Quoting Anders Dahnielson <anders(a)dahnielson.com>:
> Frustrated I took a stab at writing a simple JACK client for the
> Freeverb3
> convolution reverb. But I can't get it to work properly. Could anyone
> knowledgeable take a look at it and help us out? I've posted the code of
> my
> (more or less copy-and-paste) attempt here:
>
> http://bb.linuxsampler.org/viewtopic.php?f=7&t=33#p291
Regarding recording impulses, one should really check Aliki on
http://www.kokkinizita.net/linuxaudio/ (manual and download on the download
page).
Sampo
Hello all,
A few days ago I asked Paul Davis what he knew concerning the AES32
and using the hdspmixer tool to access it's matrix. Currently it
doesn't support it, and given that I haven't been able to contact
anybody who developes hdspmixer or the AES32 driver, he was the only
one who I could find who could at least answer any questions. The
basic answer was "there isn't anybody to fix it for you." :P
So, I've spent the last two days messing with the hdspmixer code and
have confirmed my initial assumption that I have no idea what the hell
I'm doing. I started by trying to fix it so that the hdspmixer would
support cards from both the hdsp and hdspm drivers. I'm sure it could
be done, but the drivers code has diverged enough that it makes it
more work than I really want to do. So I yanked out all code that is
specific to hdsp and made it so that it will compile against the hdspm
code. Basically it's a version of hdspmixer for the AES32 and MADI
cards. (I'm not sure how the channel mapping in MADI works out, so
that's not in it yet)
The good news is that it will compile. The bad news is that it
immediately segfaults. And gdb doesn't really tell me anything useful,
but then again I really don't know how to use it.
Therefor I'm shamelessly soliciting any help that I can get. I really
want to be able to use the AES32, but I'm starting to get in over my
head. The hdspm driver is a bit silly without a functional matrix
mixer for it. I can send the code to anybody who would be willing to
give advice, give me code fixes, fill in the missing bits for MADI,
take over the project, or just provide more confirmation that I really
don't know what the hell I'm doing.
Oh, and as a little bonus I discovered that the hdspm driver only has
a dummy typedef for hdspm_peak_rms_t (as well as a few other
functions) so there is no code for input_peaks, output_peaks,
input_rms, playback_peaks or playback_rms. :P
-Reuben
Dear all,
The Linux Audio Conference 2008 in Cologne (Feb 28th - Mar 2nd 2008)
is just one month away now. The programme is shaping up, concerts are
being organized and coffee is about to be ordered.
To help us with planning the LAC2008 we kindly ask you to register now
at the conference website. This helps us to estimate how many visitors
we may expect, what individuals the audience is made of, and allows to
produce name tags for all attendees so that it becomes easier to
identify each other.
To register, please use the "Registration" form at
http://lac.linuxaudio.org
Also we now have put accommodation info plus some maps of the
conference location online. You can find these on
http://lac.linuxaudio.org under "Visitor Info".
Finally if you're living in Cologne or nearby: We are looking for
volunteers who would like to help out in any way, e.g. to host artists
and paper presenters in their flat. If you want to offer your help,
please contact the LAC2008 orga team at lac(a)linuxaudio.org
The LAC2008 chair is looking forward to have another great conference
with you all.
All the best
--
Frank Barknecht and Martin Rumori
Chairs of LAC2008
Hi,
I was planning to spend some effort at LAC this year to get jack
support in my Video Editor right, and I was wondering what would be
the best model to implement.
My App has a simple timeline that can have an arbitrary number of
tracks, and each track can hold clips at arbitrary positions. Each
clip is one audiofile or audiotrack from a videofile, which does not
make much of a difference. Eventually I also perform operations like
resampling or filters per clip.
Potential Models:
A)
1 Thread that connects to the jack-callback with a ringbuffer, and
that does everything, mixing, effects, disk-IO, etc.
Has the advantage that it is probably not very difficult to implement.
However, compute intensive stuff that is not IO bound happens outside
the jack-thread, which I guess is not optimal.
B)
1 Disk thread+ ringbuffer per clip. Advantage: compute intensive stuff
is in jack-thread. Disadvantage: does not scale to long timelines with
many clips and therefore many threads.
C)
1 Disk thread+ringbuffer per Track. Advantage: could eventually be
implemented such that compute-bound stuff happens in jack-thread, or
at least that "some" of this happens in the jack-thread. The number of
threads is likely to be small enough to handle that.
D)
1-n Single Diskthread(s) for everything, and 1 Ringbuffer per Track.
Make number of Disk-Threads limited, and eventually add more for
multi-core/SMP, and let those feed into per Track Ringbuffers
according to some home-brew scheduling algorithm? Wouldn't waste a
thread per track?
E)
Be lazy and reuse something that someone else has written. ;-)
So, what suggestions can you make about what I should do? I know that
there will be some not so nice corner cases when seeking, the
necessity to reset ringbuffers, scrubbing during playback, etc.....
My "Latency" Requirements are probably fairly moderate, so I could
compromise in that respect.
Another "Use-Case" that I would like to implement is
"Project-Nesting", preferable in "Real-Time", and only additionally by
rendering the nested project into a file. For this I think it would be
ok to have a big enough buffer between the current project and the
embedded project, and do the "on the fly rendering+IO" in the
"Disk-Thread".
what do you think?
Cheers
-Richard
--
Don't contribute to the Y10K problem!
We are jubilous to announce CLAM 1.2 'GSoCket plugged-in release'.
We had to wait for some months to make this release as we had to
redeploy the multiplatform release infrastructure [1]. Thus, the
feature buffer for this release is pretty full. It incorporates both,
the results of the Summer of Code [2] students work and the
involvement of David and Pau with Barcelona Media Foundation Audio
Research Lab[3].
We want to thank the involvement of GSoC students Hernan Hordiales[4],
Bennet Kolasinsky[5], Greg Kellum[6], Andreas Calvo, Roman Goj[7] and
Abe Kazemzadeh, Google Inc, and Barcelona Media audio lab members
for their precious involvement in CLAM.
[1] http://clam.iua.upf.edu/testfarm/
[2] http://clam.iua.upf.edu/wikis/clam/index.php/GSoC_2007
[3] http://www.barcelonamedia.org/index.php/linies/10/en
[4] http://h.ordia.com.ar
[5] http://bennettdoesclam.blogspot.com
[6] http://gregkellum.com
[7] http://ro-baczek.blogspot.com
A summarized list of changes follows. See also the CHANGES files[8]
for details, or the development screenshots[9] for a visual guided tour.
As usual binary packages for Windows, MacOSX and several flavors of Linux
are available to download.
[8] http://iua-share.upf.edu/svn/clam/trunk/CLAM/CHANGES
[9] http://clam.iua.upf.edu/wikis/clam/index.php/Development_screenshots
Summary of changes:
The most exciting feature is the new plugin system (acalvo)
which enables third party algorithms to be distributed separately
from the core binaries. LADSPA plugins support has been enhanced
and a first iteration on FAUST[10] integration. The wiki[11] contains
very nice how-to's that cover most of that.
[10] http://faust.grame.fr/
[11] http://clam.iua.upf.edu/wikis/clam
Most of the GSoC work come as plugins: a SMS Synthesizer (gkellum),
a Voice synthesis/analysis (akazem) and some some cool guitar effects
(hordia). Also not included as plugins but in the main repository
several enhancements have been done on the SMS transformations (hordia)
and the tonal analysis (rgoj).
Some interesting work has been done on the Barcelona Media Audio Lab
on having a system to simulate 3D room acoustics which can be reproduced
on several exhibition systems. Some precomputed room databases are
available to try. Check the wiki NetworkEditor Tutorial for more
information.
Regarding the applications, Network Editor incorporates new usability
enhancements, a new on-line Tutorial and a new Spectrogram like view.
The Annotator received Bennet Kolasinsky attention improving its the
flexibility of its interface, the practical effects are multiple
segmentation and low-level descriptors panes and that we are pretty
close to visualization and auralization plugins.
Enjoy.
The CLAM Team
The PCI card I'm working with was made to receive the signal as a 32 bits frame (16 bits by channels) each time I write to the file device so I have to turn the src_data.data_out frames (2 floats) into a unique 32bit unsigned long and write it:
1) "unsigned long frame;" : I declare an unsigned long that will receive the entire frame.
2) "frame = (unsigned long) ((src_data.data_out[sent]))>>16;" : I convert the float at the index equal to sent and initialise the frame with it. So my frame look like this 0000000000000000XXXXXXXXXXXXXXXX Where Xs represent one channel signal.
3) "frame |= (((unsigned long) (src_data.data_out[sent+1])) & 0xFFFF0000);" Apply a mask to the 16 first low bits of the data_out float value and aplly a OR between it and frame. It should look like this :
yyyyyyyyyyyyyyyy0000000000000000 | 0000000000000000XXXXXXXXXXXXXXXX = YYYYYYYYYYYYYYYYXXXXXXXXXXXX
Where Ys is the other channel signal.
Is it clear enough,I hope so :)
I have to emphasize that this stuff seems to work because when I send the src_data.data_in frames buffer this way, the signal is perfect.
> Message du 11/02/08 14:12
> De : "Erik de Castro Lopo" <mle+la(a)mega-nerd.com>
> A : linux-audio-dev(a)lists.linuxaudio.org
> Copie à : "Mathieu Dalexis" <dev.audioaero(a)orange.fr>
> Objet : Re: [LAD] (no subject)
>
> Mathieu Dalexis wrote:
>
> > unsigned long frame;
> > frame = (unsigned long) ((src_data.data_out[sent]))>>16;
> > frame |= (((unsigned long) (src_data.data_out[sent+1])) & 0xFFFF0000);
>
> Sorry, but what is all this stuff?
>
> Erik
> --
> -----------------------------------------------------------------
> Erik de Castro Lopo
> -----------------------------------------------------------------
> "Projects promoting programming in natural language are intrinsically
> doomed to fail." -- Edsger Dijkstra
> ---------------------------------------------------------------------------------------
> Orange vous informe que cet e-mail a ete controle par l'anti-virus mail.
> Aucun virus connu a ce jour par nos services n'a ete detecte.
>
>
>
Hello,
I try to perform dynamic upsampling on a signal (SR=44,1kHz) to double the sample rate.
Here is the code I wrote to test it. It works using a raw audio file (cdparanoia -r 1) with a 1Khz sin signal.
When I look to my output signal with an oscilloscope it looks quite good but I get some noise (glitches?) that makes it really sound bad.
Is there something I do wrong ? Is it a libsamplerate problem ?
**************************
******My code*************
**************************
#include <stdio.h>
#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#endif
#include <math.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/types.h>
#include <sys/time.h>
#include <samplerate.h>
int
main(int argc, const char *argv[])
{
 int TAILLEBUF = TAILLEBUF;
 int TAILLEBUF_OUT = TAILLEBUF;
 int16_t * p_readbuf= (int16_t *) malloc(sizeof(int16_t)*TAILLEBUF);
 float * pf_readbuf= ( float *) malloc(sizeof(float)*TAILLEBUF);
 float * pf_writebuf= ( float *) malloc(sizeof(float)*TAILLEBUF_OUT);
 int fini=0;
 int finiRead=0;
 int fd,fdread;   /* sound device file descriptor */
Â
 struct timeval tv;
 tv.tv_sec = 0;
 tv.tv_usec = 120000;
 int error;
 SRC_STATE * src_state;
 SRC_DATA src_data;
 src_data.src_ratio = atoi(argv[3]);
 src_data.input_frames=0;
 src_data.end_of_input = 0;
 src_data.data_out = pf_writebuf;
 src_data.output_frames=TAILLEBUF_OUT/2;
 src_state = src_new(2,2,&error);
 /* open sound device I'm working on my own sound card */
 fd = open("/dev/AudioDriver/data", O_WRONLY);
 if (fd < 0) {
   perror("open of /dev/AudioDriver/data failed");
   exit(1);
 }
/*My raw sound file*/
 fdread = open("./cdda.raw", O_RDONLY);
 if (fd < 0) {
   perror("open of cdda.raw failed");
   exit(1);
 }
 fd_set writefds[1];
 FD_ZERO(writefds);
 FD_SET(fd, writefds);
 int k=0;
      while (!finiRead) {
   /* read TAILLEBUF/2 Frames */
   int j;
   if(read(fdread,p_readbuf,TAILLEBUF*sizeof(int16_t))!=TAILLEBUF*sizeof(int16_t))
         finiRead=1;
  Â
   int i=0;
   /*Convert from 16bit integer to Float*/
         for(i=0;i<TAILLEBUF;i++)
    {
      pf_readbuf[i]=((p_readbuf[i])<<16) ;
      }
     Â
   if(src_data.input_frames==0)
    {
      src_data.data_in = pf_readbuf;
      src_data.input_frames=TAILLEBUF/2;
    }
   tv.tv_sec = 0;
   tv.tv_usec = 120000;
   fd_set writefds[1];
   FD_ZERO(writefds);
   FD_SET(fd, writefds);
   while(select(fd+1, NULL, writefds, NULL, &tv)!=1){  Â
    tv.tv_sec = 0;
    tv.tv_usec = 120000;
    fd_set writefds[1];
    FD_ZERO(writefds);
    FD_SET(fd, writefds);
   }
   int sent = 0;
   Â
    if((error = src_process(src_state,&src_data))!=0)printf("error\n");
  Â
    while(sent < src_data.output_frames_gen)
      {
       /*Convert my 2-floats frame to an unsigned long frame containing my to channels*/
        unsigned long frame;
        frame = (unsigned long) ((src_data.data_out[sent]))>>16;
        frame |= (((unsigned long) (src_data.data_out[sent+1])) & 0xFFFF0000);
        if(write(fd, &frame, sizeof(unsigned long))!=sizeof(unsigned long))
      {fini = 1;
      }
        sent += 2;
      }
 exit(0);
}
****************************
****************************
****************************
I hope that you can understand my english and that my code is clear enough for you to help me.
Thanks in advance.
Mathieu
I received a private reply to my request regarding
a jack-midi equivalent to pmidi. This may be of
interest to others as well.
Jpmidi
<http://www.geocities.com/kellinwood/jpmidi/index.html>
does the job nicely.
Unlike pmidi, jpmidi is a 'resident' program - it
creates it own command prompt, allowing e.g. to
connect it and start / stop playing (it controls
and follows jack's transport).
Given this it's odd that it doesn't have a 'load'
command. To switch to another midi file you have
to quit and restart.
--
FA
Laboratorio di Acustica ed Elettroacustica
Parma, Italia
Lascia la spina, cogli la rosa.
LAC 2008: bandwidth to burn and volunteers needed
The 6th annual Linux Audio Conference is taking place in Cologne, Germany, Feb
28th to March 2nd, 2008. As with each previous year this year's conference will
be streamed live over the internet in ogg theora via icecast. The stream server
is up at: http://lac2008.khm.de:8000/
There is nothing to see at the moment, but keep checking over the coming days
as we hope to have a test stream up soon.
This year we are in the unique situation of having a Gigabit link donated by
CITIZENMEDIA: http://www.ist-citizenmedia.org/ They have asked us to use up as
much of their bandwidth as we can so they can see how well the link performs.
This year the core team, Joern Nettingsmeier and myself, are recruiting
volunteers to spread the workload. To that end we have set up a mailing list
and irc channel to coordinate our efforts. We will also have a wiki shortly.
If you will be coming to Cologne for the conference please consider signing up
to help. If you are not coming, please enjoy the fruits of our labors by
watching the streams and participating via irc.
stream team mailing list: http://zhevny.com/mailman/listinfo/lac-streams
general conference chat: #lac2008 on irc.freenode.net
stream team tech talk: #lac2008-tech on irc.freenode.net
Thanks,
Eric Rz.
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
The Aqualung development team is pleased to announce the latest
release of the Aqualung music player.
Aqualung is an advanced music player originally targeted at the
GNU/Linux operating system, today also running on FreeBSD, OpenBSD and
Microsoft Windows. It plays audio CDs, internet radio streams and
podcasts as well as soundfiles in just about any audio format and has
the feature of inserting no gaps between adjacent tracks.
The ChangeLog is included below.
Please note that our website has moved to http://aqualung.factorial.hu
Please update your links and bookmarks.
Enjoy,
Tom
2008-02-10 Tom Szilagyi <tszilagyi at users dot sourceforge dot net>
* Aqualung 0.9beta9.1
http://aqualung.factorial.hu
This is a bugfix, stability and performance oriented release also
containing a few updates to existing functionality. By using this
version, your Aqualung will be more stable, and in some cases
significantly faster. All users are encouraged to upgrade.
The project homepage has moved to http://aqualung.factorial.hu
Please upgrade your pointers and bookmarks.
Notable changes:
* Playlist code refactoring for improved performance. Please note
that incompatible changes have been made to the playlist format:
this means that your old playlists won't be parsed, you'll have to
re-create them.
(NOTE: Music Store contents are unaffected. If we ever change the
Music Store format in a backward-incompatible way, we will provide
tools to migrate your precious store data.)
* Fix threading problems that caused random crashes for some users.
* Fix lurking bug that sometimes resulted in getting stuck at the
beginning of a track when Sinc interpolator sample rate converters
were used.
* Modified the way of opening ALSA output to achieve non-exclusive
driver access.
* New, more versatile title string generating templates. Make sure
to check the documentation.
* Added option to periodically save playlist.
* Several fixes concerning command line file and playlist loading,
esp. with the -L flag.
* Better metadata handling for external files in playlist.
* Increment CDDB revision number on resubmitting an existing
disc. This is essential for correcting existing CDDB entries,
otherwise the CDDB server rejects the submission. The latest CVS
version from http://libcddb.sf.net is required for this to work
(Aqualung-Win32 is built with this version).
* OpenBSD-related portability fixes. Aqualung should now compile
cleanly and be fully functional OOTB on OpenBSD 4.2.
* Updated German, Hungarian and Italian translations. Added
Russian translation.
