OK, perhaps a stupid question: does jackd mix? In other words, let's
say we have two or more audio applications outputting to Jack; will Jack
mix the streams? Or does it send them each independently to the sound
driver?
J.E.B.
Hi folks.
On my quest to get more then two channels from Alsa out to the 9632
I stepped over this:
I figured that there is just a pcm0p under /proc/asound/card0.
Shouldn't there be pcm0p...pcm7p reflecting the 8 Alsa stereo channels on
the HDSP?
What am I mixing up here?
THX for your patience.
\Klaus
Hi,
Anyone know what happened to this website? There was a lot of good
info on this site about setting up multiple M-Audio Delta 1010 audio
interfaces on Linux for use with Jack, ALSA & Ardour.
Thanks...
--
Kevin
Well. Have no clue why, but after viewing about 90 minutes of Joost
video, my everything-audio-works configuration stopped working.
Automatic updates are turned entirely off, I'm thinking this must be a
HAL thing, not sure how to debug. Anyone know what I should try to
debug? As of this moment, using any of three configurations including
the one where everything goes direct to ALSA hardware by name, the only
audio I have is Jack :-)
J.E.B.
Hey list,
It's been a rough week (it's only Tuesday!), so I thought I would
record something sad. I wrote a little tango I'm calling Tangled.
Blogspam:
http://www.hardbop200.com/2009/10/20/tangled/
Actual file:
http://www.hardbop200.com/blog/wp-content/uploads/tangled.ogg
Just two pianos and a drummer. Recorded into Ardour on Debian
(Squeeze) using Hydrogen as a drummer. No plugins, no mixing, just
playing.
Honestly, the progression is nothing original, and I'm a little
embarrassed at my solo w/ players like Ken and Atte on the list, but
oh well. :)
Enjoy,
Josh
--
Josh Lawrence
http://www.hardbop200.com
Hi Georg and others.
Find enclosed my HDSPscan script. I wrote it to identify the channels ids
for using it with amixer.
It connects each in- to output, which makes it easy for you to identify the
routing you're looking for.
It is a two-stage approach to reduce the number of test cases.
First it will connect every single input - one after another - to "all
outputs" to identify the input you're looking for.
Then it'll connect the chosen input to every single output one after
another.
The test tone is generated by the script (sox) itself.
Hope some of you guys find it beneficial.
It saved me a hell lot of time.
Have fun
\Klaus
HDSPscan
------------------------------------------------------------------------------
#!/bin/bash
#
# written by Klaus Schulz 10/25/2009
#
# This script scans the matrix mixer of RME HDSP 9632 for input/output
connections by using amixer
# Your current setting will be removed. Use the script at your own risk.
#
# RME HDSP 9632 ( as per definition)
# inputs: 0-7 (analog), 16-23 (adat), 24-25 (spdif) 26-43 (playback
channels)
# output: 0-7 (analog), 16-23 (adat) 24-25 (spdif), 26-27 (line out)
track="/dev/shm/play.wav" # will be generated by the script!!!!!!!!!
alsainterface="plughw:0,0"
inputs_left=( 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 )
inputs_right=( 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43
)
let cnt_i=${#inputs_left[@]}-1
outputs_left=( 0 2 4 6 8 10 12 14 16 18 20 22 24 26 )
outputs_right=( 1 3 5 7 9 11 13 15 17 19 21 23 25 27 )
let cnt_o=${#outputs_left[@]}-1
function generate_track () {
which sox > /dev/null 2>&1 || { echo "Please install sox first!"; exit 1 ;
} ;
sox -r 44100 -b 16 -c 2 -n $track synth 2 sine 300-3300 gain -18 # gain in
db
}
function init_matrix () {
echo -n "Initializing hdsp-matrix..."
for output in $(seq 0 27); do
for input in $(seq 0 43); do
echo -n "."
amixer cset numid=5 $input,$output,0 > /dev/null 2>&1
done
done
echo
}
function scaninputs () {
for i in $(seq 0 $cnt_i) ; do
in_l=${inputs_left[$i]}
in_r=${inputs_right[$i]}
for y in $(seq 0 $cnt_o) ; do
out_l=${outputs_left[$y]}
out_r=${outputs_right[$y]}
echo "Connecting input:$in_l/$in_r to output:
$out_l/$out_r"
amixer cset numid=5 $in_l,$out_l,32768 > /dev/null 2>&1
amixer cset numid=5 $in_r,$out_r,32768 > /dev/null 2>&1
done
aplay -D$alsainterface -fcd "$track" > /dev/null 2>&1 &
echo "Playback started..."
echo -n "Any sound ? (y/(N)):"
read z
if [ "$z" == "y" ] ; then echo "Identified inputs: $in_l/$in_r" ;
break ; fi
for y in $(seq 0 $cnt_o) ; do
out_l=${outputs_left[$y]}
out_r=${outputs_right[$y]}
echo "Connecting input:$in_l/$in_r to output:
$out_l/$out_r"
amixer cset numid=5 $in_l,$out_l,0 > /dev/null 2>&1
amixer cset numid=5 $in_r,$out_r,0 > /dev/null 2>&1
done
done
if [ "$z" != "y" ] ; then echo "No working connection identified..." ; exit
1 ; fi
}
function scanoutputs () {
for k in $(seq 0 $cnt_o) ; do
out_l=${outputs_left[$k]}
out_r=${outputs_right[$k]}
echo "Connecting input:$in_l/$in_r to output: $out_l/$out_r"
amixer cset numid=5 $in_l,$out_l,32768 > /dev/null 2>&1
amixer cset numid=5 $in_r,$out_r,32768 > /dev/null 2>&1
aplay -D$alsainterface -fcd "$track" > /dev/null 2>&1 &
echo "Playback started..."
echo -n "Any sound ? (y/(N)):"
read z
if [ "$z" == "y" ] ; then echo "Identified outputs: $out_l/$out_r" ;
break ; fi
amixer cset numid=5 $in_l,$out_l,0 > /dev/null 2>&1
amixer cset numid=5 $in_r,$out_r,0 > /dev/null 2>&1
done
}
echo "Generating test-track..."
generate_track
init_matrix
echo "Scan inputs..."
scaninputs
init_matrix
echo "Scan outputs..."
scanoutputs
echo
echo "Result: inputs: $in_l/$in_r outputs: $out_l/$out_r"
exit 0
------------------------------------------------------------------------------
On Sun, Oct 25, 2009 at 9:53 AM, Georg Rudolph <georg.rudolph(a)schwaben.de>wrote:
> Klaus Schulz schrieb:
> > Hi Giso and others.
> >
> > I managed to get a basic routing working! I wrote a script which
> > initializes the entire matrix, connects one after another every input
> > to every output and plays a sound per setting.
> > I identified my alsainputs on 16/17 and analog outputs on 12/13 for my
> > analog extension board (out1 out2).
> >
> > This has not very much to do with the channel assignment I've seen
> > everywhere.
> >
> > With the script I can easily scan my setup now.
> >
> > Finally I need to find out how to get more then 2 channels from Alsa
> > to the same card without using Jack.
> >
> > The pity is really that I can not control the attenuation on the
> > hw-outputs with amixer. I'll do that on the application side.
> >
> > THX for all the hints so far.
> >
> > Cheers
> > \Klaus
> >
> Hallo Klaus,
>
> thanks for sharing this effort. I'm a happy user of a multiface II and
> would be very much interested in controlling remotely the submix
> (hardware mixer). This especially can't be substituted be software
> mixing with jackminimix for example. Would it be okay if you send me
> your script, so I could do a test too, and report back?
>
> I use perl scripts to control ardour, gtklick, jackminimix, once
> sooperlooper too, per osc, per an infrared remote, the mceusb2.
>
> If the hdsp could be controlled by amixer commans it wouldn't be so much
> needed to add an osc interface. Although, I like the osc interfaces.
>
> Best Regards,
> Georg Rudolph
>
>
Hi
I'm thinking about adding data files to my next CD. The pressing company
said it's the same price, if I can figure out how to deliver a disc
"that works as supposed". However that's where the problem lies :-)
I master with ardour, to get the transitions right + stuff in the ISRC
codes. This gives me an export.wav and an export.toc, that I burn like this:
cdrdao write export.toc
Now the pressing company told me that I need to add the audio first
leave the session open, and then add the data files. I looked at --multi
option for cdrdao, but couldn't really figure out what to do. Basically
I'm not even sure if it would be possible to add a "data-session" later
and if that should be "closed", or...
Have anyone any experience with this or something similar? Any advice or
help appreciated!
--
Atte
http://atte.dkhttp://modlys.dkhttp://virb.com/atte
Hi Klaus,
I hope you don't mind that I copy the reply to LAU (there are a lot RME
hdsp-users who probably know more about it that I do).
Klaus Schulz wrote:
> 1. Could you please share your .asoundrc. I guess it is the .asoundrc
> where you set "DSP".
It is not defined in the .asoundrc, but on ALSA itself: Try "cat
/proc/asound/cards" and find out the ALSA-name of your card, e.g. here
on my office PC it is:
0 [PAD ]: Digi96 - RME Digi96/8 PAD
RME Digi96/8 PAD at 0xfd000000, irq 21
1 [DSP ]: H-DSP - Hammerfall DSP
RME Hammerfall HDSP 9632 at 0xfeae0000, irq 21
2 [ICH5 ]: ICH4 - Intel ICH5
Intel ICH5 with AD1985 at irq 17
(the hdsp9652 is at home).
> 2. 32768=0db refers to the playback controls, They run at 0db@32768
> and +6db@64k
> However this seems to be somewhat different on the ouptuts. 0db=max
> position. I'd guess this should be 64k by default.
> Question: What do I actually control with below commands? I guess
> the playback-inputs only. How do I control the output channels?
> Probably I need to set it to 0db to.
The mixer in the hdsp cards can control routing from hardware inputs to
hardware outputs (for 'zero' latency monitoring) and the software output
to hardware outputs.
Hardware inputs are channels 0-25, software outputs 26-51, hardware
outputs 0-25. I don't think you can control the playback attenuation
with amixer. It might be (but that is a guess) that hdspmixer attenuates
the output channels only in software (by adding the output attenuation
to all relevant input/playback channels); When controlling only via
amixer I do not have to configure the output attenuation.
The channel mapping is also discussed here:
http://www.mail-archive.com/alsa-devel@lists.sourceforge.net/msg06019.html
> 3. I tried without success:
>
> /usr/bin/amixer -c 0 cset numid=5 26,2,32768 #first playbackchannel to
> first output of analog extension board (ch2)
> /usr/bin/amixer -c 0 cset numid=5 27,3,32768 #first playbackchannel to
> first output of analog extension board (ch3)
The channel numbering starts with zero. I do not know if the extension
board uses the same channels as ADAT out, maybe try also channels 8-15.
Cheers!
Giso
Adriano Petrosillo wrote:
> [ ... ]
>
> Oh, and I honestly don't remember if I already reported this, but volume
> meters at the bottom behave erratically (I think the refresh rate is WAY
> too high), and there is also a glitch in the layout: there's a small
> black box at the beginning of the volume meters. It's just a cosmetic
> problem, but still, I'd consider it as a bug :-)
Erratically? Ok. And yes, that constant little black segment does indeed look
like an "off by one" problem.