Time coherence - how important and what speakers?

New Speakers?

Are you designing or upgrading any new bookshelfs?........

The Europa was upgraded many months ago, and I have been working on other designs, both smaller and larger. However, I feel uncomfortable using this forum for promotion, but I do appreciate your asking. Such info will be on the website, and anyone may email me or call and I'd be happy to give them an idea what is to come, prior to the site's publication.

But the first shipments of the new Continuum 3 (replaces the C-2) must come before finishing the site's last pages. The increased sales of existing models had slowed the C-3 production down, but we are almost there. I will post a notice here, if the moderator feels that is appropriate, when the website gets up and running.

For a couple of months, I had wanted to offer my summary of this "time coherence" debate, because while I had made some points about the effect on sound quality and the physical limits to "perfect" time-coherence in any design, it still seemed to be hard for some to visualize what a time coherent speaker did for the sound, for the "waveform".

I believe part of that comes from the way the electronic age has let us visualize sound- still struggling to somehow see it directly, like we can with a water wave.

We have filmed the movement of the ear drum- we know it moves in and out with local changes in air pressure. Is that a linear response- air pressure to mental response? No. But on what we each hear from a given stimulation, we will generally agree. We don't call that stimulation a "wave"- we call it a sound. Maybe that's mom, or Mozart, a large bus passing by, or something we've never heard before.

We know a mic diaphragm also moves in and out- we can see the corresponding voltage rise and fall on the `scope. The `scope face freezes for us a 10th of a second of the diaphragm's motion- and we see there's a wave pattern going up and down.!

But to the mind- all we know is the pressure on the body went up and down.

Time-coherent speaker design means preserving that sequence of pressure changes- the same sequence of pressure variations the mic turned into voltage variations.

Non-time coherent design means the "wave pattern" we see on the scope does not matter- we can present the eardrum with a new sequence of pressure variations, and expect somehow that this will sound the same, or nearly so.

Why do some think that pattern can be changed? When they claim, "Mathematically, and audibly, that is just a collection of particular tones. The ear doesn't care EXACTLY when they arrive- just as long as they do, sort of near their original sequence- say, within a couple of wavelengths or so (~720 degrees of phase shift)."

And there is the mistake. We know in our heads those are a bunch of individual tones- we can hear them, and point to their source. We know that mathematically as well, and can now see this via a computer's FFT.

They are heard as tones, seen as tones... they are NOT tones when impinging on the body- they are a series of apparently random pressure fluctuations imposed upon it.

For me there is no choice; a speaker must be capable of making that original sequence of pressure variations traverse your body. This is an event, a series of events- described by the frozen-in-time "envelope" or "wave packet" seen on a `scope. Our minds decode the tones out of that sequence. And when they occur. And how loud they are.

Most speaker designers throw out the "When" and use tests that ignore any distortions of "When".
A sinewave frequecy response test doesn't care "when".
An FFT throws out the "when".
A pink noise test ignores the "when".

I think keeping track of distortions in the time domain, the "when" of every moment-by-moment variation in pressure, is as important as preserving the pressure (loudness or amplitude) of any particular variation.

Speaker design is easy when you ignore the time domain, but once you hear the difference, you cannot ignore the importance of reproducing the "when". Perhaps one of the non-time-coherent speaker designers could tell us why they believe the "when" is unimportant.

But to do so, they will somehow ignore the most fundamental concept- that Sound is air pressure changing in time. That is all it is. Pressure vs. time, a particular pressure change at a particular moment.

I believe reproduction approaches "hi-fi" when any change in the air pressure next to our bodies as time flows is dictated more by the sound/by the music, than the loudspeaker.

Best regards,
Roy Johnson
Green Mountain Audio

NSM, Role Audio, and maybe Jordan single driver speakers are time and phase coherent.
NSM states "Time coherent" at their website here:
NSM

Any comments Roy?

Cdc, NSM is blowing smoke up your ... There is no way on heaven or earth that the speaker shown in your link is time coherent. As I posted early on in this thread, very few speakers are actually time coherent, and I've never ever seen one with drivers mounted to a flat vertical baffle that comes even close. Sure, many manufacturers would love to have you believe it, but one look at the step response is all you need to tell otherwise.

Roy, great summary of philosophy, which I agree with 100%.

Meadowlark has a simple and easy-to-visualize example of this on their website, www.meadowlarkaudio.com.

I've said straight out what I find wrong with the sound of higher-order crossovers and the philosophy of that sort of design approach, but have not critiqued the finer points of any design. That holds here, but I do appreciate Cdc's wanting to know more. I have commented on Audio Asylum about the Jordan approach, and can add no more here. I can say that at least NSM is implementing many good things! Karls is right though about the flat baffle, but not exactly because it is flat and also perpendicular to the floor.

Remember that any speaker design which permits cabinet reflections reduces much of its potential, including any attempt at time-coherence. Reflections always cause any crossover design to be "fudged" and the driver choices to be less than optimal.

In first-order crossover speakers, all those lead to a low speaker height (tweeter too close), because that's the only way to produce a decent tone balance from what is still a twisted phase response. I will not present the math for that here (couldn't even begin to), nor would I want to give it away.

That height is also too low to offset the flat-baffle/tweeter-too-close placement, which is Karls' point. One also measures a greater +/- dB response variation through the crossover region, even when sitting at that "proper" height.

Finally, the listener is often encouraged to not toe-in these speakers, so less reflections are heard in that best seat. But this reduces the strength of the center-fill, so the speakers are moved closer together. All of this makes the head position even more twitchy.

It is the reflections that steer a designer to all these choices, and away from true time coherency. But if you read the NSM and Jordan-designs reviews, you can see what coming closer to time coherence accomplishes, including clarity, dynamics, ease of placement, freedom to play all music, and to use any component, especially when the 1st-order crossover circuit employed is simple and made with high quality parts.

I am aware of NSM's/Leo Massi's/Audio Physics/Michael Green's room placement guidelines, and I do not recommend those for any speaker. Ours are closer to Cardas', as are most firms, and described in our manuals.

Got a link for you to try:
Doo Wop Horses
If the sound "clicks", erase your browser's caches and re-visit when there's less web traffic. Takes about 30 seconds to load thru a 56k modem. Let it load completely (wait for the fence to appear), then click on each horse. Whoever wrote this is quite accomplished!
Makes you remember why we want fidelity!

Best,
Roy
Green Mountain Audio