Linux PA WAS [linux-audio-user] More Mandrake 9.2 fine tuning...Dell latitude live fx box

Russell Hanaghan hanaghan at starband.net
Wed Feb 11 12:14:01 EST 2004


On Tue, 2004-02-10 at 07:23, Greg Reddin wrote:
> --- Russell Hanaghan <hanaghan at starband.net> wrote:
> > THis
> > Saturday evening will be the test at the good ol' "Lucky Dawg"
> > saloon!
> > Yup, you read it right...The "Lucky Dawg"!! Reeks of Banjo music
> > dont?
> > :) But it's a fun low pressure gig. 
> > 
> 
> Keep us posted on how this goes.  I've had visions of using my Linux
> box as a live processor for months mow.  Most of the gigs I do are
> low pressure so I'd be a good guinea pig for it :-)

So far, I have managed to run Jack and Ardour utilizing the swh set of
LADSPA plugs. Primarily, my need is for an auxiliary path only using
reverbs and eq for vocal and instrument ambiance on the PA.I have jack
running with minimal Xruns at 1024, 2, Realtime on my Latitude CPXj 650,
256 megs ram, 20g. Latitude runs an ESS Maestro3 PCI sound chip. Latency
is there but not really noticeable for this application of realtime
processing. I'm using MDK 9.2 with the Multimedia kernel and as many
daemons turned of as I can. I use X and IceWM for GUI.

Had a rehearsal last night and I was very impressed!! It stayed stable
and I only used "Freeverb" but it is a nice warm reverb and the
flexibility is there. Compared to Alesis units or the crappy Zoom multi
fx digital pedal I've been using it was surprising. Well worth the
effort and this is minimal use but no time right now to experiment more.

> 
> So to fork the topic completely, here's my vision.
> 
> Phase 1:  Use a Linux DAW for EQ and dynamics processing for the PA.
> Phase 2:  Add a multi-effects processor to use from an aux or group
> on the board.
> Phase 3:  Use a midi controller and ditch the analog board.  Linux +
> controllers + amps becomes the PA system.
> Phase 4:  Add software to help analyze a room and set up EQs, etc.

WHOA!!! Are we recreating HAL here?? 8~}

Obviously the lack of ins/outs here is the first issue...unless using
RME or Echo Layla or similar and then only like, 8 inputs...On standard
cards the headroom specs might be an issue on inputs before distortion.
I don't have a clue what I'm running at (Prolly close to -10 dbm unity
gain) but no real clue...just watching the VU's in Ardour for that and
listening for distortion. On the cheap side, I have a CII docking
station for the laptop which has 2 PCI slots. I have a SB Live in it.
Now, multiple cards dont work when trying to sync for rec/playback...but
for this it wouldn't matter. That would give me at least 2 stereo ins
and 4 stereo outs to play with. Should be able to just add emu10k to
.asoundrc and fly it next to the Maestro 3. Hopefully the routing can be
done in Jack. I had thought of this to process a single mic insert for
my female vocalist. Her voice has mucho hi mids and a sidechain or
similar setup would be ideal to reduces those pesky earshattering freqs.
DOnt know how the processing power will go on this rig. Maybe tonight
I'll try it.
> 
> Here's some use cases for an ideal world after all this is set up.
> 
> 1.  One sound engineer mixes house and monitors for a club band. 
> He's up on stage listening to the monitor mix and adjusting it with a
> PDA over a wireless network :-)

Love this idea!!! 
> 
> 2.  Two separate engineers for house and monitors.  Everything is
> routed through one computer (or a cluster) to multiple controllers
> (as opposed to analog splitters they now use).  The monitor engineer
> has a controller and builds the monitor mix.  The FOH engineer has a
> controller to manage the FOH mix.
> 
> 3.  Multiple computer setup.  Headless processor that takes all
> inputs and outputs and does all processors.  Laptop controllers that
> do the GUIs.  For that matter, the processor box(es) wouldn't even
> need to be running X.  It could be controlled via networking. 
> Everything is fault-tolerant.
> 
> 4.  Installations.  Render a 3D design of the room.  Use this
> information to spec out what system is needed.  When the system is
> installed, set up mics around the room and use the analysis software
> to set up EQs and processing.

This had occured to me in simpler form. A basic spec. analyser should be
easy but the soundcards freq. effect and response would need to be known
for a baseline reference I would think. My ears have always been a
better tool than any spec analysis I've used but that aside...this might
be an easy one not including the 3d rendered room. That will take some
CPU juice me thinks??
> 
> Ok, so my dreams are waaay bigger than what is possible right now,
> but I think the master processing and effects processing are doable. 
> Once you get into that you really start wanting individual channels
> mixed in the computer, but you also want tactile control over each
> channel.  We'll see where it goes.
> 
> Sorry for the diversion.
> Greg
> 
> __________________________________
> Do you Yahoo!?
> Yahoo! Finance: Get your refund fast by filing online.
> http://taxes.yahoo.com/filing.html



More information about the linux-audio-user mailing list