Research Project: Infinity IL10 Speaker Review & Measurements

[pozz]

We all talk Klippel / Toole / Olive, aren't we?

I get tired of discussing again and again any insights of the three gentlemen down to the smallest detail. Sometimes, please do not be offended, it reads like a biblical interpretation. Obviously, some of the participants have not had much contact with science. I live in it, trapped. Poor me. I need some air.

It's no problem. What I get tired of is an attack on the passing, easy-to-dismiss observations when it's much more productive to address the data. It's not like much of the stuff in Toole et al is new to this forum. It's been discussed endlessly, especially the irrelevance of THD. What we continue to have difficulty with is understanding the source of differences for similarly-measuring speakers, either for Amir or when other members conduct their own listening tests at home.

Read again the 1st post and you'll find this:

"Combine that with the fact that unless one is trained, hearing distortion is difficult and you or at least I arrive at the fact that distortion matters once you take care of tonality. "

My point is that any sighted listening contains bias and trained listeners are not immune to that. I have backed up that claim with a link from Olive's blog containing his research. As my point is not related to Amir it cannot be "ad hominem" - I stated it clearly in this post, so you may consider first reading someones post before accusing him of ad hominem attack.

Let me put it simple for you: seeing the distortion measurement first and than claiming you are hearing it in a sighted listening test and that it is not related to sighted bias because you are a trained listener is simply false, as explained in Olive's article.

Fair point. I tend to skip over the listening impression section. That said, you went far out of your way to point to unreliability given that Amir had already said that he could be subject to sighted bias there. More than enough reason to give the benefit of the doubt, keep the discussion civil and give a better answer in terms of all the assembled research, prime among which is the measured data itself.

Thing is, if we go with Amir's unreliability then things become really uncertain and the path forward unclear. How do we examine the measured results and data? If we say about the spins: no, the place of distortion is underestimated, then I don't see a way forward. If we say that spins are entirely wrong for assessing measured performance then what's better?

Unless I've not been following along, the only alternatives that have been proposed concern power compression, IMD results or inaccurate NFS measurements. No one as far as I know has put forward different research showing that there is another, better way of measuring speaker performance and given concrete reasons why which have not already been addressed in the Olive/Toole research.

So, in that case, we have to assign some reliability. Given that Amir's a trained listener we can assume 1) that his reports of tonal differences are more consistent than those of others and 2) that he is better capable than many of hearing spectral differences due to resonances and mechanical issues. That's enough for me to say that the best way forward for understanding his impression of this speaker is to find others which, again, are largely similar. My money's on sound power because it makes sense: it is about the total distribution of energy radiated by the speaker from all sides into the room, which is best captured in mono testing.

The only conclusion I've come to is that there is more work to be done, and the way forward is testing and data analysis. The fact that there are some inaccuracies due to measurement setup is significant, but not overly so given how small they are. However, the difference @napilopez showed in the main FR curves is both large and significant. Finding large differences is step one. Establishing their source and contribution is step two, and indeed the goal. I do not think I am wrong here.

 

[pozz]

@pozz, would this be ad honimem?

In my post I clearly stated that my opinion is not related to Amir in person:
"Btw, don't take it personal, I wouldn't trust Olive's sighted listening test done in your room after peeking at measurements. I simply don't believe in sighted test done by a single person. I treat it as personal opinion, not measured fact. "

But here Amir's post seems to be addressed to me personnaly and not to my opinion, which perfectly fits into "ad hominem" definiton, so do you dare to ask him "if that is really necessary"?

Why would you enter the thread only to concentrate on the most trival aspects, which you moreover know to be trivial? Why do you persist in sarcastic equivocation? What have you achieved? You're satisfied with fouling to win a point?

 

[RayDunzl]

Just addressing the larger point of many audio enthusiasts thinking they can readily hear distortion.

We've all heard distortion.

Identifying, classifying, quantifying, correlating, and sometimes even recognizing distortion would, to me, be different tasks.

 

[GelbeMusik]

Daily reminder that harmonic distortion is an engineering conceit and not a perceptual metric. ... but if we're talking about the subtle grunginess in the mids and treble, harmonic distortion may point to the problem but it is an unreliable and useless way to compare and measure.

Not useless. There shall be no "problem". Again it is about cutting corners. It is absolutely possible to design a speaker without that nasty peaks in distortion. It may cost a dollar more or two--literally, but is it worth to aim for the most cheap rendition of a speaker? It is a decandent luxury product anyway, so?

Or how about when Amirm says that he's professionally trained to hear minor traces of distortion like he's some kind of audiophile Jason Bourne.

It may be possible. Again, if distortion is easily avoidable, what is all the discussion about? When I read Amirm's distortion graphs and there are peaks: skip, no further reading, really! No discussion, that's just fail, end of story. Of course for Harman the case looks different. Would the customer recognize the cheap skating? Big science.


For me as an DIYer the asthetics of a speaker starts with fine objective data. It should be capable of delivering a pleasent, elevated sound volume without strain. There is a lower limit, for sure. Otherwise it would never deserve the tag "high fidelity", and to me would it be just worthless. ( A binary parameter ;-) My listening window goes to +30° / - 10° in the vertical, which deviates from Olive's ideas a lot for some reason.

O/k, I won't bother You any further. From my perspective the findings of Toole / Olive are too often used to easy handed justify a general decline in quality. Even the test program of music is quite humble. As I really disliked "Fast Car" right from the start--I experienced its advent as a vinyl record in HiFi studios in person, it wonders me, why there is a 2015 remaster which sucks even the very rest of musicality out of it. If it is so easily altered, but still a "most revealing" discriminator for speaker quality, what should I say? I'm out.

 

[pozz]

From my perspective the findings of Toole / Olive are too often used to easy handed justify a general decline in quality.

You mean like a relentless pursuit of FR regardless of industrial quality?

 

[napilopez]

We've all heard distortion.

Identifying, classifying, quantifying, correlating, and sometimes even recognizing distortion would, to me, be different tasks.

yeah, that's a better way of putting it. In the context of the conversation, I would mean identifying negative distortion (as in, distortion that sounds bad or detracts from the musical experience/neutrality).

 

[GelbeMusik]

You mean like a relentless pursuit of FR regardless of industrial quality?

Nope. Of course FR is important. While dicussing the case, though, I've found so many caveats against the Toole / Olive "science" that I'm a bit shocked. I celebrate the attempt to establish a common (!) standard for home and studio. Same with the spinorama as its foundation. This I keep, well done, guys!

But,

(artificial beak)

the keen followship here seems to miss the clue. I've used an upmost qualified 12 inch bass/mid plus a top notch competent waveguide for a decade now. People tell me, that along some spinorama my speakers could easily be replaced with 4 inch woofer plus a bare dome tweeter?! Spikes in distortion figures, never mind. LOL, rolling on the floor ;-) Not the least my "system" costs less than any of those alternatives.

The advice is based on a survey which made common folks place a verdict on speakers listening to a musical program devoid of any inherent asthetics? Natural considerations regarding the validity of the setup are hidden behind statistics.

Again, High Fidelity is a convenience feature. If the music is right, You won't care too much. In this sense, go on.

Example: while searching for an alternative for "Fast Car" I came along Kate Bush of "Running Up That Hill" fame. I think Sony or Technics eventually bought it. Again, goose bumps. Why? It is music in the end, despite its studio origin. And due to hifi I have some muscle at hand to deal with that necessarily elevated level ... Somebody put some brains into it. Poor Tracy Chapman, I'm absolutely convinced she was tricked into the sad recording.

I'll be back with a counterexample for the hyped-to-death Brchhrdt fourhundred ess

 

[ROOSKIE]

Your doctor's opinion about what is wrong with you is much more accurate than you doing the same analysis by googling.
.

Amir with all due respect, this may be where some of the schism comes into play.
You are incorrect with this statement.
You doctors onion about what is wrong with you is not more accurate.
It is however more likely to be accurate. It likely has a higher probability of being accurate but it is not inherently "more" accurate.
This would not be semantics, it is the core issue. It science and logic.

This is why you should always get second and third opinions and even google yourself and ask questions. It is very likely many of us have had our doctors give horrid advice and or misdiagnose an issue. I know my mom is alive due to second and third opinions, including her own research.

Anyway, this is a great discussion on what I consider to be the hottest topic on this site. Harmonic distortion.

 

[Blumlein 88]

On Distortion:


I find my little JBLs to become unappealing at higher levels. They're fine for casual listening, especially if not being dmanding of the imaging.

It doesn't have to be too too high a level before they create that "grunge" sound.


Two speakers: JBL LSR 308 and MartinLogan reQuest

The frequency response, after correction, as measured here, at the listening position (10 feet), of each, is very similar - target was "flat".

That leaves dispersion and distortion as suspects for the displeasure.

Dispersion differences affect the 'stereo", but don't change distortion.

So, distortion seems (to me) to be the fault noticed at higher SPL with the little JBLs.




Simplistic measure:

Speakers in the same room, JBL adjacent to but outboard of the ML.

82dB is not a particularly loud level, less than a watt of power applied.

JBL cone

ML electrostat

Swept sine THD, dB relative at a higher level:

View attachment 70705

And THD % relative to the fundamental

View attachment 70706


*the JBLs are my daily drivers for most video and casual music when I'm inattentive. When I really want to hear music, I power up (literally) the ML. They've never sounded "strained". 110dB unweighted "flat FR" peaks, no problem. But, they're big speakers, and I have plenty of amplification to goose them.

I think part of what grunges up the LSR series is they do compress the signal at higher signal levels. As in the DSP chips compress signals actively not a passive result. This keeps peak signal levels under control letting the small speakers belt out a bit louder overall signal.

 

[amirm]

This is why you should always get second and third opinions and even google yourself and ask questions.

Well, I give you an example of googling. My doctor prescribed medication that I thought was causing a side effect. In the next visit I tell him and he says no, that is not likely to be the case. I go home and google. The first hit I get on side effects of that drug is what I was experiencing. Now get this: the hit was to the actual research results of the drug by the company! Right there, in black and white it says this is one of the known side effects.

So on next visit I tell my doctor he is wrong and that I read my side effect was caused by the medicine. He asked me where I had read it. I say it was in the documentation for the drug certification. He calmly explains to me that in drug trials they document anything that the subjects reports whether it is actually caused by the medication or not. He said doctors that prescribe these medications thousands of times build up their own body of evidence of what side effects are likely and what they are not. Well it turned out he was right as the side effect was caused by something else.

In this case, I am not a random dude reading research about controlled listening tests. It was part of my professional job and responsibility to practice it. Based on that, I am saying you must not play debating games saying you read that sighted tests are no good and that is that. The world doesn't turn that way. Not with your doctor. Not with the Ice Cream taster. Not with trained audio listeners.

As I keep saying like a broken record, none of this means 100% reliability. If we have such a crystal ball, we would close this forum and just go by that data point. We don't have it. We need to instead rely on collective wisdom and make educated guesses.

 

[ROOSKIE]

On Distortion: ....................

So, distortion seems (to me) to be the fault noticed at higher SPL with the little JBLs.

I appreciate the efforts.

Harmonic distortion, which is what you have in the chart is one category of distortion. There are others.
I really want to point this out so that folks use the term Harmonic distortion when they refer to such distortion.
There are many types of distortion, so the general term "distortion" can be misleading. A frequency error is distortion, IMD is different from HD/THD, compression is different yet, lack of dynamic reach is a distortion, high noise floor, phase distortion, transient response, lack of scale, Noise such as port chuffing, woofer bottoming, resonances, ect
The list is longer than my hobbyist brain can remember.

This isn't to be picky but to keep moving toward some clarity. That way we can figure out what is what. Is the Harmonic distortion affecting Amirs listening in the way claimed or is it actually some other aspect of the replay.

Is Harmonic Distortion what is affect your experience JBL308 or is it something else or a combo?
I personally found the JBL308 (while sounding good enough)to sound a bit murky at all volumes. Maybe you notice it more at higher levels? Maybe it is compression? Maybe it is in fact HD?

 

[Blumlein 88]

Well, I give you an example of googling. My doctor prescribed medication that I thought was causing a side effect. In the next visit I tell him and he says no, that is not likely to be the case. I go home and google. The first hit I get on side effects of that drug is what I was experiencing. Now get this: the hit was to the actual research results of the drug by the company! Right there, in black and white it says this is one of the known side effects.

So on next visit I tell my doctor he is wrong and that I read my side effect was caused by the medicine. He asked me where I had read it. I say it was in the documentation for the drug certification. He calmly explains to me that in drug trials they document anything that the subjects reports whether it is actually caused by the medication or not. He said doctors that prescribe these medications thousands of times build up their own body of evidence of what side effects are likely and what they are not. Well it turned out he was right as the side effect was caused by something else.

In this case, I am not a random dude reading research about controlled listening tests. It was part of my professional job and responsibility to practice it. Based on that, I am saying you must not play debating games saying you read that sighted tests are no good and that is that. The world doesn't turn that way. Not with your doctor. Not with the Ice Cream taster. Not with trained audio listeners.

As I keep saying like a broken record, none of this means 100% reliability. If we have such a crystal ball, we would close this forum and just go by that data point. We don't have it. We need to instead rely on collective wisdom and make educated guesses.

Anecdotes:

Okay, my Uncle is in the hospital for horrible balance problems and passing out a couple times unexpectedly each day. His doctor is reviewing his medicine taking things away one at a time and eventually restarts it all with half of it not being used. The problem comes right back. I'm there with nothing better to do so look at all the various side effects. I find one of the meds lists what amounts to an almost exact description of his problem. This list broke things down much more. As you say companies have to report even one report on something and less reported things are usually not actually due to the medicine. But this effect though listed as rarely encountered was from multiple reports and not likely meaningless.

Next time the doc is in I ask him about it, indicate I know they have to report everything, but this rarely encountered side effect was one listed several times. He looks at what I have, said it is the same list he uses because it breaks things down more precisely and said he missed this. Stops the medicine and the problem goes away.

Proving nothing about anything either way of course except in any given real world problem just have to keep hammering away at all the possibilities until something gives. A pro is more likely to get it, and more likely to know what to ignore. Sometimes weird enough problems just need luck and perseverance finding the solution. A motivated layman with hours of time on his hands, and no other things to look at can find such things even pros miss. Problems take what they take to find solutions to.

 

[ROOSKIE]

Well, I give you an example of googling. My doctor prescribed medication that I thought was causing a side effect. In the next visit I tell him and he says no, that is not likely to be the case. I go home and google. The first hit I get on side effects of that drug is what I was experiencing. Now get this: the hit was to the actual research results of the drug by the company! Right there, in black and white it says this is one of the known side effects.

So on next visit I tell my doctor he is wrong and that I read my side effect was caused by the medicine. He asked me where I had read it. I say it was in the documentation for the drug certification. He calmly explains to me that in drug trials they document anything that the subjects reports whether it is actually caused by the medication or not. He said doctors that prescribe these medications thousands of times build up their own body of evidence of what side effects are likely and what they are not. Well it turned out he was right as the side effect was caused by something else.

In this case, I am not a random dude reading research about controlled listening tests. It was part of my professional job and responsibility to practice it. Based on that, I am saying you must not play debating games saying you read that sighted tests are no good and that is that. The world doesn't turn that way. Not with your doctor. Not with the Ice Cream taster. Not with trained audio listeners.

As I keep saying like a broken record, none of this means 100% reliability. If we have such a crystal ball, we would close this forum and just go by that data point. We don't have it. We need to instead rely on collective wisdom and make educated guesses.

Yes, this journey often happens as well. But I am glad you double and triple checked man, really is a good idea. I already mentioned my mom but I have several stories where that extra effort paid off big with medical situations.

To be clear, I in no way expect you to be accurate all the time. Obviously I want you to be open enough to adapt to good evidence, beyond that it is much better to have a strong personality running this site vs a mushy soul. Lots of data and opinions tossed around here. It does indeed take wisdom to parse it out and make a smart choice.

By the way, IMHO you heard what you heard. I just want to know what it was in terms of the measurements. I have heard various things myself and am trying to correlate the sound with the available data.

 

[bobbooo]

I don't really think typical listening room has average reverberation time of 0.3 sec, at least not in EU. If I have to guess it would be closer to 0.5 sec.

Olive mentions an average measured reverberation time in domestic rooms of 0.4 seconds in Part II of the paper on his speaker preference formula (I've included the following paragraph too due to its relevance to other discussions in this thread):

The title of the study by Bradley referenced above is "Acoustical measurements in some Canadian homes", which is exactly what it says on the tin. I would imagine the average European lounge / listening room is smaller than the average Canadian one, and as reverberation time is proportional to room size, the 0.3 seconds of Olive's listening room is likely to correspond to the European average pretty well I'd say, probably even better than to the average Canadian / American room (assuming the latter two are approximately the same).

 

[mhardy6647]

It is the difference between theory, reading stuff online, and practicing and understanding it.

darned tootin' (and true in all disciplines).

To quote the late, great Earl Weaver:

 

[Rick Sykora]

As a research project, is it possible with DIY means to deliberately construct speakers with specific sound power DI? @617 @Rick Sykora

The short answer is yes. Naturally, a better approach would be a single speaker that could (predictably) change its directivity. While a dipole could be used, another approach would be to use a CBT. I do not have much experience with CBT design, but @Bjorn is already working on just such a design.

 

[Blumlein 88]

Looking more closely at the predicted in room responses, I'd venture the excess and unsmooth response above 4 khz is where the grunginess is coming from. The variable being what Amir hears as grungy and lacking clarity. One of those cases where were I there and heard the same thing and Amir said this sounds grungy to me, then I'd have a good reference for what he means.

Audio Technica AT4033 microphones have an uneven and slightly hot response in that range, and recording with that microphone I had similar feelings about it. It seems okay, but just a bit rough and grungy. Clarity never quite was achieved either. Like it had to be louder than usual to hear what was going on.

I'd alter my earlier suggestion and suggest a 3 db shelf above 4 khz.

 

[ctrl]

Harmonic distortion and IMD are just a different way that the same non-linearity manifests itself. THD is the effect as measured with a single tone, while IMD is with multiple tones.

You can't say that. With the two-tone or multitone measurement there are other distortion mechanisms that do not occur with the THD measurement.
There are distortions caused by amplitude modulation (AM) and frequency modulation (FM).
More information can be found on the Klippel website and in the purifi-audio.com blog:
Doppler distortion vs IMD?
Low frequency harmonic distortion is almost inaudible. So what’s the point of low distortion drivers?

Howdy just to point out test tones are a far cry from music and so unfortunately only represent the audibility threshold with the tones.

But you should also consider the result of the evaluation:

With the 80dB masker, HD3 is theoretically audible from an attenuation of -53dB (0.2%), with the 100dB masker from about 1%.
Thus @amirm's perception may well be correct, since with 1.5% HD3 the detection threshold is far exceeded.

The masking of the test sound is thus no longer given at -53dB attenuation (with the 80dB masker) - for a test sound with 3x base frequency, which corresponds to our HD3 in terms of frequency.

The measured 1.5% [email protected] of the loudspeaker corresponds to a damping of only -36dB. This is far, far above the perception threshold of a test tone.

Some quotes from Geddes/Lee study about distortion perception:

Don't discuss it with me, discuss it with Dr. Geddes.

Have you actually read the paper and what are you trying to say?
@amirm does not only show THD in the distortion measurements, but also HD2...HD5.

What does Dr. Geddes say in this paper? Actually nothing more than that THD alone is not enough to reflect the hearing impressions realistically.

He therefore suggests weighting the individual harmonic distortions in a certain way and calls this "Gedlee Metric". As others suggested before, this depends on the masking of the harmonic distortions, but also on the amplitude itself (as shown in post#118, the masking also depends on the sound pressure). There is another peculiarity (inclusion of the phase) that is relevant for amplifier technology.

Especially after the GedLee-paper you have linked, the reference made by @amirm to the raised HD3 is relevant, because it is less well masked.

What do we do (at least I do that ;-)) here in the forum when the measured harmonic distortions are examined?
We consider the order of the harmonic distortions and "weight" them accordingly, since higher order HD are less well masked.

Then the sound pressure should also be taken into account during the measurement. Therefore, the measured 1.5% HD3 at 86dB is already pretty bad. Probably 5% HD3 at 105dB would be less of a problem.

Together with the 4dB peak between 4-5kHz and the possibly slow decay at 1.2kHz ... a combination from hell.

 

[youngho]

Looking more closely at the predicted in room responses, I'd venture the excess and unsmooth response above 4 khz is where the grunginess is coming from. The variable being what Amir hears as grungy and lacking clarity. One of those cases where were I there and heard the same thing and Amir said this sounds grungy to me, then I'd have a good reference for what he means.

Audio Technica AT4033 microphones have an uneven and slightly hot response in that range, and recording with that microphone I had similar feelings about it. It seems okay, but just a bit rough and grungy. Clarity never quite was achieved either. Like it had to be louder than usual to hear what was going on.

I'd alter my earlier suggestion and suggest a 3 db shelf above 4 khz.

I was going to comment earlier, before Q's phallus-waving, but I think the PIR bumps at 700 Hz and a bit above 1 kHz are also problematic. I also thought that it was quite likely that Amir was responding to resonances but misidentifying distortion as the culprit, but that's only my speculation.

 

[ROOSKIE]

Then the sound pressure should also be taken into account during the measurement. Therefore, the measured 1.5% HD3 at 86dB is already pretty bad. Probably 5% HD3 at 105dB would be less of a problem.
.

Hmmm. With all due respect, why in world would 5%THD at 105db be less of a problem than 1.5% at 86. What evidence do you have to support your claim? Lets have it.

But you should also consider the result of the evaluation:

The masking of the test sound is thus no longer given at -53dB attenuation (with the 80dB masker) - for a test sound with 3x base frequency, which corresponds to our HD3 in terms of frequency.

The measured 1.5% [email protected] of the loudspeaker corresponds to a damping of only -36dB. This is far, far above the perception threshold of a test tone.

I did consider them. That was a test of the audibility threshold with test tones. threshold... This just when one can start hearing this Hd with a simplified sound. This says nothing about music, which is fantastically complex and no one should infer anything about music as that was not tested in that test at all.
They (music and test tones)may be the same, they may be fairly different, they may be wildly different.