Shadome, thanks for your replies. Let me clarify some speaker-design information you may have been given in error. About the sound of MQA- I have spent little time with it.

The idea of linear phase in speakers is tricky to understand without going through the math, but the following definitions clarify the principle:

’Phase’ is always a relative term. Example- If two cars leave the same location 12 seconds apart and arrive at the same destination still 12 seconds apart, we can say their relative phase did not change.

Now, if those two cars leave side by side and arrive side by side, we can call that time-coherent behavior for want of a better descriptor. And both were still traveling ’in-phase’, ’phase-correct’ or ’phase-coherent’. Time-coherence automatically implies phase-coherence but not vice versa.

When measuring a speaker, we assign 0 degrees at the moment its first sound arrives at the mic. Most always, this is will be at a high frequency coming from its tweeter, before that tweeter goes into a breakup mode, at 10kHz or higher in a modern design.

Now, as we go down the scale looking for when each lower tone arrives, most every speaker lets those lower tones drift backwards in time more and more the lower down we go. If it does this smoothly, that is linear phase. If it does not drift backwards at all, that is time-coherent behavior.

Let the starting phase be set to 0 degrees up at 10kHz.Down at 7kHz, our software measures the phase as ’60 degrees Positive’.This tells us three things:

- The time delay at 7kHz is 60/360, one sixth, of the 360-degree period of a 7kHz cycle = 1/6 x 1/7000 second = 0.023msec. In space, this is equivalent to ~8mm behind the location of the 10kHz location. (0.023msec x 0.340meters/msec speed of sound). You could say ’it sounds like and measures like’ 7khz source is 8mm behind 10kHz source.

- ’Positive degrees’ means ’time delay’, because of how the (unseen) math equation was set up.
- On the ’scope, a 7kHz pulse (a ’beep’) of sinewaves from the tweeter moved to the left by 1/6 of a 7kHz wave cycle. It started late by that amount and thus ends late by that same amount.

Now, down at 3.5kHz, one octave lower, if the phase change is 120 degrees total (of 3.5kHz), then we say the rate of phase change is ’linear with frequency’ and ’60 degrees per octave’ if we like.

Such a speaker is linear-phase, because its designer has also taken into account the mechanical phase shifts of its drivers. Its rate of Phase Change is SMOOTH, not ’jerky’ (= ’non-linear’), which leads to a ’smooth transition between drivers’ as Stereophile’s John Atkinson writes.

Time coherence is a much tougher performance standard for speakers. It is possible, by the way, to correct or offset most all of the mechanical delay in the drivers.

When a speaker distorts in the time domain, this blurs imaging and details, and more important if one’s ears are open to it, ’how’ the music is being played and its emotional content.

I hope this is of some help! Let me know if I can make anything more clear.

Best regards,

Roy Johnson

Green Mountain Audio