Hi, I have problems with midi-output on my OS11.2/Gnome. I installed fluidsynth and qsynth. When I try to play a midi-file with note-edit, the sound is very distorted, unrecognizable. That's on my snd-hda-intel card. What could be the problem ? I also do have a soundblaser live in the machine. When I take this as a midi-output, I don't have any sound, unless I install a soundfont with asfxload. Is it possible to have the soundfont automatically loaded on boot ? I have seen something in /etc/udev/rules and /etc/alsa.d, but I don't understand these. P.S. with OS11.2/KDE fluidsynth and qsynth worked fine, but I had other sound-isues, so I re-installed and used gnome, which seems better for multimedia. Thanks for any pointers and/or remarks to help to solve this. Koenraad Lelong. -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
On Thursday, December 17, 2009, Koenraad Lelong wrote:
Hi,
I have problems with midi-output on my OS11.2/Gnome. I installed fluidsynth and qsynth. When I try to play a midi-file with note-edit, the sound is very distorted, unrecognizable. That's on my snd-hda-intel card. What could be the problem ?
It is hard to say without more information. It happens only with Noteedit, or also with other MIDI applications? Try FluidSynth 1.0.8 or 1.0.9, but not older. The newest release 1.1.0 has some problems with QSynth, that will be fixed in the not yet published 1.1.1 release. Sorry for all the mess ;) In order to isolate the problem, my advice would be to try a simple scenario like this one: in one terminal, start fluidsynth: $ fluidsynth -a alsa file.sf2 (you may want to try "-a pulseaudio" or "-a oss" if you want) The command line interface of fluidsynth provides a shell-like session, where you can try commands. For instance, you may want to set the gain value to a lower value with "gain 0.4". Try the "settings" command to see all the current settings. Later, you would be able to set most settings using the QSynth friendly GUI. In another terminal, use the aplaymidi (from alsa-utils) utility: $ aplaymidi -l ... $ aplaymidi -p FLUID song.mid
I also do have a soundblaser live in the machine. When I take this as a midi-output, I don't have any sound, unless I install a soundfont with asfxload. Is it possible to have the soundfont automatically loaded on boot ? I have seen something in /etc/udev/rules and /etc/alsa.d, but I don't understand these.
I don't have an Audigy, but yes, this card has a MIDI hardware synth, and need a soundfont. The RPM package awesfx provides the /etc/udev/rules.d and /etc/alsa.d scripts to load soundfont files into the card. You should copy your .SF2 files into the /usr/share/sounds/sf2 directory, and write the name of your preferred one in the configuration file /etc/sysconfig/sound following the key name SOUNDFONT-FILES. Regards, Pedro -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
copy your .SF2 files into the /usr/share/sounds/sf2 directory, and write the name of your preferred one in the configuration file /etc/sysconfig/sound following the key name SOUNDFONT-FILES.
If you want a GUI tool for this task: YaST2->System->/etc/sysconfig Editor-> Hardware->Soundcard->SOUNDFONT_FILES. A good and free SoundFont for SB cards and FluidSynth is General User GS. http://www.schristiancollins.com/generaluser.php packman package: http://packman.links2linux.de/package/GeneralUser Regards, Pedro -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
On 17.12.2009 17:16, Pedro Lopez-Cabanillas wrote:
copy your .SF2 files into the /usr/share/sounds/sf2 directory, and write the name of your preferred one in the configuration file /etc/sysconfig/sound following the key name SOUNDFONT-FILES.
If you want a GUI tool for this task: YaST2->System->/etc/sysconfig Editor-> Hardware->Soundcard->SOUNDFONT_FILES.
Or use "Others" -> "Install SoundFonts" in yast sound card module. It copies SoundFonts from the driver CD and installs them automatically. -- Best Regards Ladislav Slezák Yast Developer ------------------------------------------------------------------------ SUSE LINUX, s.r.o. e-mail: lslezak@suse.cz Lihovarská 1060/12 tel: +420 284 028 960 190 00 Prague 9 fax: +420 284 028 951 Czech Republic http://www.suse.cz/ -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
On Tuesday, January 19, 2010, Ladislav Slezak wrote:
On 17.12.2009 17:16, Pedro Lopez-Cabanillas wrote:
copy your .SF2 files into the /usr/share/sounds/sf2 directory, and write the name of your preferred one in the configuration file /etc/sysconfig/sound following the key name SOUNDFONT-FILES.
If you want a GUI tool for this task: YaST2->System->/etc/sysconfig Editor-> Hardware->Soundcard->SOUNDFONT_FILES.
Or use "Others" -> "Install SoundFonts" in yast sound card module. It copies SoundFonts from the driver CD and installs them automatically.
Nice. I don't have an Audigy/Live/AWE card. I suppose that this option is hidden for any other soundcard models, right? Anyway, there are more SoundFonts available out there in addition to those provided by the manufacturer driver CD. For instance, GeneralUser by Christian S.Collins or the collections listed in http://hammersound.net . Is it possible to install and configure them using yast? And what about soft-synths like FluidSynth to render MIDI events, needed by owners of simple sound cards like the integrated devices in most laptops and cheap PCs? But looks like the original poster could configure his Audigy already, and he is now struggling with distorted sound output. Probably because he is using PulseAudio and the defaults don't work well for his system. It would be nice to have a yast module to configure PulseAudio's /etc/pulse/daemon.conf and /etc/security/limits.conf For my Intel HDA integrated soundcard (asus laptop), these are the working values that I needed to change: * File: /etc/pulse/daemon.conf realtime-scheduling = yes realtime-priority = 5 default-sample-rate = 48000 default-fragments = 6 default-fragment-size-msec = 48 * File: /etc/security/limits.conf (my userid is a member of the audio group). @audio - rtprio 70 @audio - memlock unlimited @audio - nice -20 Regards, Pedro -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
On 19.1.2010 16:35, Pedro Lopez-Cabanillas wrote:
On Tuesday, January 19, 2010, Ladislav Slezak wrote:
Or use "Others" -> "Install SoundFonts" in yast sound card module. It copies SoundFonts from the driver CD and installs them automatically.
Nice. I don't have an Audigy/Live/AWE card. I suppose that this option is hidden for any other soundcard models, right?
Oh yes, it's displayed only for AWE/Live/Audigy cards. I forgot to mention that...
Anyway, there are more SoundFonts available out there in addition to those provided by the manufacturer driver CD. For instance, GeneralUser by Christian S.Collins or the collections listed in http://hammersound.net . Is it possible to install and configure them using yast?
Well, in theory yes, but you would have to burn them on a CD. Yast searches files directly on CD/DVD medium, you cannot choose a sound font file manually... (I had a feture request for that but I had to drop it in favor of other features and bugfixes.) -- Best Regards Ladislav Slezák Yast Developer ------------------------------------------------------------------------ SUSE LINUX, s.r.o. e-mail: lslezak@suse.cz Lihovarská 1060/12 tel: +420 284 028 960 190 00 Prague 9 fax: +420 284 028 951 Czech Republic http://www.suse.cz/ -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
Pedro Lopez-Cabanillas schreef:
On Thursday, December 17, 2009, Koenraad Lelong wrote:
Hi,
...
$ fluidsynth -a alsa file.sf2 ... In another terminal, use the aplaymidi (from alsa-utils) utility: $ aplaymidi -l ... $ aplaymidi -p FLUID song.mid
Hi Pedro, I tried the above and that works. When I was trying to determine the versions of the software, I got a notification of an update of qsynth. I installed it, and after killing some remaining instances of it I started it and it works now. fluidsynth is 1.0.9 qsynth is now 0.3.4 ...
following the key name SOUNDFONT-FILES. Modifying this with yast's sysconfig editor loads the soundfont at boot.
Regards, Pedro
Many thanks Pedro. Now I have to try KDE again. Regards, Koenraad Lelong. -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
Koenraad Lelong schreef:
Pedro Lopez-Cabanillas schreef:
On Thursday, December 17, 2009, Koenraad Lelong wrote:
Hi,
... $ fluidsynth -a alsa file.sf2 ... In another terminal, use the aplaymidi (from alsa-utils) utility: $ aplaymidi -l ... $ aplaymidi -p FLUID song.mid
Hi Pedro,
I tried the above and that works. When I was trying to determine the versions of the software, I got a notification of an update of qsynth. I installed it, and after killing some remaining instances of it I started it and it works now.
fluidsynth is 1.0.9 qsynth is now 0.3.4
...
following the key name SOUNDFONT-FILES. Modifying this with yast's sysconfig editor loads the soundfont at boot.
Regards, Pedro
Many thanks Pedro. Now I have to try KDE again.
Regards,
Koenraad Lelong.
Hi, I finally tried again, after having some troubles again. I seems that if fluidsynth runs as root, it (mostly) works fine but when I run it as an ordinary user, I get distorted output. I didn't use qsynth this time, altough the problems are similar. Any more suggestions ? What can I do to help this get sorted out ? Regards, Koenraad Lelong. -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
On Saturday, January 16, 2010, Koenraad Lelong wrote:
I finally tried again, after having some troubles again. I seems that if fluidsynth runs as root, it (mostly) works fine but when I run it as an ordinary user, I get distorted output. I didn't use qsynth this time, altough the problems are similar.
Any more suggestions ? What can I do to help this get sorted out ?
It is hard to diagnose with only a "distorted output" description. Maybe it is a problem with xruns? in this case, try increasing the buffer sizes. If you are using the alsa audio output: $ fluidsynth -a alsa -oaudio.period-size=1024 -oaudio.periods=4 soundfont.sf2 It is much more comfortable using QSynth, the fields are labeled "Buffer Size" and "Buffer Count" in the audio tab of the setup dialog. The minimum nubers are very dependent on the soundcard model. Did you see warnings like these? fluidsynth: warning: Failed to pin the sample data to RAM; swapping is possible. fluidsynth: warning: Failed to set thread to high priority The above warnings are about the realtime priority, and can be solved adding the following lines to the file /etc/security/limits.conf and adding yourself to the audio group: @audio - rtprio 90 @audio - memlock unlimited Please read this article: http://en.opensuse.org/JackLab/Assigning_real-time_priorities_with_PAM BTW, have you tried kmid2? http://kde-apps.org/content/show.php/KMid2?content=116404 Regards, Pedro -- To unsubscribe, e-mail: opensuse-multimedia+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-multimedia+help@opensuse.org
participants (3)
-
Koenraad Lelong -
Ladislav Slezak -
Pedro Lopez-Cabanillas