Re: [linux-audio-user] hardware mixing - what it _actually_ is?

On Mon, Jan 22, 2007 at 04:58:13PM -0800, Mark Knecht wrote:
> I often mix almost everything that's live at the time in at the
> card level, including vocals. Vocals come in dry and go out
> immediately to a hardware reverb to get a wet mix for headphones. The
> dry version goes on to the system to be recorded, whereas the wet
> version is what the singer hears. This allows me to use any reverb
> setting they are happy with but still preserve the dry vocal for later
> mixing.

Heh. Vocal monitor reverb is a bit of a special case... Even having no
hardware reverb here, I can get a similar effect by:

* Using hardware monitoring for the dry vocal.
* Adding a send from the vocal track(s) in Ardour to a bus
* Adding 100% wet reverb to the bus.
* Mixing a bit of the bus output to the hardware output feeding the
  headphones.

This works because reverb is fairly unique in that a bit of predelay
on the effect is actually desirable, and buffer latency is as good a
source of predelay as any :-)

I haven't tried it with a very large jackd buffer though.

> I do not really consider the above scenario 'hardware monitoring'
> although I can see how others might, since I use the hardware mixer to
> insert the vocals into the stream being played by the computer. Zero
> latency, happy singers, etc.

Shrug. Sounds like the definition of hardware monitoring to me.

-- 
Paul Winkler
http://www.slinkp.com