> Number 2, There is clearly a difference between the two recordings, so
> maybe
> someone in Tennessee has a tin ear?
I just ran the two audio clips through SpectraPlus FFT software. The .MP3
output is read directly by SpectraPlus internal to the sound card with no
audio cabling necessary.
The O2 has a sharp AM audio response cutoff at exactly 3 kHz (6 kHz of BW),
as expected. On the FFT display, it is clearly a very sharp DSP cut-off
with no detectable skirt. Audio response cuts off at 3 kHz and nothing
escapes beyond that point. The O2 low-end response extends to below 80 Hz,
but rolls off smoothly below 90 Hz.
On the TS-870, the AM response is also quite sharp, but cuts off at
approximately 3.7 kHz. What we are all hearing is the significance of the
speech intelligibility in the area between 3 and 4 kHz. That region does
more to add speech intelligibility than any other area of the audible
spectrum. This should come as no surprise when one looks at the
Fletcher-Munson family of loudness curves. At approximately 3.5 kHz, the
ear is most sensitive to absolute sound pressure level. Why? Because the
ear canal terminates into the tympanic membrane and forms a closed
acoustical pipe whose average resonant frequency is -- about 3.5 kHz. The
exact point of resonance is a function of the length and diameter of our
individual ear canals. So, a mere 700 Hz of added bandwidth makes a
significant difference in intelligibility and is one reason why ESSB
proponents have been advocating the use of extended-bandwidth SSB when
conditions permit.
Although things start sounding even better beyond 4 kHz, diminishing returns
sets in quickly. In the interest of allowing operators the use of maximum
speech intelligibility versus spectrum conservation, 4 kHz for SSB and 8 kHz
for AM is an excellent choice for maximum occupied bandwidth when operating
conditions allow it.
Paul, W9AC
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|