advertisement


Speaker/Room Measurement Witchcraftery

Thanks, and for the link. I was reluctant to make assumptions but that is what I thought you mean’t and I agree.

Indeed, it is what makes Martin Logan’s a challenge if more than one listener. However badly some perceive them and their measurements they have given me much listening pleasure over the years, and their entry model is lower in price than Quads and particularly MBLs. That Martin Logan are a long established company may indicate that I am not alone in enjoying their sound. Oddly enough their variable off axis response means that they can be tuned for taste and done right can give exceptionally realistic sound, albeit with a narrow sweet spot. So many things are a compromise and pushing the envelope in one direction can mean limitations in others.

In the end it is knowing what works for us as individuals and measurements can help in that quest but nothing can substitute for listening in our own unique living space.
Dipoles have a ragged off-axis but there isn’t a great deal of it , I seldom see them used with wide spectrum absorption behind the speakers, which I would expect given that they output as much from the rear as from the front, presumably listeners enjoy the reflected output?

Keith
 
Quite so Ayya, I was thinking of car analogies and it is bit like expecting a Ferrari to perform well in town or on rutted country lanes whilst ignoring the qualities when used as intended. I personally find DSP very useful in getting the best out of speakers but it is not a complete panacea. I suspect mix and match is the most practical depending on one’s circumstances.
 
Dipoles have a ragged off-axis but there isn’t a great deal of it , I seldom see them used with wide spectrum absorption behind the speakers, which I would expect given that they output as much from the rear as from the front, presumably listeners enjoy the reflected output?

Keith
Yes, I think we do rather like the contribution of the rear radiation! I suspect different owners employ different degrees of damping. My preference FWIW is heavy curtaining, enough to reduce smearing but not negate that wonderful sense of airiness in which the sound does not appear to be coming from loudspeakers. I think it is a case of balancing to individual taste. As is often pointed out Quads are without par for listening to string quartets; it is such a shame for mankind that such extraordinary music isn’t better known or loved ;) :cool: :D. These emoticon things are starting to look like my Chord DAC!
 
Last edited:
I'm not familiar with Toole's research. In a nutshell, assuming that the listening position is perfectly on-axis, what are the consequences of an uneven off-axis response vs an even off-axis response?
 
I'm not familiar with Toole's research. In a nutshell, assuming that the listening position is perfectly on-axis, what are the consequences of an uneven off-axis response vs an even off-axis response?
To simplify somewhat, humans perceive direct sound and short-delay ("early") reflections from nearby room boundaries as combined into a single sound. Thus you hear a composite average frequency response in a typical room from direct sound and off-axis early reflections.

At a crossover frequency the bigger (LF) driver typically has a narrow "beam" (basic physics) so the level of off-axis reflections may be well down. And the smaller (HF) driver typically has a wide "beam" so the level of off-axis reflections may be well maintained. You may therefore perceive an unwanted step in frequency response across a crossover frequency.

A good loudspeaker will match driver "beam" widths at crossover frequencies (e.g. with a waveguide on the HF driver) to make sure the off-axis frequency response has no discontinuity to smooth out the perceived composite response.

This has been known for some time. Well before Toole assembled the key measurements into his single spin-o-rama test suite.
 
Also Linkwitz posited all early reflections are undesirable (the direct sound and early reflection combination suffers from time smear).

He therefore recommends speakers should be at least 1m from side walls. Above this the reflections are processed differently - but he agrees off-axis response is key there too.
 
Also, also ... Toole says off-axis response up to 10kHz is what's important, since above that boundaries tend to absorb anyway. (Just need to worry about angles for toe-in relative to LPs above 10kHz.)
 
There is nothing wrong with your measurements, you did a good job. They do exactly what I would expect for a room with those dimensions and I have measured and listened to many similar rooms, including my own.

OK, problem with that definition is that Im afraid its your speakers that have the poor... correction, no its not poor per-se, its actually very typical..... time alignment issues, something which the Kii doesnt. So you need to look for a different explanation

Your time alignment - sharp tweeter impulse followed b y wider mid/bass, followed by wider sub. So a low frequency fundamental having harmonics that play through the mid will be about a 1.5 mS early and through the tweeter 3 ms early.

imp2_zpsqdnphgb0.png

Any chance you can label the parts that correspond to the tweeter, mid and bass?

As far as I can see the tweeter bit is obvious, however:
1) Mid is wired out of phase (does the mid start, inverted, at 0.2mS?)
2) Where does the bass start?
3) What is the peak at ~3.3mS?
 
This has been known for some time. Well before Toole assembled the key measurements into his single spin-o-rama test suite.

It sure has been known for some time. Here’s a bit from ATCs brochure ..

“4. Dispersion and Directivity
The relationship between direct and reverberant sound is very important in high performance loudspeakers. It is clear that not only must the on-axis magnitude response be accurate and linear but also that the behavior off-axis must be both broad and even with frequency exhibiting no abrupt dips in amplitude.The aim should be to achieve a horizontal dispersion of +/-80 deg. With a -6dB @ 10KHz and a vertical dispersion of at least +/-10 deg.To ensure that in a well behaved room with a good RT vs frequency characteristic,The reverberant sound will be consistent with the direct sound in the listening area.
To achieve this criteria the highest performance loudspeakers will be either small two way systems with bass/mid drive units up to 160mm diameter or preferably three way designs where the midrange is no more than 75mm diameter and crosses over from the bass drive unit at around 300 Hz. The tweeter in this system should not be greater in diameter than 34mm and should be crossed over from the midrange at around 3 KHz.
In a well behaved room when listening to a stereo pair of loudspeakers with a good relationship between direct and reverbant responses you will first hear the direct sound and then the reverberant field. It is generally agreed and probably true that the reverberant field masks periodic signals, however, it is also apparent that the ear has a precedence effect which means that for impulsive sounds the ear can hear phase dependent effects. Therefore, I believe that any critical judgment of repro- duced sound is made principally on the first arrival or direct sound which gives most of the phase related cues and also the low level detail which is quickly lost in the reverberant field.
However, the way we perceive magnitude band balance and the full energy of percussive or impulsive sounds, is dependent upon the power response of the loudspeaker or how evenly it excites the reverberant field with frequency.
Clearly it is impossible to exclude from such a relationship the effect of room acoustics, however, for the purpose of discussing loudspeaker performance we will assume that the listening room has been properly treated and has no serious intrusive acoustic problems.
It should also be stated here that the use of DSP to equalize loudspeaker room interface problems is not an acceptable solution to that problem in critical listening environments if it involves modifying the direct sound from the loudspeaker. A dramatic effect of poor midrange dispersion, common in many two way loudspeaker systems, is demonstrated by recording engineers making incorrect magnitude band judgments and applying equalization, usually to the upper midrange, in an attempt to compensate for the apparent lack of energy in that region. Many examples of pop recordings are available which demonstrate this characteristic. That is, a hard strident upper midrange which masks high frequencies, and makes vocals sound recessed while accentuating the bass.”
 
Any chance you can label the parts that correspond to the tweeter, mid and bass?

As far as I can see the tweeter bit is obvious, however:
1) Mid is wired out of phase (does the mid start, inverted, at 0.2mS?)
2) Where does the bass start?
3) What is the peak at ~3.3mS?

Just left to right. M pk at 1.4 and bass at 3. Not sure what the 3.3 ms Peak is
 
If I have interpreted the above linked Toole/Harman paper correctly, an anechoic off-axis response that closely mimics the anechoic on-axis response leads to an in-room response that closely mimics the anechoic on-axis response (above the Schroeder frequency at least). Which would therefore imply that, for such speakers, the subjective "smearing" or "lack of focus" you often get from untreated horizontal (side wall) first reflections will have little effect on the in-room frequency response at the on-axis listening position, is that a reasonable conclusion? If so, then where does all the comb-filtering at higher frequencies come from when you make in-room measurements? First reflections in the two other axes? Secondary reflections?
 
If I have interpreted the above linked Toole/Harman paper correctly, an anechoic off-axis response that closely mimics the anechoic on-axis response leads to an in-room response that closely mimics the anechoic on-axis response (above the Schroeder frequency at least). Which would therefore imply that, for such speakers, the subjective "smearing" or "lack of focus" you often get from untreated horizontal (side wall) first reflections will have little effect on the in-room frequency response at the on-axis listening position, is that a reasonable conclusion? If so, then where does all the comb-filtering at higher frequencies come from when you make in-room measurements? First reflections in the two other axes? Secondary reflections?
the side wall early reflections even with a speaker with well behaved off axis are still detrimental. the distance between the speakers and the side wall and your listening position, even if it would be 100% symetrical would still be detrimental. and then you have the floor, ceiling and back wall ( behind the listening position) early reflections.

not that big of a deal, a 2x4 early reflection panel at the ceiling, floor, side and back wall can be all it takes for a free reflective zone
 
Just left to right. M pk at 1.4 and bass at 3. Not sure what the 3.3 ms Peak is

The tweeter to mid is clearly an aspect of the design, due to the mid being connected in opposite polarity to the tweeter. But I don't understand where the delay between the tweeter and the bass comes from.

I reckon the 3.3mS hump is the bass reflecting off the front wall. The drivers are 60cm from the wall , so the path difference is 120cm, which is 3.5S (at 343m/S).
The spike at 3.4mS is most likely the tweeter output bouncing off the side walls.

The picture is confused by the mid being opposite polarity, so the plots are never going to look right. And of course this design can never be truely time-coherent, however the bass is rather more time-coherent that you seem to be claiming. JA reckons this time-incoherence is not generaly audible:
https://www.stereophile.com/content/measuring-loudspeakers-part-two-page-3

However what is known to be audible is the "envelope" distortion caused by poor transient response and that is where my ZRBs really score.
So apologies if I confused time coherence and envelope distortion in my earlier posts.

FWIW, a pair of Quad 989s set up 2/3rds of the way down a longish room sounds superb and IMO does not audibly suffer envelope distortion like most boxed (especially ported) speakers do.
 
However what is known to be audible is the "envelope" distortion caused by poor transient response and that is where my ZRBs really score.
What is envelope distortion? And do you perhaps have a link to the known to be audible part?
 
Ditto. What is envelope distortion.

The delays are a combination of different physical locations of the driver acoustic centres and the phase shift created by the crossover.

I'm not claiming anything, the time alignment is in black and white.

Time coherence will be less of an issue the lower in frequency you go but the difference between aligned and non aligned is audible. I thought you were previously concerned about time alignment?

So, your bass is out of phase with your mid. You have a significant dip at what looks close to a crossover frequency. Have you tried putting them in phase and seeing the effect on the dip?
 
Setting aside the discussion on envelope distortion for a moment, I would welcome comments on the Impulse response below:

Impulse-Responce.jpg


Please could the experts tell me how this looks for time coherence and where the various drivers contribute.
 


advertisement


Back
Top