Hi,
what do people use for damage control (prevention likely is a case for
deadcats though I am not sure whether their absorption may be too much
for this application) when condenser mics occasionally "plop" due to
wind?
One can hear this effect a few times on the outdoors "wasp" video
<https://youtu.be/vKCdTh7h8f8> between 5:00 and the end (5:22). What I
tried in that recording is using a high pass filter (at about 100Hz)
with "Soft limiter" which supposedly maps something like -3dB to +∞dB to
-3dB to 0dB. I think it was "Invada". The result is still unpleasant.
Are there any better approaches, like some sort of smooth gating?
As a note aside, I got really annoyed at snd_oxfw (Firewire via ALSA, no
Ffado available) for a Mackie Onyx Satellite. At 96kHz sample
frequency, xruns using Ardour about once per minute. Don't remember
this being the case when I tried last time (with a Thinkpad T61 and an
older CPU). Regardless whether using the built-in Ricoh Firewire
controller of a T420, or an Expresscard controller with TI chip, using
Jack or not, connected to mains power or not. I finally threw in the
towel and used the analog TLR outputs on the Satellite into an RME
Hammerfall (via Expresscard-to-Cardbus Adapter). I am not sure I really
was satisfied with the result: I suspect that it went through A/D+D/A
already before arriving at Main Out of the Satellite. The video from
that attempt was sort of badly lit so I ultimately ditched it anyway.
The one linked above was instead recorded with an Alesis iO|14 audio
card, using Ffado and Jack (this requires blacklisting the ALSA Bebob
driver which only produced hacked-up audio last time I tried it). I did
not want to move my regular equipment (a large Mackie Onyx mixer) into
the yard and at least had the Alesis still around from earlier
experiments. I actually found the noise level of the Alesis a bit nicer
than when using the Satellite preamps, but then the Satellite produces
lacklustre phantom voltage (something like 35V or so).
A final note on the mics (leaving the Linux-specific realm): I used
hypercardioid Oktava MK-012 capsules (I think 0.5" diameter membranes).
Would cardioid or even omni be less sensitive to wind as a rule? Or is
this a solid "it depends" or "naaah"?
All the best, thanks for any hints
--
David Kastrup
Hi folks,
Back in 2012 I scripted some post-installation customization for when I
would build a new Linux machine. I suspect that all of what I did is no
longer needed for today's systems. I'm writing to see if someone can
confirm that.
What I did came from these links:
https://wiki.linuxaudio.org/wiki/system_configuration#the_kernelhttps://wiki.linuxaudio.org/wiki/system_configuration#cpu_frequency_scalinghttps://wiki.linuxaudio.org/wiki/system_configuration#hardware_timers
My new system runs the 4.17.17 kernel along with the rest of Fedora-27.
What do you think? Do I need to taylor the new machine in the above ways?
Thanks!
P.S.: My script is below, for the curious.
#!/bin/sh
# $Id: 36-linux_musicians,v 1.4 2012/11/16 05:56:14 kevinc Exp $
DATE=`date +%F-%T`
# http://wiki.linuxmusicians.com/doku.php?id=system_configuration#the_kernel
sudo cp /etc/default/grub /etc/default/grub-${DATE}
sudo cp /boot/grub2/grub.cfg /boot/grub2/grub.cfg-${DATE}
# cp /etc/default/grub grub
sudo perl -p -i -e 's/(^GRUB_CMDLINE_LINUX=\"nomodeset.*rhgb) quiet$\"/$1
quiet threadirqs\"/' /etc/default/grub
sudo /sbin/grub2-mkconfig -o /boot/grub2/grub.cfg
#
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#cpu_frequen…
# echo -n performance \
# | sudo tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
sudo touch /etc/rc.d/rc.local
cp /etc/rc.d/rc.local rc.local_cpufreq
cat << EOF >> rc.local_cpufreq
#!/bin/sh
for cpu in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
do
echo -n performance > \$cpu
done
echo 3072 > /sys/class/rtc/rtc0/max_user_freq
EOF
sudo cp rc.local_cpufreq /etc/rc.d/rc.local
sudo chmod 755 /etc/rc.d/rc.local
#
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#hardware_ti…
/bin/rm -f 40-timer-permissions.rules
cat << EOF > 40-timer-permissions.rules
KERNEL=="rtc0", GROUP="audio"
KERNEL=="hpet", GROUP="audio"
EOF
sudo cp 40-timer-permissions.rules /etc/udev/rules.d/
sudo chmod 644 /etc/udev/rules.d/40-timer-permissions.rules
cat << EOF > 60-max-user-freq.conf
# High Precision Event Timer for performance audio/MIDI
dev.hpet.max-user-freq=3072
EOF
sudo cp 60-max-user-freq.conf /etc/sysctl.d/
sudo chmod 644 /etc/sysctl.d/60-max-user-freq.conf
cat << EOF > 60-vm.swappiness.conf
# Delay start of use of swap partitions
vm.swappiness = 10
EOF
sudo cp 60-vm.swappiness.conf /etc/sysctl.d/
sudo chmod 644 /etc/sysctl.d/60-vm.swappiness.conf
cat << EOF > 93-audio_limits.conf
# Increase priority of audio applications
# # maximum realtime priority
@audio - rtprio 90
# maximum locked-in-memory address space (KB)
@audio - memlock 2000000
EOF
sudo cp 93-audio_limits.conf /etc/security/limits.d/
sudo chmod 644 /etc/security/limits.d/93-audio_limits.conf
exit
Hi folks,
Back in 2012 I scripted some post-installation customization for when I
would build a new Linux machine. I suspect that all of what I did is no
longer needed for today's systems. I'm writing to see if someone can
confirm that.
What I did came from these links:
https://wiki.linuxaudio.org/wiki/system_configuration#the_kernelhttps://wiki.linuxaudio.org/wiki/system_configuration#cpu_frequency_scalinghttps://wiki.linuxaudio.org/wiki/system_configuration#hardware_timers
My new system runs the 4.17.17 kernel along with the rest of Fedora-27.
What do you think? Do I need to taylor the new machine in the above ways?
Thanks!
P.S.: My script is below, for the curious.
#!/bin/sh
# $Id: 36-linux_musicians,v 1.4 2012/11/16 05:56:14 kevinc Exp $
DATE=`date +%F-%T`
# http://wiki.linuxmusicians.com/doku.php?id=system_configuration#the_kernel
sudo cp /etc/default/grub /etc/default/grub-${DATE}
sudo cp /boot/grub2/grub.cfg /boot/grub2/grub.cfg-${DATE}
# cp /etc/default/grub grub
sudo perl -p -i -e 's/(^GRUB_CMDLINE_LINUX=\"nomodeset.*rhgb) quiet$\"/$1
quiet threadirqs\"/' /etc/default/grub
sudo /sbin/grub2-mkconfig -o /boot/grub2/grub.cfg
#
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#cpu_frequen…
# echo -n performance \
# | sudo tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
sudo touch /etc/rc.d/rc.local
cp /etc/rc.d/rc.local rc.local_cpufreq
cat << EOF >> rc.local_cpufreq
#!/bin/sh
for cpu in /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
do
echo -n performance > \$cpu
done
echo 3072 > /sys/class/rtc/rtc0/max_user_freq
EOF
sudo cp rc.local_cpufreq /etc/rc.d/rc.local
sudo chmod 755 /etc/rc.d/rc.local
#
http://wiki.linuxmusicians.com/doku.php?id=system_configuration#hardware_ti…
/bin/rm -f 40-timer-permissions.rules
cat << EOF > 40-timer-permissions.rules
KERNEL=="rtc0", GROUP="audio"
KERNEL=="hpet", GROUP="audio"
EOF
sudo cp 40-timer-permissions.rules /etc/udev/rules.d/
sudo chmod 644 /etc/udev/rules.d/40-timer-permissions.rules
cat << EOF > 60-max-user-freq.conf
# High Precision Event Timer for performance audio/MIDI
dev.hpet.max-user-freq=3072
EOF
sudo cp 60-max-user-freq.conf /etc/sysctl.d/
sudo chmod 644 /etc/sysctl.d/60-max-user-freq.conf
cat << EOF > 60-vm.swappiness.conf
# Delay start of use of swap partitions
vm.swappiness = 10
EOF
sudo cp 60-vm.swappiness.conf /etc/sysctl.d/
sudo chmod 644 /etc/sysctl.d/60-vm.swappiness.conf
cat << EOF > 93-audio_limits.conf
# Increase priority of audio applications
# # maximum realtime priority
@audio - rtprio 90
# maximum locked-in-memory address space (KB)
@audio - memlock 2000000
EOF
sudo cp 93-audio_limits.conf /etc/security/limits.d/
sudo chmod 644 /etc/security/limits.d/93-audio_limits.conf
exit
Greetings,
Recently my computer rebooted after a power outage and after having an
uptime of many months. I'm running Fedora 23 here, with Fernando's rt
kernel from Planet CCRMA, on hardware that includes an AMD FX6300, 16G
memory, a big hard drive, and nVidia graphics. For those months of
uptime I've enjoyed a smooth-running system with very few issues. The
CPU runs in performance mode, and as far as I can tell the system has
been sweetly optimized for realtime audio.
Alas, after rebooting I now have extremely annoying issues with noise
from my wireless mouse when running VCV Rack, which is a problem I never
had before the reboot. I suspect IRQ assignment, but I'm awfully rusty
at troubleshooting so I'm asking the group for advice.
My checklist so far:
1. Running the correct kernel ?
Linux The6300 4.6.7-200.rt14.1.fc23.ccrma.x86_64+rt #1 SMP PREEMPT RT
Sat Oct 1 16:06:12 PDT 2016 x86_64 x86_64 x86_64 GNU/Linux
2. CPU in Performance mode ?
'cat /proc/cpuinfo' reports 'cpu MHz: 3500.000' for all cores
3. rtprio set ?
$ rtirq status
PID CLS RTPRIO NI PRI %CPU STAT COMMAND
169 FF 70 - 110 0.0 S irq/8-rtc0
748 FF 69 - 109 0.0 S irq/19-snd_hda_
764 FF 69 - 109 2.0 S irq/20-snd_ice1
166 FF 68 - 108 0.0 S irq/1-i8042
85 FF 50 - 90 0.0 S irq/9-acpi
138 FF 50 - 90 0.0 S irq/22-ahci[000
150 FF 50 - 90 0.0 S irq/17-ehci_hcd
151 FF 50 - 90 0.0 S irq/19-ehci_hcd
153 FF 50 - 90 0.0 S irq/16-ohci_hcd
154 FF 50 - 90 0.0 S irq/16-ohci_hcd
155 FF 50 - 90 0.0 S irq/18-ohci_hcd
156 FF 50 - 90 0.0 S irq/18-ohci_hcd
159 FF 50 - 90 0.0 S irq/18-ohci_hcd
430 FF 50 - 90 0.0 S irq/14-pata_ati
431 FF 50 - 90 0.0 S irq/15-pata_ati
706 FF 50 - 90 0.0 S irq/7-parport0
1062 FF 50 - 90 0.0 S irq/26-enp2s0
1178 FF 50 - 90 0.8 S irq/27-nvidia
...
Permissions are correct, and I have already tried switching the wireless
dongle to a different USB port. So what am I missing, what have I
forgotten ? The machine is the same as before the reboot, the kernel is
the same, and as far as I can tell the audio system is the same as it
ever was. I'm at a bit of a loss here, any and all help will be
appreciated.
Best regards,
dp
Hi,
I am a user of Archlinux ARM with Raspberry pi3. When starting JACK, the
following error appears.
Failed to connect to session bus for device reservation: Unable to
autolaunch a dbus-daemon without a $DISPLAY for X11.
To bypass device reservation via session bus, set
JACK_NO_AUDIO_RESERVATION=1 prior to starting jackd.
I attempted to append JACK_NO_AUDIO_RESERVATION=1 to jackdrc, but jackdrc is
missing anywhere. Please tell me how to avoid errors.
Minie
--
Sent from: http://linux-audio.4202.n7.nabble.com/linux-audio-user-f5.html
Hello all,
Updates are available at
<http://kokkinizita.linuxaudio.org/linuxaudio/downloads/index.html>
for all of the following:
aeolus-0.9.7
ambdec-0.7.1
clthreads-2.4.2
clxclient-3.9.2
ebumeter-0.4.2
jaaa-0.9.2
jack_delay-0.4.2
jack_utils-0.0.1
japa-0.9.2
jconvolver-1.0.2
jkmeter-0.6.5
jmatconvol-0.3.3
jmeters-0.4.4
jnoisemeter-0.2.2
octofile-0.3.2
tetraproc-0.8.6
yass-0.1.0
zita-ajbridge-0.8.2
zita-alsa-pcmi-0.3.2
zita-at1-0.6.2
zita-bls1-0.3.3
zita-convolver-4.0.2
zita-dpl1-0.3.3
zita-jclient-0.4.2
zita-lrx-0.1.2
zita-mu1-0.3.3
zita-njbridge-0.4.4
zita-resampler-1.6.2
zita-rev1-0.2.2
For most of them this is just small bug fixes, general maintenance,
and above all a systematic cleanup of the Makefiles.
Also the jacktools python package (presented at LAC 2018) is
available now. I actually did upload the files just before
the conference, but forgot to update index.html...
Ciao,
--
FA
Ah, a couple of responses while running an errand! Here goes:
Francesco:
i run aeolus through jack. i attempted to run it through alsa, but was
told that aeolus could not connect to alsa. This happened when setting the
sample rate to both 44100 and 48000.
It turns out that stops is installed, though it is the version that came
with the distro. it just never occurred to me that stops and aeolus would
be installed as separate programs. Could it be that the newer compiled
version is butting heads with this?
David:
The architecture is 64bit. My apologies for not mentioning that. My
projects are not quite in order enough for me to upgrade to 18.04, though
it was my plan to do so in the next few weeks.
That's what i was thinking: The path to definitions is not clearly
defined. i've never learned the format for setting those paths. i've half
a head to uninstall stops and recompile the newer version to see if that
solves anything. Given the morning's plans, though, i'm a few hours away
from that....
Thank you, Francesco and David, for your responses.
tom
Well, well, well, Francesco, that fixed the problem immediately! i've
spent the last few minutes listening to a Bach fugue in a variety of
temperaments. Aeolus doesn't remember the path, but i can jigger with a
script. This can now be considered solved.
Thank you so much for your help!
tom