the polarity will show-up in the frequency response, but it won't change the impedance plot.
 

Waxx

Major Contributor
Joined
Dec 12, 2021
Messages
1,931
Likes
7,688
Location
Wodecq, Hainaut, Belgium
So you're telling me that you made changes in the crossover that altered the impedance but didn't alter the frequency response?... unless you built an APF, I find this very hard to believe.
It change the frequency response, but not in a way that it's obvious to see in a frequency response in most cases. But if the phase is not right, the timing of the signal is not right and that causes distortion at the crossover point. But that is also not obvious that the source is that the phase of the drivers are not aligned. Therfor measuring phase of drivers is essential to develop a good crossover i think.
That's relative phase. If the drivers are 180 degrees out-of-phase at the crossover point the sound waves will cancel. You can make smaller phase shifts by moving the woofer & tweeter back-and-forth relative to each other.

It can happen the crossover introduces a 90 lag in the woofer and a 90 degree lead in the tweeter (or midrange in a 3-way). In that case they are 180 degrees out-of-phase and reversing the polarity to the tweeter (or midrange) will correct it.


Reversing the polarity will show-up in the frequency response, but it won't change the impedance plot.
There you make the mistake that you think phase is always the same over the frequency response of a driver, but it's not, it's just like impendance frequency depending and a big part depending on the form of the cone. The acoustic centers of the drivers can be aligned and so, but that does not mean the phase is aligned. Phase can only be measured (most driver companies don't publish it) and is heavy influenced by the crossover. And the best crossovers are those who make the phase of the drive aligned at the wide region of the crossover point. That should be crossover 101 stuff , but is very often forgotten. R.E.W and DATS can measure that very easely and are freeware (REW) or relative cheap (DATS).
 

hardisj

Major Contributor
Reviewer
Joined
Jul 18, 2019
Messages
2,907
Likes
13,908
Location
North Alabama
But that is also not obvious that the source is that the phase of the drivers are not aligned. Therfor measuring phase of drivers is essential to develop a good crossover i think.

If it doesn't show up in the measurements but does show up in the impedance then something is wrong somewhere.

of course measuring phase of the drivers is important. But the phase of the completed speaker (as I said earlier) doesn't tell us anything that isn't in the FR. The FR (phase + magnitude) is the end result.

Phase of a speaker in a system compared to another speaker (i.e., bookshelf + subwoofer) matters. But that's relative phase (as mentioned by another poster above). The absolute phase of just one tells us nothing that isn't already captured in the FR.
 

Sokel

Master Contributor
Joined
Sep 8, 2021
Messages
5,844
Likes
5,787
Audio reproduction is not neutral, if any frequencies are emphasized or deemphasized with regard to the source, or if sounds, that are not part of the source are reproduced. Sounds from the source must reproduced with the correct timing.
So,no audio reproduction can ever be neutral in domestic environment unless we turn it to an acoustically treated dedicated space and make use of DSP correction,isn't it?
No wonder people are fighting like cats about audio then.
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,089
Likes
7,550
Location
Bjerringbro, Denmark.
Whether a sound can be heard or not, can be determined by ABX tests.

I would change that to: Whether a specific sound or distortion type can be heard or not, can be determined to a degree beyond mindless nitpicking by ABX testing with relevant test signals. However, whether those sounds or distortions will be masked by music and become nonrelevant to our hobby, requires ABX testing with actual music.
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,089
Likes
7,550
Location
Bjerringbro, Denmark.
So,no audio reproduction can ever be neutral in domestic environment unless we turn it to an acoustically treated dedicated space and make use of DSP correction,isn't it?

Not even then. No matter how you design an listening space acoustically, it will always be riddled with compromises. You can get somewhat close though.

No wonder people are fighting like cats about audio then.

Indeed :D
 
Top Bottom