[DanWiggins]

Kevin Haskins wrote:

What is the best way to set levels for distortion testing?   Obviously if your working with a 6.5" driver and trying to normalize it somewhere for unit to unit distortion test you have to calibrate it to a level.   If you pick an individual frequency you may end up on a peak or dip of an individual driver.   If you pick an average over a range you have to choose the range & that might be fraught with peril if your unit has a large dip or peak in that range.

Correct on all counts - it's difficult to do.  I tend to set a nominal level as averaged (1/12th octave spacings) over the center 2 octaves (or most important range; for a 6.5" woofer, that would be typically 250 Hz to 1 kHz).

The other way is to show a true relative distortion level to the output; graphs that show the fundamental (frequency response) along with the harmonics are good, but you must interpret the harmonics relative to the fundamental level.  You can't just look at the harmonics and say "lower is better!" with those types of graphs.

For example, at the 6.5" woofer HD test, look at the Extremis and the AA6.5 (often compared against each other).

Down in the deep frequencies - right at 20 Hz, we see the following for the first two harmonics:

Extremis: -52 dB/2nd    -44 dB/3rd   (0.25% and 0.63%)

AA6.5: -56 dB/2nd   -56 dB/3rd   (0.16% and 0.16%)

MANY have looked at that and said "see!  The AA6.5 has less THD than the Extremis - XBL² is junk, it doesn't work!"

BUT, look at the output levels:

Extremis: -2 dB

AA6.5: -16 dB

So the Extremis was generating 14 dB MORE output at those distortion levels!  Now, you typically cannot just add the output offsets to the distortion levels (it's typically highly conservative; THD in a driver does not increase linearly with excursion), but for the quickest shot, we'll do it:

Extremis: -50 dB/2nd    -42 dB/3rd   (0.31% and 0.79%)

AA6.5: -40 dB/2nd     -40 dB/3rd   (1% and 1%)

Pretty different now, eh?  That big advantage is completely gone once we consider output level.  This is what you would expect - worse "conservative" case - if the output levels were equalized.

Fundamentally, THD testing is a tricky thing.  IMHO you need to always consider the following:

1.  Measurement setup - am I in the proper acoustical setting (near/far field), is the ambient noise low enough to accurately measure the levels I want?

2.  Am I setting the relative SPLs in a significant/relevant way?

3.  Is the data as presented easy to understand/simple to reference to each other, for the target audience?

[DanWiggins]

ShinOBIWAN posted:

However much depends on the baffle size when doing nearfield

Not with regard to the near-field/far-field transition.  It's strictly a function of the diameter of the piston and the wavelength.  Measuring within the near-field is something that should be avoided, unless you acknowledge the significant issues related to such.

believe what John originally started was doing measurements for his own curiosity and to better what he builds. He decided at some point to share this data and, because of popular demand plus the lack of independant testing, has now grown the website for the benefit of level headed enthusiasts. Unfortunately you can't keep all the idiots out of such a place and you get misinformation spreading. The keyword here is context.

And I applaud him for it!  It is interesting data to see, at the very least.  My only issue is that sometimes - well, actually MOST of the time - the data as presented isn't really indicative of the driver performance (because of near/far field issues), AND the graphs are not scaled to each other.  So I often get e-mails (like 4-5 a week) from someone who asks "why does the Extremis have such bad THD at Zaph Audio, especially down low?" without understanding the issue with output levels.

For the typical reader, the graphs are difficult to understand; I'd like to see the THD levels "normalized" to a flat frequency response, to get a real "apples to apples" comparison of what each driver does at a given SPL level.

If I get 4-5 per week, there's probably 20-30X that many each week not asking, just assuming it as the truth...

[Danny Richie]

Dan,

I completely agree with your last post. I don't get 4 or 5 a week but I have had more than I want to spend my time responding to.

I'd be nice to be able to communicate and have some dialog to share measuring information and strive towards some standardized testing, but when you send an e-mail to a guy and get, "Don't email me or call me, I'm not interested in having a "discussion" with you in much the same way I'm not interested in petting a rabid Chihuahua after I kicked it off my leg.", then that makes it pretty tough.

I know that if another industry professional contacted me with any interest I would be inclined to listen.

[Kevin Haskins]

Dan,

I completely agree with your last post. I don't get 4 or 5 a week but I have had more than I want to spend my time responding to.

I'd be nice to be able to communicate and have some dialog to share measuring information and strive towards some standardized testing, but when you send an e-mail to a guy and get, "Don't email me or call me, I'm not interested in having a "discussion" with you in much the same way I'm not interested in petting a rabid Chihuahua after I kicked it off my leg.", then that makes it pretty tough.

I know that if another industry professional contacted me with any interest I would be inclined to listen.

To be fair to John, you talked down to him as though he were and idiot.    Your not going to get far with anyone when you tell them you think they are "early on the learning curve".      So... I think you stepped into the Chihuahua comment.      

All of us are people and you have to realize that people don't respond well to being considered ignorant, even those for whom it is true (maybe especially those who for whom it is true).

[Danny Richie]

To be fair to John, you talked down to him as though he were and idiot.

That certainly was not the intent, but I can see how it can be taken as such. I'm certainly not perfect either.

But keep in mind that his animosity towards me came about long before that. His unfounded and demeaning comments made about the M-130 were several years ago, and long before I ever said a word to him. I have always been the snake oil salesman in his eyes.

[Daygloworange]

To be fair to John, you talked down to him as though he were and idiot.    Your not going to get far with anyone when you tell them you think they are "early on the learning curve".      So... I think you stepped into the Chihuahua comment.      

Kevin, I believe you're assuming that, more than anything. Sometimes you truncate your thoughts on the keyboard while responding to someone. Danny is anything but condescending in person.

All of us are people and you have to realize that people don't respond well to being considered ignorant, even those for whom it is true (maybe especially those who for whom it is true).

Again, Kevin, not all of us respond the same. I'm a blunt person. I appreciate candor. I prefer to not waste time doing something twice, or having to undo something. If I'm truly early in the learning curve, that's something I need to know. If I don't know enough to know that already, then I'd rather have someone point it out, and I am grateful.

Cheers

[Kevin Haskins]

Well...   that is easy to understand also.   

I don't think most engineers are going to see casual listening test as a valid method of reliable data.    You make absolute statements about the audibility of various unmeasurable phenomena without any actual repeatable data to support it.   

I'm not pointing my finger, to some extent we all do it.    John makes a lot of statements about which distortion he thinks is "definitely audible", or that "high x-max" drivers sound worse without any real explanation for the comment based upon either physics, measured data or any other support.   It is said based upon his subjective experience which we all know is flawed.   He also spends more time covering his tracks though by making a lot of disclaimers about his subjective comments.

You on the other hand don't make the disclaimers as obvious.   It becomes an absolute statement of truth rather than a subjective preference.   When you do that about something that has zero physical data to prove it (the audibility of wire break-in) you open yourself up wide for people with technical backgrounds to call "snake-oil!".    John probably long ago placed you in the snake-oil category.    If you just cover your tracks better with a couple disclaimers concerning the subjective nature of your opinions you won't draw as much fire.

[Watson]

Let me clarify a few things...

The white trace is an Extremis mounted in a cabinet.

The other traces are a raw Extremis, no baffle, no cabinet.  Essentially a free-air driver being measured.

I appreciate the clarification, though perhaps you should have mentioned this when you posted the original graph, otherwise it is arguably quite misleading in that it significantly overemphasizes the effect of the nearfield measurement.

[Danny Richie]

Let's flip back a few pages to Dan's comments about near field measurements and see if he is correct.

Here is the M-130 measured in the near field at 1", 2", 4" and 8". Measurements are in order from Red, to Orange, to Yellow, and then to Green.



Hey look a dipped area in the near field that drops off (goes away) with distance. It's right in the 2kHz range. A near field anomaly? It looks like it to me. Where is that crow eater?

It looks like a little baffle step loss is starting to occur as we get further away too. It is in a 8" wide baffle (A/V-1 box), and the responses are not quite the same. is it any wonder that a 1 watt/1 meter point is an industry standard?

Since the response changes with distance in the near field then what distance do we pick as a reference?

It looks like Dan was right about this.

He also stated that issues with frequency response will effect THD as well.

Lets see if that is true. Here are more measurements of the M-130. All Red lines are at 1", all Orange lines are at 2", all Yellow lines are at 4", and all Green lines are at 8". I set the output level as high as I could without overloading the mic to get the highest level of signal to noise ratio.

Here are second order distortions.



Note that as distance, and amplitude, and near field anomalies changed, so did distortion. There are some peaks where there were dips and dips where there were peaks. Which is accurate? 

How about 3rd order harmonics?



Now I see even more changes.

How about 4th order harmonics?



How about 5th order harmonics?



It looks like Dan was right about that too.

Boy these near field distortion measurements are tricky huh?

[Watson]

And I applaud him for it!  It is interesting data to see, at the very least.  My only issue is that sometimes - well, actually MOST of the time - the data as presented isn't really indicative of the driver performance (because of near/far field issues), AND the graphs are not scaled to each other.  So I often get e-mails (like 4-5 a week) from someone who asks "why does the Extremis have such bad THD at Zaph Audio, especially down low?" without understanding the issue with output levels.

For the typical reader, the graphs are difficult to understand; I'd like to see the THD levels "normalized" to a flat frequency response, to get a real "apples to apples" comparison of what each driver does at a given SPL level.

Normalization would do more harm than good, because it misrepresents how the drivers are typically used in practice.  (If most speakers were built using active equalization on the low end, then you would have a point, but such setups are in the minority, except in the car audio and subwoofer realms.)

Zaph's data does not misrepresent the Extremis in a typical context -- quite the opposite in fact.  What is happening is that the Extremis is generating more output down low, but that output is lower quality in that it has higher distortion.  Zaph's data accurately conveys this.

[ShinOBIWAN]

Exactly what is the purpose of John's website?  I can't quite figure that out.  Apparently, he's not related to any of these companies, doesn't use any of the these products, doesn't really understand what his results indicate, and yet publishes results under the guise of knowing more than he really knows.  If his measurements aren't to be used as a reference, then what are they to be used for? 

So, what the heck is the purpose of John's website? 

Well if you don't know that then clearly your in the wrong place

See the bit where I talked about idiots.

[Kevin Haskins]

 Sometimes you truncate your thoughts on the keyboard while responding to someone. Danny is anything but condescending in person.

Na.... my thoughts are naturally truncated at the source.   

Danny has a great person-to-person interface and it helps him tremendously.    On the other hand I've seen him talk down to people.  There is a reason people perceive it and I've heard it repeated over and over again.

I don't say this to be harmful.   It is blunt... and I think Danny should understand that is how he is coming across and it can be a big part of the reason for someone like John coming back at him with a Chihuahua comment.

Also I have a lot of respect for the work Danny has done with the High School Physics class.    It brought Danny up several notches on my personal evaluation chart.   

[ShinOBIWAN]

You're just not getting it. Please re-read my last post.

I could say the same about you. Its just me not agreeing with you, OK?

[Danny Richie]

I don't think most engineers are going to see casual listening test as a valid method of reliable data.    You make absolute statements about the audibility of various unmeasurable phenomena without any actual repeatable data to support it. 

   

Kevin I agree with what you said about this before in this thread. It takes a lot of time and money to properly document this stuff. It's not easy.

I have set up A/B and blind A/B comparisons of this stuff before with groups of listeners only for it to become a joke to them because the differences were too great and apparent. If there is no challenge, it's no fun and no one wants to play.

I have never had a person come here and listen to any A/B comparison of anything that has been debated as making a difference or not, and NOT have them be able to hear a difference.

Unfortunately a lot of systems out there aren't capable of resolving a lot of the subjective differences in some things and some people just don't get it because of that.

You on the other hand don't make the disclaimers as obvious.   It becomes an absolute statement of truth rather than a subjective preference.   When you do that about something that has zero physical data to prove it (the audibility of wire break-in) you open yourself up wide for people with technical backgrounds to call "snake-oil!".    John probably long ago placed you in the snake-oil category.    If you just cover your tracks better with a couple disclaimers concerning the subjective nature of your opinions you won't draw as much fire.

You may have a point to some of that, but your experiences aren't mine.

Here is the big difference Kevin. You and I don't agree on these things either, but that's okay. It has never kept us from getting along. I call you (and I have) and can call you if I want to and talk about anything. You call me (and have) and discuss anything too. We see each other at shows, we get along, we have no issues. I don't hate you because you don't agree with me or visa versa. We don't have targets on our backs for each other. I don't feel any need to get even with you or show you up. I have never seen you do any of that to me either. We are competitors with some things but still get along great.

One day I'd love to share our experiences together (listen to some of these subjective things) and I would bet that in the same situation we'd hear and notice the same things. My door is always open to anyone that wants to listen.

John on the other hand does not share this attitude, does he? There's a big difference!

By the way Kevin, you went way up on my chart when you adopted those kids. 

[ShinOBIWAN]

Let's flip back a few pages to Dan's comments about near field measurements and see if he is correct.

Danny, quit beating your chest.

On John's website the M130 is under this section:

Outdated driver tests:
Note: these are old an less accurate. Don't put too much faith in them.

And within that section, here's his M130 FR plot:



Your talking about nothing. There is no 1.9Khz dip in his measurements.

[Daygloworange]

Well if you don't know that then clearly your in the wrong place

I can guarantee you, we have different thoughts as to the reason as to the existence of his website.

Its just me not agreeing with you, OK?

You have a real knack for pointing out the obvious. The reasons why it doesn't make sense and resonate with you, speaks volumes.

See the bit where I talked about idiots.

I wouldn't be constantly reminding us of all the idiots out there, given some comments here. 

Cheers

[Danny Richie]

There is no 1.9Khz dip in his measurements.

You're right. Thanks for helping me make my point.

So why was it included in his comments? This is what he had to say bout the M-130.

GR M130 - coated paper cone, plastic frame w/6 holes, fake phase plug, good motor venting. Breakup node at 5.5kHz with energy storage problem. Strange suckout at 1900hz, not visible in gated or smoothed curves.

[ShinOBIWAN]

I can guarantee you, we have different thoughts as to the reason as to the existence of his website.

I questioned your motives before but now I have no doubt.

In case you were wondering, John measures drivers picks the best price vs. performance ratio and then uses them in his own designs which he makes available for us lot to do as we wish on the proviso that you get no support. The aftermath of that is the driver tests. I really know I'm banging my head against the wall with you so I won't say much more than that.

You have a real knack for pointing out the obvious. The reasons why it doesn't make sense and resonate with you, speaks volumes.

We're talking opinions here. I stated mine and you yours and then we both objected to each others. So it seems you missed the point and still think its 'obvious'.

I wouldn't be constantly reminding us of all the idiots out there, given some comments here. 

Yeah idiots, they're everywhere.

[ShinOBIWAN]

There is no 1.9Khz dip in his measurements.

You're right. Thanks for helping me make my point.

So why was it included in his comments? This is what he had to say bout the M-130.

GR M130 - coated paper cone, plastic frame w/6 holes, fake phase plug, good motor venting. Breakup node at 5.5kHz with energy storage problem. Strange suckout at 1900hz, not visible in gated or smoothed curves.

Re-read this part, from Johns site, again:

Outdated driver tests:
Note: these are old an less accurate. Don't put too much faith in them.

Your talking about something that is there for posterity. People are then ignoring this and making a whole book out of it. The measured data shows nothing except a well behaved driver.

[Kevin Haskins]

You may have a point to some of that, but your experiences aren't mine.

Here is the big difference Kevin. You and I don't agree on these things either, but that's okay. It has never kept us from getting along. I call you (and I have) and can call you if I want to and talk about anything. You call me (and have) and discuss anything too. We see each other at shows, we get along, we have no issues. I don't hate you because you don't agree with me or visa versa. We don't have targets on our backs for each other. I don't feel any need to get even with you or show you up. I have never seen you do any of that to me either. We are competitors with some things but still get along great.

One day I'd love to share our experiences together (listen to some of these subjective things) and I would bet that in the same situation we'd hear and notice the same things. My door is always open to anyone that wants to listen.

John on the other hand does not share this attitude, does he? There's a big difference!

I wouldn't get too snuggly wuggly.   I've been pissed at you ever since you had my amp in that review and didn't use my adaptor cable.     

Truth be told I've been enjoying seeing you in the hot seat.   That is the evil side of Kevin though... not the snuggly wuggly side.