Linkwitz Orions beaten by Behringer.... what!!?

The lack of any plain-and-simple definition and explanation is both confounding, and off-putting . . .

Confounding, yes, but this shouldn't "put you off". There is no deffinition to this stuff. When I first heard the term I thought it was just another meaningless Audiofool phrase, but now I take it more seriously. But I still can't define it subjectively or objectively. That doesn't mean that it isn't a real attribute. The more it is discussed the more likely some convergence on a rational deffinition and objective metric are likely to happen. I've learned a lot by this discussion. I think that idea that the speaker/room interface is a factor is intriguing because I hadn't thought of that. I am still not convince that thermal isn't an aspect. I think that Tom is right on the money that a MTF type of measurement might correlate to "dynamics". Is any of this clear? Yea, clear as mud, just like every other endevour starts out.
 
Very interesting - you understand the perceptual issues of distortion. I would not have expected this much difference.

I think the difference in the measured distorsion (in post http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-120.html#post3394911) comes from the differences in directivity between the speakers.

The used STEPS software works with a continous signal, so reflections are not gated out. As the frequency goes up, the directivity of the JBL 2445 goes up => less measured distorsion.
 

ra7

Member
Joined 2009
Paid Member
I think the difference in the measured distorsion (in post http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-120.html#post3394911) comes from the differences in directivity between the speakers.

The used STEPS software works with a continous signal, so reflections are not gated out. As the frequency goes up, the directivity of the JBL 2445 goes up => less measured distorsion.

Interesting. I knew that STEPS was including the reflections. Is this another way of looking at the MTF that Tom has been talking about? Does this result show something useful that correlates with the perception of greater dynamics from the horn/JBL combo?

Edit: I would also point out that while the reflections caused some peaks and dips in the frequency response, and the smoothing evened out this response, the B&W was audibly distorting when producing 100db (106 db at the mic at 1m) at 2m. My physical response of wincing was much more pronounced while running the test for the B&W than for the horn, which was much more smoother and composed.
 
Last edited:
Interesting. I knew that STEPS was including the reflections. Is this another way of looking at the MTF that Tom has been talking about? Does this result show something useful that correlates with the perception of greater dynamics from the horn/JBL combo?

I still have to read the references Tom posted, but I do not think these are comparable. I would expect however that repeating your measurement outside would give a better comparison of the distorsion characteristics of both speakers...
 
I think the difference in the measured distorsion (in post http://www.diyaudio.com/forums/mult...ns-beaten-behringer-what-120.html#post3394911) comes from the differences in directivity between the speakers.

The used STEPS software works with a continous signal, so reflections are not gated out. As the frequency goes up, the directivity of the JBL 2445 goes up => less measured distorsion.

Expanding on the idea of "it's the indirect sound field, stupid", here's another hypothesis: distortion becomes more audible when D/R decreases.
 
I was thinking along the lines of this:


Just take a look at Zwicker book cover, no need even to actually open the book :D

You'll see the masking threshold gets more horisontal as level increases. The harmonics are inhibited. Fundamentally you cannot hear something below the threshold.

So it can be said room reflections decrease the perception of nonlinear distortion.

attachment.php



- Elias
 

Attachments

  • 51UGn1lHyKL._BO2,204,203,200_PIsitb-sticker-arrow-click,TopRight,35,-76_AA300_SH20_OU15_.jpg
    51UGn1lHyKL._BO2,204,203,200_PIsitb-sticker-arrow-click,TopRight,35,-76_AA300_SH20_OU15_.jpg
    18.2 KB · Views: 444
Just take a look at Zwicker book cover, no need even to actually open the book :D

You'll see the masking threshold gets more horisontal as level increases. The harmonics are inhibited. Fundamentally you cannot hear something below the threshold.

So it can be said room reflections decrease the perception of nonlinear distortion.

attachment.php



- Elias

Let's open the book: how is the masking threshold measured? What type of signal under what conditions?
 

ra7

Member
Joined 2009
Paid Member
Right, reflections make the main signal seem louder. And louder signals mask better. But the brain is expecting to hear the harmonics in a voice, say. And if the reflections mask the harmonics, shouldn't that be perceived as distortion? Isn't the picture less clear to the brain in this case? And if by directivity, the harmonics can be more easily detected, i.e., the reflected field is less louder, then shouldn't that be perceived as less distorted?
 
Just take a look at Zwicker book cover, no need even to actually open the book :D

You'll see the masking threshold gets more horisontal as level increases. The harmonics are inhibited. Fundamentally you cannot hear something below the threshold.

So it can be said room reflections decrease the perception of nonlinear distortion.


- Elias

From what I have read about reflections:

- The information in the direct sound and the reflections is combined
- It’s incorrect to assume that the precedence effect is some sort of masking phenomenon, which by blocking out later reflections, prevents us from being confused
- The loudness of the perceived sound is increased by the later arriving reflections
- The apperent position of the source comes from the first arriving sound
 
If what some of you are calling dynamic is the direct to reflective ratio I think we have a problem here. If more direct sound on axis becomes the definition than why don't I or Earl make a horn that has something like a waveguide with say a 20 degree dispersion pattern and shove everything at you on axis and almost eliminate any reflective waves except for a first reflection from behind your head. Yes I have taken this to an extreme, but that was to make a point that it is not the ratio of D to R that is giving the dynamics in my eyes. Yes I think that it is part of the equation, but it is more than just those two factors. The same with Tom's posting about light waves and diffraction doesn't seem to correlate here in a very logical way. I think that you have to take phase and group delay into the picture, those two functions would be as important as the on axis frequency response. If the phase and group delay produced by the speaker are shifting across the band how can you discount those things in the dynamic reproduction of the original waveform? Yes, I think that there needs to be a common definition so that we can all speak the same language, otherwise we are all speaking gibberish to each other and none of use can understand and communicate clearly about the subject. Just my two cents of thought on the matter today.
 
Administrator
Joined 2004
Paid Member
Yikes! Super busy here today. Some real nice discussion going on.

RA7, thanks for the measurements. It's easy to see the differences - but which speaker do you think sounds more dynamic? Or are they basically equal? I've often wondered if a dynamic passage were recorded from 2 different speakers would they actually show differences in amplitude? I mean average to peak amplitude.

OT, and on the subject of recorded pianos, years ago I used to do a lot of music and sound FX for theater on my trusty Revox decks. For one play I recorded a selection of piano pieces by Schubert. I used a pair of old Schoeps tube mics I had gotten from the ORTF. One mic near the piano (dry) and one at the other end of the studio (wet). During playback I could fade from dry to wet, then out, instead of just a simple volume reduction fade out. It made a lovely transition, moving the piano away as it faded. I was quite pleased with the effect. Nobody, not one person, ever noticed. :(
 
From what I have read about reflections:

- The information in the direct sound and the reflections is combined
- It’s incorrect to assume that the precedence effect is some sort of masking phenomenon, which by blocking out later reflections, prevents us from being confused
- The loudness of the perceived sound is increased by the later arriving reflections
- The apperent position of the source comes from the first arriving sound

Mostly right, but your first tick is only true for the first 5-10 ms or so for most frequencies. Your second point is quite true and often not understood. Your last point might be misleading since this "generally" true, but not always, and your second point should also be considered here because while the location is suggested by precedence, the "blur" of this location is highly complex.
 
If more direct sound on axis becomes the definition than why don't I or Earl make a horn that has something like a waveguide with say a 20 degree dispersion pattern and shove everything at you on axis and almost eliminate any reflective waves except for a first reflection from behind your head.

If it was feasible I would make the directivity narrower. But having studied this in some detail, there is a sweet spot at 45 degrees (90 degrees inclusive). Less gets problematic in one aspect and more gets problematic in another.
 
Hi Guys
I wanted to be clear, I wasn’t suggesting that the MTF is the be all and end all acoustic measurement but that those with more time and there is certainly interest, should look into it.

First, the problem they had determining intelligibility for large scale sound should sound very familiar, the result could part the microphone, it’s preamp, part the electronics, part the loudspeaker transducers, part or strongly the system directivity and mounting in the room relative to the people and probably more stuff I didn’t think of.
In other words, like a stereo, it’s a bunch of things and many of which you have no control over individually and may confuse the issue.

So, over many steps they arrived at a “measuring” STIpa which does a good job predicting the intelligibility based on a special test signal fed to the speaker system. This way, one can measure the loudspeaker and room portion of the sound system without any of the source related stuff affecting the results.
Now, since we are talking about “how much” information remains within the audio one is hearing at the listening position that makes it seem interesting to me from a stereo / signal preservation point of view.

Since the root of this kind of measurement is based on measuring what remains of the rapid amplitude modulation of signal ,the loudspeaker (plus the room / directivity etc etc) the MTF and you guys are hot on dynamics, it seemed logical to throw out the observations I had made.
Wonder why we don’t “see everything” in measurements?, we don’t measure everything, things like preserving / delivering amplitude modulation for example .

As Earl suggested this would be more of a comparative indicator than a diagnostic tool because it in no way tells you why you get the measurement you did or what to do about it. It’s not everything, but it is i think something, because our hearing system acuity seems to be roughly centered around the voice range.

Dewardh, in a faithful reproducer, one (to me) has a goal more like a process lens or apo telescope, one does not want any bokeh or chromatic, spherical aberration, the highest possible MTF, smallest possible airy disc.
Like I explained above for sound, the MTF measurements the STIpa is based on, are measurements of how much of the signal dynamic remains at the listening position which should be interesting since we are talking about dynamics and we don’t normally measure this domain.
Yes, it was developed for something else, but that doesn’t mean there may be something to be learned / of value in this area.
Anything that helps progress the state of our art or understanding of what we love is a good thing, have at it guys, back to the thread already in progress.
Best,
Tom Danley
 
Hi Earl
Well I can describe what I saw at the seminar but “what that is” in the processing world would be much more your domain than mine.

The fellow who designed the STIpa measurement described that they used 7 bands between 125 and 8KHz.
These bands had a gap between each of them so they did not overlap (was possible to deal with each individually).
The amplitude of the signal in these bands was amplitude modulated, my impression from the graphic is was an “on off” Morse code style modulation, not modulation with a sine signal etc.
In that way, it would be like the black / white bands in the optical world. The rate of modulation is increased (if I remember they stopped at 12 or 15 Hz) . The degradation usually happened at the high rates first and lowest rates last, also just like in optics.

In the examples of reduced MTF, what they showed was the “off” periods had sound, it was (more or less) the reduction in the peak to trough that showed a reduced mtf at that frequency, the cumulative effect of the bands ( weighted to a speech spectrum) turned into the STIpa..

During two of the breaks, I pressed the fellow about how high a modulation rate they considered and so on. He said they had measured up to around 30Hz rate but that range wasn’t needed for speech intelligibility. I asked a number of questions about localization etc and he said "well maybe" but was clearly focused on voice stuff.

I have found a couple video’s (product demonstrations so ignore that) which ever so lightly discuss the STIpa process but more importantly you can hear the test signal, download it from the link they mention (and analyze it?)

STIPA.mov - YouTube

STIPA-Basic-Overview.mov - YouTube

Also, ARTA has an “MTF measurement” which is done at a number of modulation rates but I am not sure how they do it.
Hopefully this makes some sense.
Best,
Tom
ugh, more snow coming