It’s All a Preference

[terryj]

Let's go back to audiophiole 101:

For a discouse on the pros and cons of on and off axis FR ,See the the posts on this forum by Roger Sanders of SandersSoundsytems. I assume Roger has tools to measure frequency response.

Peter Aczel for one postulated that the time domain was possibly the most aspect of speaker design/performance. Hience his love of the DCM Time Window. His refence was the Beveridge 2SW. He was also a big fan of the Quad and wrote a favorable review of the MartinLogan CLS 1.

Still don't get it. There seems to be a denial of the harmon findings. Simply put (hope I have it right enough) 'The majority of listeners prefer smooth Fr on and off axis'.

How can that be denied?? Unless of course they are lying, misrepresenting or deliberately being deceitful.

Given that audio 101 should be about reproducing what was recorded, how then does such a finding fly in the face of that?

Would anyone disagree with the following...'most listeners prefer amplifiers that accurately reproduce the signal in'??

Why does there suddenly become a switch as soon as we hit speakers?? I don't get it. Sure, there is ALWAYS room for personal preferences, from wherever the differences come, but surely shirley we have a minimum starting point??

We are talking hi end systems here are we not?

Whilst I am more than happy to include other factors such as time (as amir well knows haha) I also completely accept that smooth FR is needed, and that that could very well be the single most important.

I don't do A/B testing any more.

Never?? You mean you don't upgrade?? If someone upgrades, do they listen and audition first? If so then how is that not an a/b?

It would then Be presumptious but entirely unlikey that the purchasers of ML did not make a thorough assesment of ML and its' arguuably more "acurate" compettetors.
Obviously they disagreed with AMIR and the other testakers.
Finally AMIR I apologize if I wrongly atrributed any characteristics of your listening preferences. We have never met nor shared a ny listenig sessions.

HOW, exactly, did they disagree with amir and the other testers???? As a laywer, would you not expect a more rigorous argument in court?

You mean they sat there, listened blind and still preferred the ML's??? Or just that they bought them in the first place. If the latter, how can you say they disagreed with amir if they have never done the test?

I have asked this a few times, apart from mep (I think, apologies without going back and checking) no-one has expressed any interest in doing this demo. From your postings I am pretty certain you too have no interest (fine as far as it goes), but why then are you (and others) here arguing the topic??

No interest in testing the hypothesis, yet completely happy to deny it. Seriously you guys, how is it different in tenor than the creationists??

IF you had done the test (or at least 'wished' you could) then ok, we might be able to see that you are happy to discuss...but no interest along with complete denial ('well, because you see')..not much of an argument is it.

It is however an excellent window into the intent and mindset.

I mentioned Dave Moulton (an excellent place for a few nights browsing as mentioned by others too). Here is an interesting thought, sort of. It takes bravery to write that episode.

Bravery you ask?! Yeah, stupid I know, but think about it. It means he has to admit a few things. His ears WERE fooled, he CAN be fooled, and he admitted it.

To me and a few others here, not that brave (silly word when you think about it)...yet it is the very essence of an audiophile. Their ears are paramount, what they hear is gospel, in some cases it is the framework upon which they hang their audiophile identity, it is what gives them any altitude they think they possess in audio land.

Imagine a mag reviewer admitting their ears may not be as good as they make out. Well, that would be the instant end of them eh......(at least they think so)

So yeah, stupid word and all, it takes bravery to write of an experience like moultons.

sad

 

[terryj]

I see theories about speaker performance all the time. Indeed, there are hundreds if not thousands of them. Every manufacture has its thing about speaker design. You can and indeed will go crazy trying to rationalize what they say against any other manufacturer. Here is the one big difference here: Dr. Toole and Crew have literally written the books here. While they are not trivial to read and understand, they require surprisingly little technical knowledge to follow. And it is not just opinions. Everything stated is backed by other research and experimental data. No stone is left unturned.

I have done a test along the lines greg 'suggested' (time, hope I got the right bit to quote!)

It was audible, and preferred, when 'time' was addressed. The things being compared were level matched and very closely matched in FR at the lp. Needed of course to isolate the time as the only change.

Without a doubt however, the overriding perceptual factor WAS FR (which is why it needed to be controlled as well as possible), yet the timing is very much delicious icing on the cake.

 

[Soundminded]

Still don't get it. There seems to be a denial of the harmon findings. Simply put (hope I have it right enough) 'The majority of listeners prefer smooth Fr on and off axis'.

How can that be denied?? Unless of course they are lying, misrepresenting or deliberately being deceitful.

Given that audio 101 should be about reproducing what was recorded, how then does such a finding fly in the face of that?

Would anyone disagree with the following...'most listeners prefer amplifiers that accurately reproduce the signal in'??

Why does there suddenly become a switch as soon as we hit speakers?? I don't get it. Sure, there is ALWAYS room for personal preferences, from wherever the differences come, but surely shirley we have a minimum starting point??

We are talking hi end systems here are we not?

Whilst I am more than happy to include other factors such as time (as amir well knows haha) I also completely accept that smooth FR is needed, and that that could very well be the single most important.




Never?? You mean you don't upgrade?? If someone upgrades, do they listen and audition first? If so then how is that not an a/b?



HOW, exactly, did they disagree with amir and the other testers???? As a laywer, would you not expect a more rigorous argument in court?

You mean they sat there, listened blind and still preferred the ML's??? Or just that they bought them in the first place. If the latter, how can you say they disagreed with amir if they have never done the test?

I have asked this a few times, apart from mep (I think, apologies without going back and checking) no-one has expressed any interest in doing this demo. From your postings I am pretty certain you too have no interest (fine as far as it goes), but why then are you (and others) here arguing the topic??

No interest in testing the hypothesis, yet completely happy to deny it. Seriously you guys, how is it different in tenor than the creationists??

IF you had done the test (or at least 'wished' you could) then ok, we might be able to see that you are happy to discuss...but no interest along with complete denial ('well, because you see')..not much of an argument is it.

It is however an excellent window into the intent and mindset.

I mentioned Dave Moulton (an excellent place for a few nights browsing as mentioned by others too). Here is an interesting thought, sort of. It takes bravery to write that episode.

Bravery you ask?! Yeah, stupid I know, but think about it. It means he has to admit a few things. His ears WERE fooled, he CAN be fooled, and he admitted it.

To me and a few others here, not that brave (silly word when you think about it)...yet it is the very essence of an audiophile. Their ears are paramount, what they hear is gospel, in some cases it is the framework upon which they hang their audiophile identity, it is what gives them any altitude they think they possess in audio land.

Imagine a mag reviewer admitting their ears may not be as good as they make out. Well, that would be the instant end of them eh......(at least they think so)

So yeah, stupid word and all, it takes bravery to write of an experience like moultons.

sad

"Would anyone disagree with the following...'most listeners prefer amplifiers that accurately reproduce the signal in'??"

Many audiophiles prefer vacuum tube amplifiers whose measurements are atrocious. Why? Perhaps their distortions mitigate distortions eleswhere in their sound system. A rolled off high end amplifier compensating for a shrill loudspeaker or poor damping factor amplifier mitigating a speaker with rolled off bass response. We don't listen to loudspeakers or amplifiers, we listen to sound fields resulting from the entire chain starting with the microphones and ending with what the room does to the sound that finally reaches the listener.

"The majority of listeners prefer smooth Fr on and off axis'."

Then the best speakers would be classic Acoustic Research and Allison. They had among the widest dispersion of all speakers. All that was necessary was to equalize them so their FR was flatter. Many manufacturers today try for limited dispersion with sharp falloff of off axis response especially at high frequencies. This is claimed to improve imaging, the holy grail of many audiophiles.

 

[terryj]

"Would anyone disagree with the following...'most listeners prefer amplifiers that accurately reproduce the signal in'??"

Many audiophiles prefer vacuum tube amplifiers whose measurements are atrocious. Why? Perhaps their distortions mitigate distortions eleswhere in their sound system. A rolled off high end amplifier compensating for a shrill loudspeaker or poor damping factor amplifier mitigating a speaker with rolled off bass response. We don't listen to loudspeakers or amplifiers, we listen to sound fields resulting from the entire chain starting with the microphones and ending with what the room does to the sound that finally reaches the listener.

All the above is true, they happen. Let's leave aside the obvious points like 'if you want a tone control buy a tone control' and move on to the important bits.

It could be educational too! Ok, what is the worst measuring amplifier we know of??? I recall a lot of derision about a wavac valve amp, but what is the worst measuring amp ever?? Dunno.

Anyways, I'd wager the worst ever amp far exceeds the best ever measuring speaker by many orders of magnitude. The question is, why the sudden gestalt switch when it comes to speakers??

Would anyone seriously consider an amp that measures like the BEST measuring speaker.....

"The majority of listeners prefer smooth Fr on and off axis'."

Then the best speakers would be classic Acoustic Research and Allison. They had among the widest dispersion of all speakers. All that was necessary was to equalize them so their FR was flatter. Many manufacturers today try for limited dispersion with sharp falloff of off axis response especially at high frequencies. This is claimed to improve imaging, the holy grail of many audiophiles.

I'll take your word for it, but point out the harmon criterion (as said by me remember) is not 'wide dispersion' per se.

On the last point, what manufacturers do you know of that aim for limited dispersion? (talking 'name brands' as it were)

I ask cause, afaik, limited dispersion/cd designs etc etc seem a relatively recent phenomenon, often (mainly) discussed on diy forums. Has it transferred to commercial in any significant way?

I have never heard one myself, on the 'curious about' list.

 

[amirm]

Perhaps you should blind yourself to Harmons resources. It seems to influence you greatly. Being flat in anechoic chamber does not replicate real world conditions. Many manufacturers show themselves sittiing in anechoic chamber. I assume they make good use of it.

Even though I have explained this three or four times it seems that the concept is still completely missed given your statement above. So once more, in more detail:

The standard anechoic chamber measurement you talk about is on-axis. This indeed is pretty far away from what we hear in our homes. At low frequencies, the room dominates and heavily modifies the response of the speaker. At higher frequencies the speaker is in control which means that we can predict the performance of the speaker in room *if* we have the right measurements. Here is the one picture you want to remember from my article on low frequency room optimization:

For the purposes of this discussion we are talking about what is above the transition frequency. There, what you hear hear is the sum total of direct sound and reflected. Since the power of the reflected sounds is nearly as much as the direct sound, it has a huge impact on the perceived timber of the speaker sound you hear.

Think of your speaker as not one, but an array. The one you think you own is sitting where it is. But from every reflection point on the wall, there is another speaker that is also playing. Except that the sound from those speakers differs from the sound from the main speaker. If I mixed these you would intuitively arrive at the right conclusion that the sum total of all of these speakers matters and not just the one producing the direct sound.

To measure the sound of those other "speakers," Harman utilizes a device called a " "Spin-O-Rama." This is a turntable which the speaker sits on and under computer control rotates "N" degrees. In front of the speaker there is an array of microphones in an arc (in an anechoic chamber). The table moves from notch to notch and measurements of the array are captured. Once done, there are 70 points worth of data in a sphere around the speaker. Here is a way to visualize it:

Each measurement point tells us the quality of the sound escaping the speaker in that direction. Harman research into this using listening tests has shown which contributions matter and which do not. And to what degree. Here is Kevin Voecks on the topic: http://www.soundstagehifi.com/index...yre-designed&catid=62:monthly-column&Itemid=3

'Each transducer is measured in the enclosure using a process we call "Spin-O-Rama," a series of 70 measurements made in a large 4? ("full-space") anechoic chamber. These measurements result in data forming a complete sphere around the speaker. This is done because our research has resulted in a group of calculated responses that have an excellent correlation to sound quality.

This high-quality data is then used in filter-synthesis software to design the system crossover networks. (While nobody really thinks about the crossovers when they look at a speaker, the crossovers are, in many ways, the heart of the system.) We can then evaluate the "Spin" data, which includes curves of the listening window, early reflections, sound power, and the directivity index. This group of measurements has a very close correlation to timbre, which is the most audible aspect of loudspeaker sound quality. If the "Spin" data is poor, the speaker will certainly sound bad, so any issues visible in the calculated curves would be investigated and solved before any listening tests.

Once the full "Spin" data results are optimized, position-independent listening tests are performed. Since small differences in the placement of a loudspeaker within any listening room result in significant audible differences at the listener’s ears, each loudspeaker must be heard when placed at the same location. We use a pneumatic "shuffler" to move each loudspeaker to precisely the same location when it is being played -- all within a few seconds, in order to maximize the sensitivity and validity of the test. We compare the speakers with the best "anchors," as well as speakers in their price range. The anchors are the best direct-radiator loudspeakers known to us -- the Salon2s, for example -- which serve as an acoustic "target." They also prevent inadvertent effects if all the competitors’ speakers happen to have a similar, non-ideal timbre, such as being excessively bright.

In general, we can outperform competitors based solely on the measurements made up to this point. However, this is where the real fun begins, where we can do blind-listening tests and aim to fine-tune the loudspeaker to perform as close as possible to the reference anchor loudspeaker. At this point, we compare the prototype to more expensive competitors to get a sense of where our performance lands relative to the (often much more expensive) competitive loudspeakers."

Saying linear response is important is iakin to saying the "son is hot."

Funny thing is, folks design speakers where they assume sun is not hot! Yes, smooth frequency response is the most fundamental truth we hold in audio. But when it comes to speakers it seems that people have resigned themselves to the fact that speakers are going to have non-ruler-flat response and as a result, put way too little emphasis on it.

Now here is the key thing: getting on-axis frequency response flat is not too hard. Getting it to be flat off-axis, is difficult! The reason is that as the frequencies go up relative to the size of a driver, it becomes more directional. This means that the driver sends less of its sound to the sides than to the front. Ideally by the time this happens, the next smaller driver picks up the task and we get wide response again. This simple rule is violated commonly. Take a bookshelf speaker with an 8 inch woofer and 1 inch tweeter. The 8 inch woofer will have to work well into Kilohertz region where it becomes very directional. Yet the tweeter is not yet able to pick up that load. Result is that in those mid-frequencies -- where the ear is most sensitive, the off-axis response drops off in power. Put that speaker in a room and mix that off-axis bouncing off the wall with your direct sound and you now have modified the timber of the speaker. Humans are awfully good at hearing timbre variations. So no wonder that speakers like B&W 802 lost out in the Harman test.

Read Rogers' white papers where he reounts many factors eg. on axis reponse(narrow sweet spot),distrotion, transient repsonse, high spl witouht brekup. Admittedly elctrostantics varying impedance acroos tthe audio spectrum causes problems. It does represent a trade off.

No, there is no trade off in this specif area. Screw up the speaker tonally and you can do all you want and it won't be as good of a speaker. The results of listening tests prove that.

Hence my refence to time domain vs frequency domain.

A reference is not an argument . Remember, if you modify frequency response, you change time and vice versa. These are not independent events.

Ratinalizing theoriies may be agonizing. I find evaluating thier implementation not difficult at all.

The problem is, their implementations lose out in listening tests. And have poor measurements to boot. So what is left is to hang one's hat on a completely ad-hoc test: a speaker in some showroom or your own home. On what basis would your conclusions be translatable to someone else?

Is not ringing a time doiamin error?

No, it is in both. Look at what happens when you EQ down a peak in frequency response:

What happened to the ringing? It got reduced. It has to. It has no choice or Fourier would be wrong.

I find it highly objectionable?Don't you? BTW just how would I go about proving that to you? (time domain v Frequency response)

You have to get technical and demonstrate the point and not just feed me buzzwords . Put forward a theory that if a speaker has bad frequency response, I don't notice it because it has time domain perfectly nailed. Show me the math. Show me listening tests.

Right or wrong eerts usually prevail over layman.

They do if you close your eyes . And hear two things compared. .

Realy I don't mean to make this personall, but it was you who said blind testing is only necessary for detecting small differences.... I remember becaus4 it came as quite a shock.

Give me quote. For now, I am surprised you would care about someone's observation about blind tests one way or the other .

First I guess you missed the part where I said I don't do A/B testing. While my mmeory is less than perfect my idea of what a piano or drum sounds like is sufficient for audition purposes. I might be an exception Sonuc memery is real. Musicians tune their instruments by ear. You may recall my reference is live music not the source.

And I addressed that. You are not situated as those musicians. They always hear their live events. You do not. You are sitting at home with a CD or LP at hand with zero clues as to how it sounded in the studio when it was mixed and mastered. That is the nature of the audio business and one of the major reasons so many disagreements exists. With no litmus test of what is live, then it is all hypothesis.

Good news is that as humans, we can tell good sound from bad. If an amp distorts, we can tell it is doing that (at least in gross amounts) without any idea of what we are playing and its relevance to live event. So yes, we do have a reference in our mind but it is not the live presentation as you say.

As to not doing AB tests, I don't buy that. I hear you saying it but I have never seen an audio person not do AB tests. You take out a cable, put in a new and say, "ah, that sounds better." Well, that is an AB test.

Agreed. It's not about ML I;m 'concerend about.Iit's waht appears to be an epihany that concerns me.

I thought I had heard it all . You are worried about me having an epiphany? Why? If I didn't have one, I would come across more convincing to you? You rather this stuff didn't make sense?

EQ only cures one problem.

Was not talking about cure. Was talking about how if I take your favorite speaker, right now, and apply an EQ to it and bring down the mid-tones by 3 db, you will hear it. Yet you deny such variations as mattering in design of a speaker. And instead like to chance fancy words like ringing and time domain that you can't articulate to me in a scientific way what the impact would be.

No problem. We are however the sum of expereince. Conflict of interest is real even if you think you are immune. I seem to remeber you asking me how I proported to know things about you when I had not made an inquiry of you. I tried to explain. It's you who have stated your chose products based on science. All you have to is clink on the link themn click on products.It's you who have endorsed SEan Olives methods. I merely made some inferences based on that knowledge.

Appearance of bias <> bias. Not in real life, and not in discussion of audio. Lest you subscribe to the school of if I look guilty, I must have committed the crime. Your point was not this anyway. You talked about my company's web site not having audio cables on it. That has nothing to do with correctness of Harman's tests.

As I have noted, Dr. Toole's work here goes years back to research at NRC when he was not even associated with Harman.

 

[terryj]

(Greg quote Is not ringing a time doiamin error?)


No, it is in both. Look at what happens when you EQ down a peak in frequency response:

What happened to the ringing? It got reduced. It has to. It has no choice or Fourier would be wrong.

To be fair to greg I thought he was referring to ringing in the speaker itself, rather than the room which you specifically referred to.

Anyway, it did remind me to point out that in the FR ringing CAN be seen (even if he is talking about a speaker ringing at 2k say). Just because it is a FREQUENCY response does not mean many other factors do not show up. Ringing may be seen better on a CSD say, but the frequency anomoly will be seen in the straight no frills FR anyway.

Put another way perhaps, any time you see 'sudden' peaks and dips in a FR you know there are accompanied phase and time issues associated with that.

I wonder if a lot of the problem is simply no experience at all with correlating what can be seen on a FR and how it manifests to the ears??

Take the 'challenge' of 'Was not talking about cure. Was talking about how if I take your favorite speaker, right now, and apply an EQ to it and bring down the mid-tones by 3 db, you will hear it. Yet you deny such variations as mattering in design of a speaker. And instead like to chance fancy words like ringing and time domain that you can't articulate to me in a scientific way what the impact would be. '

Heck, make it 0.2 db (over a large range) and I bet you'd hear it. Unless you have had the opportunity to play with this stuff in your own chair then it remains just a 'theory' or some such???

 

[Soundminded]

Even though I have explained this three or four times it seems that the concept is still completely missed given your statement above. So once more, in more detail:

The standard anechoic chamber measurement you talk about is on-axis. This indeed is pretty far away from what we hear in our homes. At low frequencies, the room dominates and heavily modifies the response of the speaker. At higher frequencies the speaker is in control which means that we can predict the performance of the speaker in room *if* we have the right measurements. Here is the one picture you want to remember from my article on low frequency room optimization:

For the purposes of this discussion we are talking about what is above the transition frequency. There, what you hear hear is the sum total of direct sound and reflected. Since the power of the reflected sounds is nearly as much as the direct sound, it has a huge impact on the perceived timber of the speaker sound you hear.

Think of your speaker as not one, but an array. The one you think you own is sitting where it is. But from every reflection point on the wall, there is another speaker that is also playing. Except that the sound from those speakers differs from the sound from the main speaker. If I mixed these you would intuitively arrive at the right conclusion that the sum total of all of these speakers matters and not just the one producing the direct sound.

To measure the sound of those other "speakers," Harman utilizes a device called a " "Spin-O-Rama." This is a turntable which the speaker sits on and under computer control rotates "N" degrees. In front of the speaker there is an array of microphones in an arc (in an anechoic chamber). The table moves from notch to notch and measurements of the array are captured. Once done, there are 70 points worth of data in a sphere around the speaker. Here is a way to visualize it:

Each measurement point tells us the quality of the sound escaping the speaker in that direction. Harman research into this using listening tests has shown which contributions matter and which do not. And to what degree. Here is Kevin Voecks on the topic: http://www.soundstagehifi.com/index...yre-designed&catid=62:monthly-column&Itemid=3

'Each transducer is measured in the enclosure using a process we call "Spin-O-Rama," a series of 70 measurements made in a large 4? ("full-space") anechoic chamber. These measurements result in data forming a complete sphere around the speaker. This is done because our research has resulted in a group of calculated responses that have an excellent correlation to sound quality.

This high-quality data is then used in filter-synthesis software to design the system crossover networks. (While nobody really thinks about the crossovers when they look at a speaker, the crossovers are, in many ways, the heart of the system.) We can then evaluate the "Spin" data, which includes curves of the listening window, early reflections, sound power, and the directivity index. This group of measurements has a very close correlation to timbre, which is the most audible aspect of loudspeaker sound quality. If the "Spin" data is poor, the speaker will certainly sound bad, so any issues visible in the calculated curves would be investigated and solved before any listening tests.

Once the full "Spin" data results are optimized, position-independent listening tests are performed. Since small differences in the placement of a loudspeaker within any listening room result in significant audible differences at the listener’s ears, each loudspeaker must be heard when placed at the same location. We use a pneumatic "shuffler" to move each loudspeaker to precisely the same location when it is being played -- all within a few seconds, in order to maximize the sensitivity and validity of the test. We compare the speakers with the best "anchors," as well as speakers in their price range. The anchors are the best direct-radiator loudspeakers known to us -- the Salon2s, for example -- which serve as an acoustic "target." They also prevent inadvertent effects if all the competitors’ speakers happen to have a similar, non-ideal timbre, such as being excessively bright.

In general, we can outperform competitors based solely on the measurements made up to this point. However, this is where the real fun begins, where we can do blind-listening tests and aim to fine-tune the loudspeaker to perform as close as possible to the reference anchor loudspeaker. At this point, we compare the prototype to more expensive competitors to get a sense of where our performance lands relative to the (often much more expensive) competitive loudspeakers."


Funny thing is, folks design speakers where they assume sun is not hot! Yes, smooth frequency response is the most fundamental truth we hold in audio. But when it comes to speakers it seems that people have resigned themselves to the fact that speakers are going to have non-ruler-flat response and as a result, put way too little emphasis on it.

Now here is the key thing: getting on-axis frequency response flat is not too hard. Getting it to be flat off-axis, is difficult! The reason is that as the frequencies go up relative to the size of a driver, it becomes more directional. This means that the driver sends less of its sound to the sides than to the front. Ideally by the time this happens, the next smaller driver picks up the task and we get wide response again. This simple rule is violated commonly. Take a bookshelf speaker with an 8 inch woofer and 1 inch tweeter. The 8 inch woofer will have to work well into Kilohertz region where it becomes very directional. Yet the tweeter is not yet able to pick up that load. Result is that in those mid-frequencies -- where the ear is most sensitive, the off-axis response drops off in power. Put that speaker in a room and mix that off-axis bouncing off the wall with your direct sound and you now have modified the timber of the speaker. Humans are awfully good at hearing timbre variations. So no wonder that speakers like B&W 802 lost out in the Harman test.


No, there is no trade off in this specif area. Screw up the speaker tonally and you can do all you want and it won't be as good of a speaker. The results of listening tests prove that.


A reference is not an argument . Remember, if you modify frequency response, you change time and vice versa. These are not independent events.


The problem is, their implementations lose out in listening tests. And have poor measurements to boot. So what is left is to hang one's hat on a completely ad-hoc test: a speaker in some showroom or your own home. On what basis would your conclusions be translatable to someone else?


No, it is in both. Look at what happens when you EQ down a peak in frequency response:

What happened to the ringing? It got reduced. It has to. It has no choice or Fourier would be wrong.


You have to get technical and demonstrate the point and not just feed me buzzwords . Put forward a theory that if a speaker has bad frequency response, I don't notice it because it has time domain perfectly nailed. Show me the math. Show me listening tests.


They do if you close your eyes . And hear two things compared. .


Give me quote. For now, I am surprised you would care about someone's observation about blind tests one way or the other .


And I addressed that. You are not situated as those musicians. They always hear their live events. You do not. You are sitting at home with a CD or LP at hand with zero clues as to how it sounded in the studio when it was mixed and mastered. That is the nature of the audio business and one of the major reasons so many disagreements exists. With no litmus test of what is live, then it is all hypothesis.

Good news is that as humans, we can tell good sound from bad. If an amp distorts, we can tell it is doing that (at least in gross amounts) without any idea of what we are playing and its relevance to live event. So yes, we do have a reference in our mind but it is not the live presentation as you say.

As to not doing AB tests, I don't buy that. I hear you saying it but I have never seen an audio person not do AB tests. You take out a cable, put in a new and say, "ah, that sounds better." Well, that is an AB test.


I thought I had heard it all . You are worried about me having an epiphany? Why? If I didn't have one, I would come across more convincing to you? You rather this stuff didn't make sense?


Was not talking about cure. Was talking about how if I take your favorite speaker, right now, and apply an EQ to it and bring down the mid-tones by 3 db, you will hear it. Yet you deny such variations as mattering in design of a speaker. And instead like to chance fancy words like ringing and time domain that you can't articulate to me in a scientific way what the impact would be.


Appearance of bias <> bias. Not in real life, and not in discussion of audio. Lest you subscribe to the school of if I look guilty, I must have committed the crime. Your point was not this anyway. You talked about my company's web site not having audio cables on it. That has nothing to do with correctness of Harman's tests.

As I have noted, Dr. Toole's work here goes years back to research at NRC when he was not even associated with Harman.

Given the important effect the room has on how a speaker will sound, what provisions has a Harman company or anyone else engineered into their product to compensate or even mitigate the effect of that variable, that is to make provisions so that their speakers sound the same or similar in different rooms with different acoustics? None that I can think of. Tone controls, equalizers, and level controls by themselves don't work because they change both the direct and reverberant sound field simultaneously. I've designed my own speakers otherwise. They are configured to alter the two fields independently of each other. One effect pulling speakers away from the walls of a room has is to reduce to a degree the effect of the room, so does building a LEDE room. So does trying to make the speakers as directional as possible. This last characteristic means you have to position and aim your speakers with great precision and you must sit in one spot with your head in a vice or you will not hear the intended effect. Is that what accuracy means? From what I've gleaned it does to a lot of audiophiles (I do not consider myself an audiophile in the commonly used sense of the word.)

Musicians are notoriously bad at judging the quality of electronic music reproducing systems for accuracy. Ask one how can you listen to recordings on that awful $79.95 player and they will tell you frequently they hear past the technology through to the music. That's one reason, another is that where they perform, what they hear is entirely different from what the audience hears. To be a good candidate to be on an evaluation panel, I'd think listeners should not only demonstrate their hearing accuity (what was Fremer's and Tellig's most recent hearing test results and how recently was it performed) but they should be experienced concert goers as well. Since most people's preponderance of listening experience relates to recorded music, what qualifies them to decide on what is accurate and what isn't? That ability relates to memory reinforced by frequently repeated exposure to the real thing. So along with their hearing tests, what is the frequency of their attendance at live performances where only acoustic instruments and no electronic sound reinforcement system gave them a reference that's reliable enough to take their opinions credulously?

 

[Phelonious Ponk]

Given the important effect the room has on how a speaker will sound, what provisions has a Harman company or anyone else engineered into their product to compensate or even mitigate the effect of that variable, that is to make provisions so that their speakers sound the same or similar in different rooms with different acoustics?

I think this is exactly what Harman is doing. Even more importantly, I think it may be the only practical, effective thing anybody can do. You're right; EQ boogers (the technical term) the direct sound as it compensates the reflected sound. And even the most affluent OCD audiophiles are not building anechoic listening rooms, because...well, among other things they don't sound good. So there are always going to be reflections. And if you want a pair of speakers to sound their best without knowing the room they're going in or what kind of reflections you'll be dealing with, what's the only practical think you can do?

Make sure the reflected sound comes from a source that is accurate, or at least a good match to the direct sound; build speakers with smooth, even response on and off axis, so when the first reflections come back and mix with the direct sound in the listeners ears, you have the delay, but not the delay of a second speaker system that sounds unlike the first.

Musicians are notoriously bad at judging the quality of electronic music reproducing systems for accuracy. Ask one how can you listen to recordings on that awful $79.95 player and they will tell you frequently they hear past the technology through to the music.

Given the gap between live music and reproduction - even on the best systems - perhaps we all should be trained to do this instead of being trained to hear the smallest distortions and artifacts. Sometimes the audiophile endeavor strikes me as self punishment.

To be a good candidate to be on an evaluation panel, I'd think listeners should not only demonstrate their hearing accuity (what was Fremer's and Tellig's most recent hearing test results and how recently was it performed) but they should be experienced concert goers as well.

Perhaps the biggest audiophile can of worms of all: The live concert as the reference. What kind of live concert? Which hall? What seat? And you're comparing it to what? A live recording of that same performance made from that seat with a pair of stereo mics? OK, so your make your own live recordings. If you know the sound of your mics and recorder really well and you have super-human auditory memory, live music can be your reference for an acoustic concert. Listen to jazz, rock, pop, country? All bets are off.

Since most people's preponderance of listening experience relates to recorded music, what qualifies them to decide on what is accurate and what isn't?

IMO, people are NOT qualified to make this decision, because there is no "accurate" when it comes to live acoustic performance. It is dependent upon the room, how many people are in it and the seating position at least. The live performance, as reference, is a fantasy. The only thing the playback system knows is the recording. That is the only reference.

That ability relates to memory reinforced by frequently repeated exposure to the real thing. So along with their hearing tests, what is the frequency of their attendance at live performances where only acoustic instruments and no electronic sound reinforcement system gave them a reference that's reliable enough to take their opinions credulously?

You need to go back to the musicians. Find ones who are also recordists. They hear the real thing, in the actual performance space, then they hear it played back through a reproduction system. Even the engineer is not in the room with the player; he is in the control room listening through microphones, system and monitors. But even the musician will not have the true reference you are looking for because the sound of an instrument from the perspective of the player is very different than the sound of the same instrument in front of the player.

The live "reference" is a fantasy. It only really exists for a guy like Barry Diamet, recording music in the performance space with two mics and electronics he is intimately familiar with. If anyone here has joined Barry for one of his sessions and stood where the mics are placed, he has, within the limits of his auditory memory, a live "reference" for one recording. But even then he has a live reference to an unrealistic performance. To achieve balance in those mics, Barry sets his musicians up placed in a line front to back, with the mic the furthest thing forward. The louder the instrument, the further back it goes. I completely understand why he does this, but no one sets up and plays a performance that way. I think what he's doing is cool and unique and subtle. But it is still a manipulation of the performance to achieve a result in recording that is very different from what you would hear, sitting in the audience, listening to an actual performance.

Recordings are a false construct. And accuracy means fidelity to the recording, as it always has, until audiophiles went looking for a reason to believe their favorite colorations were somehow more accurate than accurate.

Tim

 

[MylesBAstor]

Given the important effect the room has on how a speaker will sound, what provisions has a Harman company or anyone else engineered into their product to compensate or even mitigate the effect of that variable, that is to make provisions so that their speakers sound the same or similar in different rooms with different acoustics? None that I can think of. Tone controls, equalizers, and level controls by themselves don't work because they change both the direct and reverberant sound field simultaneously. I've designed my own speakers otherwise. They are configured to alter the two fields independently of each other. One effect pulling speakers away from the walls of a room has is to reduce to a degree the effect of the room, so does building a LEDE room. So does trying to make the speakers as directional as possible. This last characteristic means you have to position and aim your speakers with great precision and you must sit in one spot with your head in a vice or you will not hear the intended effect. Is that what accuracy means? From what I've gleaned it does to a lot of audiophiles (I do not consider myself an audiophile in the commonly used sense of the word.)

Musicians are notoriously bad at judging the quality of electronic music reproducing systems for accuracy. Ask one how can you listen to recordings on that awful $79.95 player and they will tell you frequently they hear past the technology through to the music. That's one reason, another is that where they perform, what they hear is entirely different from what the audience hears. To be a good candidate to be on an evaluation panel, I'd think listeners should not only demonstrate their hearing accuity (what was Fremer's and Tellig's most recent hearing test results and how recently was it performed) but they should be experienced concert goers as well. Since most people's preponderance of listening experience relates to recorded music, what qualifies them to decide on what is accurate and what isn't? That ability relates to memory reinforced by frequently repeated exposure to the real thing. So along with their hearing tests, what is the frequency of their attendance at live performances where only acoustic instruments and no electronic sound reinforcement system gave them a reference that's reliable enough to take their opinions credulously?

Have you ever heard a LEDE room? I have and they were disasters

 

[Soundminded]

I think this is exactly what Harman is doing. Even more importantly, I think it may be the only practical, effective thing anybody can do. You're right; EQ boogers (the technical term) the direct sound as it compensates the reflected sound. And even the most affluent OCD audiophiles are not building anechoic listening rooms, because...well, among other things they don't sound good. So there are always going to be reflections. And if you want a pair of speakers to sound their best without knowing the room they're going in or what kind of reflections you'll be dealing with, what's the only practical think you can do?

Make sure the reflected sound comes from a source that is accurate, or at least a good match to the direct sound; build speakers with smooth, even response on and off axis, so when the first reflections come back and mix with the direct sound in the listeners ears, you have the delay, but not the delay of a second speaker system that sounds unlike the first.



Given the gap between live music and reproduction - even on the best systems - perhaps we all should be trained to do this instead of being trained to hear the smallest distortions and artifacts. Sometimes the audiophile endeavor strikes me as self punishment.



Perhaps the biggest audiophile can of worms of all: The live concert as the reference. What kind of live concert? Which hall? What seat? And you're comparing it to what? A live recording of that same performance made from that seat with a pair of stereo mics? OK, so your make your own live recordings. If you know the sound of your mics and recorder really well and you have super-human auditory memory, live music can be your reference for an acoustic concert. Listen to jazz, rock, pop, country? All bets are off.



IMO, people are NOT qualified to make this decision, because there is no "accurate" when it comes to live acoustic performance. It is dependent upon the room, how many people are in it and the seating position at least. The live performance, as reference, is a fantasy. The only thing the playback system knows is the recording. That is the only reference.



You need to go back to the musicians. Find ones who are also recordists. They hear the real thing, in the actual performance space, then they hear it played back through a reproduction system. Even the engineer is not in the room with the player; he is in the control room listening through microphones, system and monitors. But even the musician will not have the true reference you are looking for because the sound of an instrument from the perspective of the player is very different than the sound of the same instrument in front of the player.

The live "reference" is a fantasy. It only really exists for a guy like Barry Diamet, recording music in the performance space with two mics and electronics he is intimately familiar with. If anyone here has joined Barry for one of his sessions and stood where the mics are placed, he has, within the limits of his auditory memory, a live "reference" for one recording. But even then he has a live reference to an unrealistic performance. To achieve balance in those mics, Barry sets his musicians up placed in a line front to back, with the mic the furthest thing forward. The louder the instrument, the further back it goes. I completely understand why he does this, but no one sets up and plays a performance that way. I think what he's doing is cool and unique and subtle. But it is still a manipulation of the performance to achieve a result in recording that is very different from what you would hear, sitting in the audience, listening to an actual performance.

Recordings are a false construct. And accuracy means fidelity to the recording, as it always has, until audiophiles went looking for a reason to believe their favorite colorations were somehow more accurate than accurate.

Tim

While what you say is true about every seat even in one performance venue sounding different from every other seat, there are certain common characteristics of the acoustic effects large venues create that cannot be captured or reproduced by our current technology no matter how expensive or elaborate. This not only includes the auditory sense of space but tonality of musical instruments as they are heard in those venues as well. In other words, if you are trying to reproduce the tone of musical instruments as heard at a concert hall with a sound system, (binaural recordings made with a dummy head at a seat in the audience played through headphones excepted) no matter what frequency response your system has, it's wrong. And that doesn't even take into account that spectral balance varies from one recording to the next.

The best that can be done with a 2 channel sound system IMO even after correcting for spectral reflection distortions and variables in the recording is to reproduce the tone of the instruments as they would be heard were they in your listening room. That done, that does not mean they will be spatially accurate. I can get my main speakers in my best system to match the timbre of the Steinway piano in the same room but there's still a big difference. The piano being largely an indirect radiator fills up half the room with sound as an enormous source, the speakers are a pair of point sources. You can tell the difference immediately. The speaker/overall system design in this regard remains still flawed.

 

[RUR]

..... EQ boogers (the technical term) the direct sound as it compensates the reflected sound...

Not in today's SOTA products. Lyngdorf, Dirac and Trinnov all use proprietary technology to differentiate, measure, and appropriately compensate for direct vs. room reflections. See for example Jan Abildgaard Pedersen, "“Loudspeaker-Room Adaptation for a specific Listening Position using Information about the Complete Sound Field”". That's 2006 and the technology has progressed significantly since then.

 

[Soundminded]

Not in today's SOTA products. Lyngdorf, Dirac and Trinnov all use proprietary technology to differentiate, measure, and appropriately compensate for direct vs. room reflections. See for example Jan Abildgaard Pedersen, "“Loudspeaker-Room Adaptation for a specific Listening Position using Information about the Complete Sound Field”". That's 2006 and the technology has progressed significantly since then.

http://www.automatedhk.com/MaterialAttachPart/93/proofread_papper_aes32.pdf

Looks like the same flawed concept to me. The system shown in figure 1 cannot differentiate between equalizing for the direct and reflected field.

Lingdorf's cheapest Steinway system looks like 2 10" subwoofers crossed over to a pair of bipolar Heil tweeters. Probably incorporates automatic room correction. For $22,000 that seems awfully steep to me. Is there more to it than that?

 

[Robh3606]

Not in today's SOTA products. Lyngdorf, Dirac and Trinnov all use proprietary technology to differentiate, measure, and appropriately compensate for direct vs. room reflections. See for example Jan Abildgaard Pedersen, "“Loudspeaker-Room Adaptation for a specific Listening Position using Information about the Complete Sound Field”". That's 2006 and the technology has progressed significantly since then.

To keep things in context to the discussion you cannot EQ a speaker to change it's directivity characteristics. If the off axis response is not smooth no amount or type of EQ can fix it. It is "broken" from the get go.

This is one of the reasons EQ was so hit or miss with older designs before a smooth directivity curve and smooth off axis response was given priorities as design goals. Any modern CD design can be EQ'd with better overall results.

Rob

http://www.audioheritage.org/vbulletin/showthread.php?6687-JBL-Technical-Notes-Volume-1-Number-11

 

[Soundminded]

To keep things in context to the discussion you cannot EQ a speaker to change it's directivity characteristics. If the off axis response is not smooth no amount or type of EQ can fix it. It is "broken" from the get go.

This is one of the reasons EQ was so hit or miss with older designs before a smooth directivity curve and smooth off axis response was given priorities as design goals. Any modern CD design can be EQ'd with better overall results.

Rob

http://www.audioheritage.org/vbulletin/showthread.php?6687-JBL-Technical-Notes-Volume-1-Number-11

The problem with that idea is that it still does not deal with the direct and reverberant fields independently. Therefore it cannot compensate for different acoustics from one room to another without affecting the direct field at the same time.

 

[Robh3606]

The problem with that idea is that it still does not deal with the direct and reverberant fields independently. Therefore it cannot compensate for different acoustics from one room to another without affecting the direct field at the same time.

Of course it doesn't. That's the point you can't EQ them independently. You need a speaker that has good on and off axis response. You have a much better chance of it working with a CD type speaker.

If you have been following this thread you should know about the 70 point measurements Harmon based companies use for there speaker designs. That type of measurement can be used to predict how a speaker will perform in a typical room.

Rob

 

[RUR]

http://www.automatedhk.com/MaterialAttachPart/93/proofread_papper_aes32.pdf

Looks like the same flawed concept to me. The system shown in figure 1 cannot differentiate between equalizing for the direct and reflected field.

As I said, this was 2006 and the technology has moved on. Today's Trinnov products, for example, use a multi-head mic to deliver frequence and time domain information to a dual core 1.8gHz processor for analysis. "Trinnov’s state-of-the art time-frequency analysis algorithms identify room modes, first reflections and late reverberation. Every acoustic aspect is analyzed and compensated with a specific technique." Details are proprietary, but specific user-preference adjustments are provided for managing early and late reflections.

 

[amirm]

Given the important effect the room has on how a speaker will sound, what provisions has a Harman company or anyone else engineered into their product to compensate or even mitigate the effect of that variable, that is to make provisions so that their speakers sound the same or similar in different rooms with different acoustics? None that I can think of.

Rob and Tim beat me to the punch . The whole purpose of this exercise with 70 point measurement and math to compute the target is to predict how a speaker sounds in a room. You saw in my graph that as I changed the position of the mic, it did NOT change the mid to high frequencies that much. This means the speaker timbre was highly room independent.

And as Rob points out, once you have a well behaved speaker, then you have a shot at using EQ to modify its sound. Because the speaker direct and indirect response are similar, both are modified as you wish. Whereas if the tonal change is coming from the fact that direct and indirect sounds are different, no EQ up stream of the speaker can fix that.

Tone controls, equalizers, and level controls by themselves don't work because they change both the direct and reverberant sound field simultaneously.

Exactly. So if the two are the same per above, then you get around this issue above transition frequency. Below transition frequency the room dominates and techniques for that involve other means that the speaker itself.

One effect pulling speakers away from the walls of a room has is to reduce to a degree the effect of the room, so does building a LEDE room. So does trying to make the speakers as directional as possible. This last characteristic means you have to position and aim your speakers with great precision and you must sit in one spot with your head in a vice or you will not hear the intended effect. Is that what accuracy means? From what I've gleaned it does to a lot of audiophiles (I do not consider myself an audiophile in the commonly used sense of the word.)

The LEDE (Live End, Dead End) concept as you may know came about from speakers that had poor off-axis response. So folks blocked the reflections as much as they could. The purposes of a speaker with great off-axis response or one that doesn't let much go off-axis is to do away with this requirement.

Research shows that we actually like reflected sounds from the sides. I will cover this in a future article, post. But the conventional wisdom that you must absorb first reflections is not correct.

Since most people's preponderance of listening experience relates to recorded music, what qualifies them to decide on what is accurate and what isn't?

What is the alternative? Not asking them? Isn't the ultimate goal to play something and enjoy it? If I enjoy one speaker more than another, shouldn't I get the one that I enjoy more?

Yes, this is a broken system where we do not capture the recording room and hence, can't know when we have duplicated it. This doesn't mean we throw our hand up in the air and pick on some other basis. The ultimate test is when I listen to a piece of music, which sounds better to me. We pick equipment this way all the time. Why not speakers?

Mind you, I had a hard time with this too until I sat through the tests and it then it all made sense.

That ability relates to memory reinforced by frequently repeated exposure to the real thing.

I don't agree fully if you mean listening to a lot of live music. You can do that as much as you like but you never know what the James Taylor song that I listened to in blind tests was supposed to sound like. The judgement then is based on what we think is "good sound." I gave the example of amp distortion. In high amounts we can tell it is bad by itself. We need not have a reference point at all. I could play an instrument that you have never heard and you can still tell me it is distorted.

Nothing about this hobby at the end of the day is to recreate the live event but to do justice to the music as delivered to us. There, having good frequency response should be at the top of the list. Relying on a dead room to get rid of reflections should not be it.

So along with their hearing tests, what is the frequency of their attendance at live performances where only acoustic instruments and no electronic sound reinforcement system gave them a reference that's reliable enough to take their opinions credulously?

Harman has tested different groups of people with such experience and without. The theme is the same in all as far as what sound we like. What varies between groups is how bad we rate something that is bad. And how accurately we score. http://seanolive.blogspot.com/2008/12/loudspeaker-preferences-of-trained.html

"To study this question, the author conducted a large study (see reference 1) that compared the loudspeaker preferences of 300+ untrained and trained listeners.... An important conclusion from this study is that the loudspeaker preferences of trained listeners can be safely extrapolated to the tastes of consumers having little or no formal listener training."

So while this is a comparison between trained and untrained listeners, we see that the core preferences are remain even in large scale tests. Look at how speaker "M" did worse than others no matter which group of listeners scored it. Ditto for speaker B.

 

[Phelonious Ponk]

While what you say is true about every seat even in one performance venue sounding different from every other seat, there are certain common characteristics of the acoustic effects large venues create that cannot be captured or reproduced by our current technology no matter how expensive or elaborate. This not only includes the auditory sense of space but tonality of musical instruments as they are heard in those venues as well. In other words, if you are trying to reproduce the tone of musical instruments as heard at a concert hall with a sound system, (binaural recordings made with a dummy head at a seat in the audience played through headphones excepted) no matter what frequency response your system has, it's wrong. And that doesn't even take into account that spectral balance varies from one recording to the next.

Agreed.

The best that can be done with a 2 channel sound system IMO even after correcting for spectral reflection distortions and variables in the recording is to reproduce the tone of the instruments as they would be heard were they in your listening room.

Agreed in broad terms. I'm sure you understand that to do this literally, the instruments would have to be recorded in your listening room. But we can get very close.

That done, that does not mean they will be spatially accurate. I can get my main speakers in my best system to match the timbre of the Steinway piano in the same room but there's still a big difference. The piano being largely an indirect radiator fills up half the room with sound as an enormous source, the speakers are a pair of point sources. You can tell the difference immediately. The speaker/overall system design in this regard remains still flawed.

Agreed again. On a good recording, even a lightly processed one, I can certainly hear the difference between a Gibson Jumbo and a Martin Orchestra Model. But they not only disperse differently than speakers, they disperse differently from different parts of the neck, body, etc. You can pretty radically change the sound of an acoustic instrument by moving the microphone. Yes, the difference is immediately evident, and we're talking about one instrument. The playback system that would be required to even reasonably emulate the spatial characteristics of a jazz quartet - in any space - is difficult to even imagine. The drum kit alone would require at least a half dozen channels, with different transducer types, most of which have not yet been invented, for each element in the kit.

Yet people believe, and will not be disuaded.

Tim

 

[LL21]

The best that can be done with a 2 channel sound system IMO even after correcting for spectral reflection distortions and variables in the recording is to reproduce the tone of the instruments as they would be heard were they in your listening room. That done, that does not mean they will be spatially accurate. I can get my main speakers in my best system to match the timbre of the Steinway piano in the same room but there's still a big difference. The piano being largely an indirect radiator fills up half the room with sound as an enormous source, the speakers are a pair of point sources. You can tell the difference immediately. The speaker/overall system design in this regard remains still flawed.

AGreed. I studied piano for 12 years...i am fortunate to listen to a nice system. reasonably well setup. Still ain't the same. I focus more on purity of tonality than anything else...i dont listen in the sweetspot that often. i focus next on deep tacticle bass/fullness of sound, then detail/decay and noisefloor. the magic for me is when these attributes manage to reveal the nuances of music. for example, youd be surprised to know how often contrapuntal themes occur in music, and are lost somehow in music playback. You dont get that one set of strings and the toher are actualy playing off of one antoher. That is the part of music playback i think systems CAN achieve. The tonality is why i stick with my zanden despite the advances in digital tech in terms of dtail and noise floor. As the system has become more rsolved, those contrapuntal themes have started to come thru. i dont try to recreate the hall, or create super deep soundstages, just the tonality, fullness of extension, and that quickness that (i think) is part of capturig those many themes in the music in understandable form.

 

[Phelonious Ponk]

Nothing about this hobby at the end of the day is to recreate the live event but to do justice to the music as delivered to us.

Thank you. Perhaps that should be my new signature.

Tim