Hi Roger and group,
I think Stefan has already answered most of the questions.
Indeed you can (should) use the same instance to lock the sampling
rate to a reference, and produce a stable output signal.
But there's a catch:
Even if the input (A/D converter) and output (D/A converter) are in
the same device ("soundcard"), their sampe rates may be slightly
different even if both are configured for the same nominal SR (like
44100 or -often better- 48000 samples/second). This doesn't seem to
be a common problem, but I have experienced it myself in one of my
PCs. One would need a control loop (soundcard's output fed back into
the input) to correct the output sampling rate, but this would make
things overly complicated and thus it isn't implemented in Spectrum
Lab (at least not directly). Also, if input and output sampling rate
are "slightly different", it becomes a nightmare to manage the audio
buffers so there is neither a sporadic input-overflow or
output-underflow. Both cases of overflow cause an audible phase
glitch in the output.
Much easier and more reliable to build a dedicated hardware, clocked
by a 10 MHz reference, using a PLL, DDS, or similar.
All the best,
Wolf .
Am 25.04.2012 11:38, schrieb Roger Lapthorn:
|
|
