[Jack-Devel] How does --hwmix work?

[Chris Caudle]

On Wed, May 3, 2017 1:05 am, David Kastrup wrote:
> Sigh, you are right.  But the question remains:  What applications will
> support this and what does it mean?

If you are recording something you presumably want or need to hear what
you are recording.  There are three ways to do this:

You connect what you are recording (microphone, keyboard, etc.) to a
hardware mixer, that mixer sends the audio to the computer audio interface
and also to the speakers.  You may need to also take output from the
computer audio interface and mix with what you are recording before
sending to the speakers (for example you are adding tracks to some already
recorded material and need to hear both the existing material and what you
are playing).

The second method is that what you are recording goes only to the computer
audio interface, the computer does any mixing required with pre-recorded
material, and the output from the computer audio device goes to the
speakers for  you to listen to.  The disadvantage is that the audio you
hear is delayed by two conversion paths (into the computer and out of the
computer) plus whatever time is taken by the software actually processing
the audio.

The third method is what is meant by hwmon, the audio you are recording
goes to the audio interface, but rather than being transferred into the
computer and requiring the software to process and send back to the
output, the audio interface hardware sends the input signal directly to
the output, optionally mixing with audio being sent from the computer
software at the same time.  That can be done at the analog stage or the
digital stage.  For example, I have an interface where the analog output
can be directly from the DAC (converting the digital audio sent from the
computer), or can be directly from the input circuitry, or  you can mix
between, all controlled with a knob and implemented with analog circuitry.
Your HDSP device can do something similar, but done after the analog to
digital conversion, so that you can control the level and mix with
software instead of using a knob on the interface hardware.

Where this makes a difference is when you are using a recording
application such as Ardour, if you configure Ardour so that it uses
software monitoring, Ardour will mix the incoming audio with the existing
material and send back out to the audio output.  If you configure for
hardware monitoring, then Ardour will only play back the existing
material, and the hardware is responsible for mixing the incoming audio
material with the output of pre-existing material so you can hear both.
Note that both method 1 and method 3 would be classifired as hardware
monitoring, it only differs in whether you have an external mixer handling
the task or you computer audio interface can handle the task.

-- 
Chris Caudle