AudioBanter.com - View Single Post

In article writes:

Not quite correct, group delay just defines about the same as time shift.
For a sine wave of a specific frequency this time shift results in a phase
shift, for other frequencies the same time shift result in different phase
shifts. If this time shift or group delay depends on frequency, the effect
is called 'dispersion'. A dispersion such that all frequencies have the
same phase shift is quite a strange phenomen, for this the time shift for
20Hz should be 1000 times the time shift for 20kHz

Don't you realize how bogus this discussion is? Phase is always
relative, but you have to define what the reference is. Let's make
this as simple as we can. You have a complex waveform that's periodic
and contains a fundamental and its second and third harmonic. If they
all start at the same time, we can say they're in phase. If we apply a
delay to this waveform, it's shifted in time, so it's phase relative
to something else hasn't changed, but the frequencies comprising the
waveform are still in phase.

Now, if we pass it through a phase shift network (an all-pass filter),
each of the harmonics will be delayed by a different amount. There are
no longer in phase. Something about the complex waveform will change.

Now if we have a stereo pair of complex waveforms that are different
on each channel but have some relationship none the less - they both
came from the same source - if we delay one with respect to the other,
we'll have a whole bunch of different phase shifts, more than we can
reasonably measure, because they're going to change before we can
write the numbers down. I believe that when someone says "I've shifted
the left channel 90 degrees from the right channel" nearly all the
time, they'll mean that they've created that phase shift at a single
frequency, typically near mid-band, with some sort of a delay.

So first we have to agree on a language,

--
I'm really Mike Rivers )
However, until the spam goes away or Hell freezes over,
lots of IP addresses are blocked from this system. If
you e-mail me and it bounces, use your secret decoder ring
and reach me he double-m-eleven-double-zero at yahoo