Linux-audio-dev

linux-audio-dev@lists.linuxaudio.org

4 participants
7538 discussions

[linux-audio-dev] ANN: set_rtlimits for PAM-less systems and perhaps others

by Jonathan Woithe

With the recent adoption of the realtime rlimits patch into the mainline kernel, people have needed a way of utilising this for their music software. There is apparently a PAM module out there which can be used if one's system uses PAM, but for people on Slackware (which doesn't use PAM) - and perhaps other distributions as well - this is not an available option. set_rtlimits is a small program I hacked up over the weekend to allow controlled non-priviledged access to realtime scheduling through the new RTPRIO resource limit. set_rtlimits needs to be setuid root but it only runs at elevated priviledges when setting the resource limits. Furthermore, the users who are permitted to use set_rtlimits are controlled through a hardcoded central configuration file (/etc/set_rtlimits.conf), as are the programs those users can run via set_rtlimits and the maximum realtime and nice priorities the users can request for each program. Programs must be specified using absolute paths, so a malicious user can't just run their own ardour binary, for example. There is no homepage for this yet; simply grab the source from http://www.physics.adelaide.edu.au/~jwoithe/set_rtlimits-1.0.0.tgz Documentation is by way of the included README and manpage. Note that this is not polished software. It's not autoconf-ised, although hopefully it won't require too much tweaking to compile on a variety of systems. It probably isn't written in a very portable way, and although it poses no security issues to the best of my knowledge (beyond the usual issues with setuid root programs and realtime priorities) it has not been extensively audited. Having said that it gets the job done for me and might be useful for others. My time is limited at present which is why the rough edges remain. Development was done under Slackware 10.0 running kernel 2.6.12-rc5. Using this program I have successfully run Ardour 0.9beta29 and muse 0.7.2pre1 with Jack 0.99.73 under 2.6.12-rc5 for short periods. Bug reports, improvements, suggestions and patches are welcome; send to the email address included in the tarball documentation. Regards jonathan

20 years, 8 months

[linux-audio-dev] OT: Survey facing design patterns and communication

by Nico

Hello everybody! We are a group of students at "Freie Universitaet Berlin". As part of our computer science studies we are going to do a survey facing the use of design patterns in communication. Examples of design patterns are "Abstract Factory", "Singleton", "Composite", "Iterator" and "Listener". If you know what we are talking about, you are welcome to take part in our survey. It takes about 5 minutes to fill out the form. Just jump to: http://study.beatdepot.de If you agree, we will send you the results of our survey. Thanks in advance for your participation! And sorry for the interruption of your discussion.

20 years, 8 months

[linux-audio-dev] Algorithm for displaying waveforms

by Them

For a small application I am developing, I need to display waveforms (read-only) from audio data loaded into my application using libsndfile. My question is about displaying a waveform in a GUI window -- I have searched around the net looking for general algorithms but haven't seen anything that describes this. I haven't tried coding this yet, so I am asking if someone can describe the basic algorithm or point me in the right direction to a book, web site, the source file in Ardour or whatever. Even just a high-level description of the basic algorithm would be good -- I am thinking somewhere along the lines that I am taking the values of samples and drawing line segments from point to point. Thanks in advance! -- Brett

20 years, 8 months

[linux-audio-dev] 30db loudness difference at these cases

by Andrew Gaydenko

Hi, All! I'm rather frustrated with my situation, so I'm not sure which mailing list is the most appropriate for my question. I have Terratec Aureon 7.1 Space card which is used with ICE1724 alsa driver (1.0.8, 1.0.9a). An appropriate .asoundrc fragment is shown below. All works fine except for at some cases a sound level is about 30db lower rather at another cases. Of course, alsamixer settings are the same. --- "Normal" cases are: --- 1. Playing back with aqualung this way: aqualung -o alsa -d default 2. Using any JACK-enabled app (ReZound, aqualung) --- "-30db" cases are: --- 1. aplay -d default <file> 2. xine engine with any frontend (amaroK, Kaffeine) Will anybody be so kind to suggest steps to find this difference reason? Andrew ////////////////// .asoundrc fragment //////////////////////////// pcm.!default { type plug slave { pcm "2x4" format S32_LE } } pcm.2x4 { type route slave.pcm surround71 slave.channels 8 ttable.0.0 0.05 ttable.1.1 0.05 ttable.0.2 1 ttable.1.3 1 ttable.0.4 1 ttable.1.5 1 ttable.0.6 1 ttable.1.7 1 }

20 years, 8 months

[linux-audio-dev] Objective-C (was: gcc, you let me down...)

by Toby

Tim Goetze wrote: > I'm pretty much sold on Python as my high-level language of choice and > very reluctant to diversify in computer language literacy any further. I feel your pain. Python is by far my favorite language ever. However, I've recently been looking for an alternative *compiled* object-oriented language, because let's face it, Python is on average 10 times slower than C. Sometimes you just can't afford it. Enter Objective-C: - STRICT SUPERSET OF C: every valid C program is a valid ObjC program. This makes it trivial to include or link to C code and libraries and to mix procedural, object-oriented and ASM code in the same *file*. - SIMPLE: ObjC is plain C with one syntax addition and a few new keywords. It only extends the C language to support Smalltalk-like object-oriented features, because that's all you're going to need. No more operator overloading, templates, references, 'const', etc. - DYNAMICALLY TYPED: messages (method calls) are delivered according to the dynamic type of the target object, not to some static type. This is how Python works. You can even send an object a message that is not specified in its interface. This might seem like a bad idea, but instead it allows for powerful delegation-based design patterns. - FAST: Objective-C performs dynamically bound message calls very quickly, about 1.5-2.0 times as long as a plain C function call! Objective-C is the language of choice for MacOS X development. GCC compiles it very well too. Using GNUstep (optional) as the system and GUI framework, you can make a GUI program that compiles almost without changes on both GNU/Linux and MacOS X! (Windows port in the working.) > > Well I really like to separate C and C++. C is unashamedly a low > > level language. C++ OTOH tries to be both low level and high level. > > In comparison to C, C++ is a poor low level language. Compared to > > Python or Ocaml, C++ is a poor high level language. Objective-C can be as low-level as C (including #define, ASM...) and as high-level as Python (albeit a bit more verbose) IN THE SAME FUNCTION! Toby -- One theory states that if anyone ever learns how to use all of Emacs, it will instantly disappear and be replaced by something even more bizarre and inexplicable. Another theory states that that's how VI was invented.

20 years, 8 months

[linux-audio-dev] [ANN] Digital Room Correction 2.6.0

by Denis Sbragion

Hello, Digital Room Correction 2.6.0 is available at: http://freshmeat.net/projects/drc Changes: A new prefiltering curve based on the bilinear transformation has been introduced. An improved windowing of the minimum phase filters used to apply the target frequency response and the microphone compensation has been implemented. A missing normalization of the minimum phase correction filter has been added. A new logarithmic interpolation has been added to the target transfer function computation. The new interpolation method simplifies the definition of the target transfer functions. Small improvements to the documentation and to the Octave scripts used to generate the graphs have been applied. A new improved version of the measurejack script has been included in the package. Some new sample configuration files, including one approximating the ERB psychoacoustic scale, have been added. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it

20 years, 8 months

[linux-audio-dev] [ann] caps-0.2.3

by Tim Goetze

The ever popular CAPS Audio Plugin Suite reincarnates as v0.2.3, a maintenance release that rectifies the last remaining denormal problems and restores the intermittently nonfunctional AmpIV gain control to its usual fine form. CAPS is a LADSPA library that enjoys worldwide favour for its high-quality instrument amplifier emulation plugins; in addition it provides a small but no less sophisticated assortment of DSP units for daily use as well as some more exotic sound generators. Upgrading is recommended; grab your copy before they are all gone: http://quitte.de/dsp/caps.html http://quitte.de/dsp/caps_0.2.3.tar.gz Please forward as you see fit. Enjoy, Tim

20 years, 8 months

[linux-audio-dev] Acoustic echo canceller released

by Jean-Marc Valin

Hi, I thought this could be of interest to some people here. I have just released an acoustic echo canceller as part of Speex 1.1.9 (http://www.speex.org/). So far, I have tested it on 8 kHz, but it should work for other sampling rates. It uses the MDF algorithm and (optionally) residual echo cancellation in the spectral domain. If anyone is interested in making an ALSA plugin out of it, I can provide assistance. Oh, and it's released under the (revised) BSD license. Jean-Marc P.S. In case some noticed, there was an echo canceller in Speex before, but this is the first version to actually work in real conditions (double talk and all). -- Jean-Marc Valin <Jean-Marc.Valin(a)USherbrooke.ca> Université de Sherbrooke

20 years, 8 months

[linux-audio-dev] photo needed

by Dave Phillips

Greetings: I've prepared a brief report on LAC 2005 for the Linux Journal, it's ready for submission but I need an outside photo of ZKM + the Kubus. Did anyone take a nice shot of the buildings that they'd like to see in LJ ? If so, let me know asap. A TIFF is preferred, but high-resolution JPG will probably do. TIA! Best, dp

20 years, 8 months

[linux-audio-dev] FreeBSD 5.4-RELEASE

by Chris Camisa

Hello! Sorry if I'm starting in the wrong place, but after several months of thinking and two weeks of working, I have a couple of questions. 1.) can FreeBSD 5.4-RELEASE operate without a sound card such that the Network Audio Server allows applications running on it to have sound heard on other speakers on the network? a.) I have "device sound" compiled into the custom kernel b.) the NASD gives an error about connecting to a block device when I try to start NASD. 2.) is arts the way to go with ALSA? When KDE starts on my 2.6 kernel Gentoo system the sound suddenly get louder, as if a new mixer takes over and bumps the master volume as KDE 3.3 loads. 3.) What is the preferred method to have multiple x86 computers playing the same stream of sound simultaneously? (within a few dozen milliseconds) a.) all the servers have NTP capability and mplayer/xine 4.) How do I have several x86 FreeBSD/linux machines all synchronize video also? Is that simply XF86 Forwarding? Yours Truly, Christopher

20 years, 8 months

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Linux-audio-dev