>
> Message: 5
> Date: Mon, 20 Aug 2007 20:37:08 +0200
> From: Fons Adriaensen <fons(a)kokkinizita.net>
> Subject: Re: [LAU] Re: Jackmp - libjackdmp.so
> To: linux-audio-user(a)lists.linuxaudio.org
> Message-ID: <20070820183708.GA6201(a)linux-2.site>
> Content-Type: text/plain; charset=iso-8859-1
>
> On Mon, Aug 20, 2007 at 05:38:34PM +0200, Stéphane Letz wrote:
>
>> Any program that correctly does dynamic linking should work with
>> ackdmp. I remember some (?? don't remember which right now )-:) that
>> were doing static linking with a specific version of libjack.so thus
>> failing to work with jackdmp.
>
> I have a weird problem with alsaplayer.
>
> I start jackdmp on hw:1. Everything works fine, except alsaplayer.
> When started it plays to both hw:0 and hw:1, at twice the normal
> speed but normal pitch. Now I guess it was linked statically
> with libjack (Grrrr), but why/how it manages to use both servers
> at the same time I just can't imagine...
Hum, how can that happens?? Do you really have jackd and jackdmp
running at the same time??
>
> The autotools files just are beyond my comprehension. Does anyone
> know the trick to make alsaplayer use the dynamic libjack (and only
> that) ?
>
> jackdmp 0.63
> alsaplayer 0.99.80-rc2
>
Recompile it with jackdmp installed? ((-:
Stephane
Hi everyone,
Qsynth 0.3.0 is now out for you to try and guess what? This marks the
point of no return to the aging Qt3 framework. Yes, Qt4 migration was
complete.
Hints from the change-log might be shallow, but nevertheless:
- Qt4 migration has comenced and is now complete. Care must be taken
with this new configuration file and location: this release starts a new
one from scratch and won't reuse any of the previous existing ones,
although cut and paste might help if you know what you'll be doing :)
- Application icon is now installed to ${prefix}/share/pixmaps;
application desktop entry file is now included in installation; spec
file (RPM) is now a bit more openSUSE compliant; initial debianization.
- Default font option names were adjusted to "Sans Serif" and
"Monospace", wherever available.
- The "keep child windows always on top" option is not set as default
anymore, because window focus behavior gets tricky on some desktop
environments (eg. Mac OS X, Gnome).
- Autoconf (configure) scripting gets an update.
Good grief. All this is rteadily available from the usual place:
http://qsynth.sourceforge.nethttp://sourceforge.net/projects/qsynth
Cheers && Enjoy &
--
rncbc aka Rui Nuno Capela
rncbc(a)rncbc.org
Hello,
Maybe I just not a master googler, but I didnt find any answer why qjackctl
limits maximum of input or output channels to 32. Also PD has this limit
compiled in and so on, but jackd starts with 64 channels with my MADI-card ?
Is there a technical reason or can I just remove the limits in the code ?
Its not so bad becaused I always used alsa-only before, but I just want give
jack a try.
BTW, I also would suggest that if a limit is needed, it should be 256 channels
(4 MADI-Cards) I thinks its the maximum you can do nowaday with linux, but if
there is the rule of computer speed, so I can think, this count will double
each 2 years, so to be save for next 4 years we need 1024 channels limit.
mfg winfried
--
--
- Winfried Ritsch
- ritsch(a)iem.at - http://iem.at/ritsch
- Institut fuer Elektronische Musik und Akustik
- University of Music and Dramatic Art Graz
- Tel. ++43-316-389-3510 (3170) Fax ++43-316-389-3171
- PGP-ID 69617A69 (see keyserver http://wwwkeys.at.gpg.net/)
--
Quoting Matthew Polashek <matt(a)tinysongs.com>:
> Ok, I'm throwing together a new box in which I will install
> UbuntuStudio with the intention of using it as a live processing rig
> for live performance using PD. I'm interested in discovering whether
> it makes more sense to get a Core 2 Quad processor that runs with a
> bus speed of 1066 or a Core 2 Duo processor that runs with a bus
> speed of 1333. The idea is to reduce latency as much as possible.
> I'll get I/O via a RME 9652 and a couple Behringer ADA8000 units.
> Also, any thoughts on how ram speed plays into this equation? I'm
> looking at the Intel dg33fb board.
Ardour is multi-threaded, but at the moment, Ardours' audio processing is
run in a single thread. This means that the audio work done by jack cannot
be spanned over two processors.
Using jackdmp will enable your system to run some jack clients in parallel
(depending on the connection scheme between them). But this only means that
different jack clients will be able to run in parallel. It doesn't make
Ardour process in multiple threads.
This is something we will work on in the future, but at the moment, this
makes the choice quite clear: core 2 duo should suffice well. A quad core
will be more future proof though. I can't really comment on bus speed or ram
speed. Like always, the faster the better, but the question is how much of a
difference it will make? My guess is that while it would be faster, it
probably is not that much faster that you could run with half the buffer
size than with the slower bus/ram.
Before you get really dissappointed by this fact, remember that while only
one cpu can be used for ardour work, it means thath the second core (or rest
of them in the case of the quad) is free to run all those other tasks in
your system. This means that the ardour gui, the desktop and other
applications will be practically unaffected by the DSP load on your box.
This is a big boon on 2 cpu systems.
Sampo
Greetings:
I am seeking assistance with a couple of issues relating to my Ubuntu
server stream box.
1. I am currently running darkice to create a stream from a Delta1010
interface. Is there a way to create additional streams from
additional inputs? I have everything configured with ALSA for the
first stereo input pair, but I cannot figure out how to create
another stream from the second stereo pair.
2. I would like to configure a box with as little config as possible
to connect to a stream on boot and pass the audio to the audio
port. I cannot find a command line util to do this, and I am not
sure how to configure something like this to automatically connect
on boot.
My goal is to build a temporary (one year) backup Studio to Transmitter
link for our station. We currently have a wireless lan connection to
campus because we are located off campus. This existing configuration
bottlenecks at about 20 listeners with our Icecast2 server on the studio
end of the network. New construction has blocked our microwave path and
reduced us to a POTS codec to feed our FM transmitter.
We are about to move our microwave transmitter to a different building
on-campus and deliver the audio over the IP connection. At the same
time, I would like to move our Icecast2 relay box to the new on-campus
location to get the connecting traffic off of the wireless link. If I
can, I would like to send a higher bandwidth stream for the microwave in
addition standard "low" and "high" .mp3 streams. Right now, if I add
the higher bandwidth stream to the current darkice config, all streams
start to skip. While I can drive everything off of one pair of analog
inputs on the Delta1010, I would like to use two seperate stereo pair so
that Emergency Alerts are not on the webstream feed.
Any suggested configurations would be most helpful. Thanks all.
MattRock
--
Matt Rockwell- Technical Director
University of Wisconsin- WSUM Student Radio
http://wsum.net/
>
> Message: 1
> Date: Mon, 20 Aug 2007 15:54:18 +0200
> From: Lars Luthman <lars.luthman(a)gmail.com>
> Subject: Re: [LAU] Re: Jackmp - libjackdmp.so
> To: linux-audio-user(a)lists.linuxaudio.org
> Message-ID: <1187618058.4917.1.camel(a)box.lars>
> Content-Type: text/plain; charset="utf-8"
>
> On Mon, 2007-08-20 at 15:24 +0200, Stéphane Letz wrote:
>> Le 20 août 07 à 15:11, Asmo Koskinen a écrit :
>>
>>> Stéphane Letz kirjoitti:
>>>> The best is always to get the latest published version (0.63 for
>>>> now).
>>>>
>>>>
>>>
>>> OK, it seems that I have try this version:
>>>
>>> asmok@ubuntu:~$ /usr/local/bin/jackdmp
>>> jackdmp 0.64
>>
>> This is the SVN version. Better use the published 0.63 version
>> (http://www.grame.fr/~letz/jackdmp_0.63.zip)
>>
>>>
>>> How about other programs? Do you know what programs can use jackdmp
>>> instead of jackd? Is there any available?
>>>
>>> Bets regards Asmo Koskinen.
>>
>> Any program that correctly does dynamic linking should work with
>> jackdmp. I remember some (?? don't remember which right now )-:) that
>> were doing static linking with a specific version of libjack.so thus
>> failing to work with jackdmp.
>
> Aren't there some differences with the threads in which certain
> callbacks are executed,
Yes, they are 2 threads, one for audio callback (Real-Time) and one
for non RT code, that is notifications from the server. But in
general this is completely transparent for applications, and I don't
know of any applications that show problems because of that (if they
are some, please tell me... ((:)
> and with MIDI?
>
Dmitry Baikov was working on that, and part of the code for MIDI in
already in jackdmp MIDI branch, although it is not complete yet.
Stephane
kara-moon have just opened a new topic on their site called 'The open
source alternative'.
They are very keen to hear about peoples experiences with regard to
setting up a Linux based studio.
One guy in particular wants to build a complete system from scratch,
and has the luxury of plenty of time.
However, I get the impression they are mostly interested in a fairly
easy to set up system, some advice and warnings, and maybe a bit of
simple encouragement.
http://www.kara-moon.com/forum/index.php?topic=892.0
--
Will J Godfrey
http://www.musically.me.uk
Thanks a lot, Julien, for your "tutorial". I have successfully set things up
and managed to get sound and all! Only problem is I have too much lag. I
think I would have to install the realtime kernel, and try my USB Audiophile
soundcard. I'm pretty worn out from these weeks of struggling though.
Yesterday I launched my Windows laptop and within two hours had completed
the MIDI to VST output process. That was so nice. I think I'll stick with it
and record the VSTI-ed output in wav files, then send them over to my Linux
pc for the audio recording and editing (which works very well without
problems). If I ever get a problem with my laptop I'll try my hand in
returning to a Linux VSTI set-up, but for now I'm exhausted, and I'd rather
cross the river if my current bridge ever fails. I hope the next Ardour
release will give an easier time to aspiring Linux musicians with specific
needs (I realize I was asking for something pretty specific. "standard"
audio and MIDI isn't that hard to get). Thankfully, the user community is
very supportive! :)
Cheers,
Dominic
On 8/20/07, Chuckk Hubbard <badmuthahubbard(a)gmail.com> wrote:
>
> On 8/19/07, Julien Claassen <julien(a)c-lab.de> wrote:
> >
> > Hi!
> > I think you might try ardour. I know it can host and record
> > VST-generated
> > sounds. At least I believe so. Anyone can confrim?
> > With your setup, it seems to me you might want to kick timidity++ out
> > of
> > your chain. Timidity plays sounds using its own sounds (GUS-patches or
> > soundfonts), if not for me it had a great latency. I don't know exactly
> > about
> > freeST, but it might have it's own midi port. So in an xterm or on the
> > console
> > you may try:
> > aconnect -li
> > and
> > aconnect -lo
> > This will display all alsasequencer midi-ports currently in existence.
> > You
> > can also use some graphical patchbay to do the same. I think ghostess is
> > a
> > nice one, just search for "patchbay", you should get a list of them.
> > Then connect your rosegarden midi-out (if it has one) to your freeVST
> > in (if
> > that has one) and connect your FreeVST jack_output to audacity. So take
> > the
>
>
> I'm also fond of the aconnect solution. You aren't reliant on how the
> sequencer's developers decided to set things up. Is there any loss of
> timing, anyone? Although right now, Csound's ALSA midi output isn't working
> for me.
> I haven't been able to get Rosegarden to load Csound's csLADSPA, perhaps
> it doesn't use LADSPA_PATH to search for plugins. It has its own plugin
> directory. I know Rosegarden is pretty and a bit more developed, but Muse
> seems to be more useful sometimes. Of course neither one has an interface
> like some of the commercial programs- yet- but I think it'll happen,
> gradually.
>
> I haven't used Ardour much, I understand it's pretty cool, but you could
> totally use Audacity for both steps 2 and 3.
> ALTHOUGH, I read that someone was commissioned to add MIDI sequencing to
> Ardour! So maybe in another few months, Ardour could be the only program
> necessary for the whole process! Future Linux composers will look back and
> say "How did they manage like that?"
>
> -Chuckk
>
This has to be one of those problems with an overly simple solution, but it
escapes me atm. I first noticed it with xmms, playing music, of any sort,
it will play fine for a while, then will stop, as if waiting to fill up a
buffer, but never starts back again, it also continues to display the play
symbol but no time advancement. I've narrowed it down to being an alsa problem,
because when I run aplay somesound.wav, aplay never exits, and the sound loops
indefinitely.