Clemens Ladisch hat gesagt: // Clemens Ladisch wrote: Would nrpacks=1 also be a good option to set for USB-midi devices? Or would it at least not do any harm?

Hi all... I am going to want a compressor limiter and some microphones but am quickly coming short on cash... any recomendations on a good and cheap compressor/limiter I can put on my vocal and bass signal paths to clean up my bad microphone technique and my really shitty bass?? Nick

Due to how the USB protocol works, all USB audio devices have the same latency. You should be able to get lower latencies by loading the snd-usb-audio driver with the "nrpacks=1" parameter, and using three periods in Jack.

Thanks for this worthy hint. Unfortunately, jack will not start with this setting on my card.

Probably this (I have the same card): jackd 0.99.54 Copyright 2001-2005 Paul Davis and others. jackd comes with ABSOLUTELY NO WARRANTY This is free software, and you are welcome to redistribute it under certain conditions; see the file COPYING for details JACK compiled with System V SHM support. loading driver .. creating alsa driver ... hw:1|hw:1|1024|2|48000|0|0|nomon|swmeter|-|32bit control device hw:1 configuring for 48000Hz, period = 1024 frames, buffer = 2 periods Note: audio device hw:1 doesn't support a 32bit sample format so JACK will try a 24bit format instead Note: audio device hw:1 doesn't support a 24bit sample format so JACK will try a 16bit format instead nperiods = 2 for capture Note: audio device hw:1 doesn't support a 32bit sample format so JACK will try a 24bit format instead Note: audio device hw:1 doesn't support a 24bit sample format so JACK will try a 16bit format instead nperiods = 2 for playback ALSA: could not start playback (Broken pipe) DRIVER NT: could not start driver cannot start driver

BTW: Is there a way to tell JACK to use 16bit so JACK doesn't need to find it by trial and error?

ALSA: could not start playback (Broken pipe)

kernel: cannot submit datapipe for urb 0, err = -28

Clemens Ladisch hat gesagt: // Clemens Ladisch wrote: Is there a way to disable bandwidth checking at runtime, that is, without recompiling the kernel and using some module option or similar?

On Wed, 2005-06-15 at 13:58 +0200, Clemens Ladisch wrote: Sounds like a good candidate for a bug report to debian.

I thought so, too, but now I found out, that my kernel image package actually is not part of Debian official, but of some unofficial Apt repository. I will go on and compile my own kernel 2.6.11 later.

kernel.org kernels are now enabling EXPERIMENTAL options by default? Doubtful.

Christoph Eckert hat gesagt: // Christoph Eckert wrote: To my knowledge, I didn't load a config file anywhere, I just did "make gconfig" and went through most of the settings by hand. There I found USB BW alloc. enabled.

Hi all, For those who seem to be interested in UBUNTU/DEMUDI, here are some steps that I resolved and some others where I'm blocking. I'm new in Linux/audio. For a first time with I would like to record my bass with rosegarden and play drums with hydrogen. My card : maudio 1010 LT which don't have Synth Support under ALSA $kinfocenter Synth devices : NOT ENABLED IN CONFIG In the file /etc/apt/sources.list, add : deb http://demudi.agnula.org/packages/demudi stable main In a terminal : $apt-get update $apt-cache search kernel $apt-get install kernel-image-2.6.12-2-multimedia-386 $apt-get install kernel-patch-2.6-realtime-lsm In a terminal : $lsmod |grep snd You need to have : snd_seq_midi .......... OK snd_seq_midi_event .... OK snd_seq ............... OK snd_rawmidi ........... OK snd_timer ............. OK snd ................... OK soundcore ............. OK snd_virmidi realtime If you don't have one, execute "modprobe" in a terminal. For example : $modprobe snd_virmidi FATAL: Error inserting snd_virmidi (/lib/modules/2.6.12-2-multimedia-386/kernel/sound/drivers/snd-virmidi.ko): No such device It was OK with ubuntu kernel !!! $modprobe realtime any=1 $apt-cache search alsaconf $apt-get install alsa-utils Then you have $alsaconf $alsamixer $alsactl $apt-cache search jackd $apt-get install jackd $apt-cache search qjackctl $apt-get install qjackctl Set qjackctl with jackd, not with jackstart with jackstart : #12:12:42.618 Patchbay deactivated. #12:12:42.906 Statistics reset. #12:12:42.980 MIDI connection graph change. #12:12:43.114 MIDI connection change. #12:12:45.920 Startup script... #12:12:45.921 artsshell -q terminate #12:12:46.252 Startup script terminated with exit status=256. #12:12:46.252 JACK is starting... #12:12:46.253 /usr/bin/jackstart -R -dalsa -dhw:0 -r48000 -p2048 -n2 #12:12:46.256 JACK was started with PID=7553 (0x1d81). #jackstart: cannot get realtime capabilities, current capabilities are: # =ep cap_setpcap-e # probably running under a kernel with capabilities disabled, # a suitable kernel would have printed something like "=eip" #12:12:46.263 JACK was stopped with exit status=1. #12:12:48.268 Could not connect to JACK server as client. to use rosegarden, I need I import some hydrogen samples in rosegarden, and I don't have any sound. Does somebody have an idea ??? May be this is the virmidi module which is not loaded ? Best regards, Benjamin

I thought so, too, but now I found out, that my kernel image package actually is not part of Debian official, but of some unofficial Apt repository. I will go on and compile my own kernel 2.6.11 later.

Hi Frank. Using a dodgy kernel eh!

I've being viewing your posts , and sort of wanting to help, scanned the synaptic list. I had 2.4.27 kernel and 2.6.8. Having now found

also that the 2 modules I've mentioned should really be loaded as default. Just giving my 2 cents worth. Hope you get it fixed. Nigel.

Latency: 4msec, still no xruns with alsaplayer (I even switched to a virtual console... ;)

I'd like to express my deepest thanks that you shared your worthy knowledge with us. I guess Frank Barknecht will be happy as well :) .

http://alsa.opensrc.org/index.php?page=Alsa+Preferred+Soundcards http://alsa.opensrc.org/index.php?page=TerratecAureonUSB5.1

Did the default behavior change recently?  If so this is a huge user-visible regression, and we need to make the kernel people aware before 2.6.12 and so it can be fixed in the stable (2.6.11.x) series.

Hi ce, On Sunday 12 June 2005 12:54, Christoph Eckert wrote: I am using a tascam us122. Its usb1.1, has two ins and two outs, phantom power for the XLR-inputs. It isn't really the cheapest (~200€) but worth its price. I run it successfully with latencies <10ms, even without special parameters for the driver.