by Austin Acton
On the eve of Mandrake's latest release, Austin Acton has published a guide to create a digital audio workstation using Mandrake 9.1. This 'HowTo' describes how to set up a professional quality audio workstation in an afternoon or less without compiling or text editing . . .
The Mandrake Audio Workstation HowTo
Summary
Beginners can skip this section. Experts can ignore almost everything else.
You can setup a professional quality audio workstation in an afternoon or less, with Mandrake Linux. No compiling. No text editing. No dependencies. It's this easy.- Use the ALSA drivers. They are all name snd-. Use draksound or mcc to do so.
- Setup a contrib repository using:
# urpmi.setup
(if you don't have urpmi.setup installed, you'll have to do that first with:
# urpmi urpmi.setup )
- Install your studio with one command:
# urpmi jackit-realtime ardour audacity alsamixergui swh-plugins cmt
- If you run KDE, you have an extra step
# urpmi kickarts
- Reboot into the multimedia kernel (it will have the suffix “mm”)
- If you're using KDE, add kickarts to your kicker panel, and diable arts with it.
- To use audacity, go ahead and run it from the menu.
To use ardour, start jack by executing:
$ jackstart -d alsa -d hw:0
then run ardour from the menu. You can monitor your system's latency in the jack window.
- Thank Mandrake and/or the audio application authors with a small donation or by helping the project in some other way.
Why GNU/Linux?
The reasons to chose GNU/Linux are many. The price is right. You have control over every element of building your system, building your applications, and modifying them as you desire. You have input into the development process. You can even adopt the code into your own application if you wish.
The main reason though, is that it's the right thing to do. Software should be free. You should know what's going on – what programs are doing. You should have control of your system and a voice for the future of the applications you use. Nobody should have a monopoly on the tools you need to get work done – be it an office suite, a web browser, an audio editor, or the operating system itself. There's power in having a choice.
If this rant doesn't convince you or fulfill your needs for information, see www.gnu.org.
Why Mandrake Linux?
There are many distributions to choose from. Mandrake offers simplicity and full-featured power in one package. There's no need for a specialty distribution, as Mandrake can easily be tailored to one task or a full desktop operating system. It's free, it has a large community of supporters and developers, and all of the packages in Mandrake Linux are freely distributable and open source. Urpmi allows you to install the applications you need, without the dependency problems often associated with RPMs, and you can keep your system up-to-date with a single command.
The main reason I've chosen it for an audio workstation is because you can set it up with little to no unix expertise, often with little to no trouble, and you can easily help improve the next version by financial or developmental contributions.
See Mandrake's website for more information.
Choosing Hardware
As far as power goes, you can get by with a simple system, but more is almost always better. You will need at least a Pentium processor, 64 MB ram, and a good hard drive to run Mandrake. If you want to do pro-audio recording, I'd suggest at least a Pentium III or Athlon processor, lots of good ram (128 MB minimum), and a big, fast (7200 RPM) hard drive. While SCSI drives are generally more powerful and longer lasting than EIDE drive, the cost pretty much offsets any benefits you'll actually see, so I'd suggest a bigger, faster EIDE drive to an equally priced but smaller SCSI drive.
For example, I have one system with a Celeron 750, 256 MB PC100 RAM, and a 5000 RPM drive. With it, I can record three or four mono tracks in full duplex, but no more. With my other system, a dual Athlon 1900, 512 MB DDR ram, and a 7200 RPM drive, I can record many stereo tracks in full duplex without artifacts. This should give you a rough idea of what you'll need.
Most core hardware (video, CD-RW, etc.) is now supported under linux as long as it's not extremely new technology. Printers are about 50% supported, so see this link for details. Modems are fully supported as long as they are hardware modems, not soft-modems. Ask your supplier/manufacturer for info on that.
If you're really into linux, you WILL need a good internet connection. Anything faster than a modem (ISDN, ADSL, cable, lan) is fine. A modem connection will possibly suit your needs, but one tires of it VERY quickly.
The Sound Card
Obviously, you'll need a sound card with a linux driver. The first rule of thumb is to avoid all USB devices like the plague. They have much larger processing overhead, the drivers lag behind PCI equivalents, and they are designed for convenience, not power. Most PCI cards have linux drivers, and they are very affordable. Some cards offer multiple inputs, external inputs, and external DAC (digital/analogue converter). These are all of benefit to those recording live sound samples, but are a waste of money if your work is mostly MIDI/synthesis based. The best place to check for hardware support is www.alsa-project.org, since that's the sound system you'll be using (hopefully). See the 'sound card matrix' for supported sound cards and sound chips.
Really, most sound cards offer adequate sound quality to do home recording. Many motherboards come with a sound chip installed. These are hit-and-miss. My A-Open motherboard with a VIA chip is terrible. Every movement of the mouse or activity of the hard drive causes noise. However, my Asus motherboard with a C-Media chip is amazing. The sound is crystal clear. If in doubt, disable the on-board chip through your BIOS setup, and get a PCI sound card. It's best to install it in the slot furthest from your video card and hard drive, and these can both cause sound interference.
As a brief guideline, chose any card that's fully supported by ALSA. Creative Labs, ESS, VIA, and Intel are popular choices. Stay away from Aureal based cards (Vortex, some Turtle Beach, some Diamond), SiS, Riptide, Digital Audio Labs, and Echo Corp cards (Darla, Gina, Layla). For more professional quality cards, check out Ensoniq, Midiman (not all supported), Yamaha (not all supported), Roland, or better yet, RME. It's important to let these companies know that linux is an important market to them. If drivers are not available from ALSA, please encourage the card manufacturer to release its specs to the ALSA team so they can develop a driver for it.
Also, if you're recording live audio, be sure your card supports true full-duplex mode. This means it can playback one track while you record another. In other words it has two D/A converters. Some cards (mainly ESS/Creative cards) try to fake full-duplex by splitting the D/A converter's capabilities in half. This mean you can record in full duplex, but an half the original sound quality. You don't want this, since many affordable cards support true full duplex.
If you're doing serious production work, read the Linux Audio-Quality HowTo. It is full of good information on studio hardware setup, but beware: much of the cards/drivers section is out of date.
Getting the Software
Most of the software you'll need is included with Mandrake Linux. You can get the core operating system by one of three ways: buy the CD's, download the CD's, or install over a network. Buying the CD boxed set is the traditional, comfortable choice for most, but isn't necessary or beneficial really. Your money is better spent going straight to the company. Consider joining MandrakeClub instead.
For a simple solution, download all three Mandrake Linux 9.1 CD iso files from your nearest mirror. For a list of mirrors go here. Each ISO is about 650 MB, so this may take a while. Be sure to use an ftp program which can stop and resume, just in case. Burn the iso files to three CD's.
Install
Installing Mandrake Linux is discussed on the CD itself, online, and in elsewhere in print.
Partitioning your drive(s) is important for an audio workstation. In the ideal case, you'll have a second hard drive for audio scratch space only. Two would be even better. If you only have one drive, I'd suggest allocating 3 GB for the root partition, and the rest for the /home partition. That way, you can delete the OS at any time (accidentally or not) and re-install without losing any of your personal files.
You'll have to chose a filesystem type. Ext3 is by far the most reliable, but XFS gives the best performance for large files. I'd suggest ext3 for the root and home partitions for sure. If you're doing mostly MIDI or small audio file work a scratch partition is not really necessary. If you're doing serious live recording, will want to use a third partition or a second hard drive for scratch space, and use an XFS file system. Just be sure to tell any audio applications to store scratch data somewhere in your home directory (/home/austin/tmp, /home/austin/.audacity, or similar), on your scratch partition (/tape, /scratch, /audio, or whatever you design), and not on the root partition (/tmp, or similar).
You'll need to install your desktop of choice. If you have a fast computer, I'd suggest GNOME or KDE as they're very popular, full-featured, and easy to use. If you have a slower computer, you'll get better performance using a lighter desktop environment. Consider Blackbox, or IceWM.
Setting Up Online Software Sources
Obviously not all the linux software in existence can be included on three CD's. You will need to access some applications online. Most of the apps you will need are going to be on the CD's, or in the Mandrake Contributions section online.
To access the contributions section, run:
# urpmi urpmi.setup
# urpmi.setup
and select a 'contrib' mirror close to you. It's that easy.
Be aware that contrib applications are not directly supported by MandrakeSoft, because they are packaged by volunteers. You can still report bugs here, but Mandrake makes no definite claim that they will support them or maintain them.
On kernels, latency, and security
The biggest problem with the linux kernel is that it was designed for server-type applications. Thus it likes to divide processing time between various users and applications. This is not good for multimedia applications. Audio must be captured on-time, and written to disk as soon as the memory buffer is full. If the kernel or hardware can't keep up with your application, you will hear dropouts (missing sound), pops (noise), or time changes in your recording.
To avoid this, you can install a specially patched multimedia kernel. This kernel is designed to avoid latency, and allow certain programs to take priority over your hardware. Obviously you should not use the multimedia kernel on a server or an unprotected network, as there is a small security risk involved. For the average workstation though, you need not worry.
To install the multimedia kernel, simply become root (use the su command), and enter:
# urpmi kernel-multimedia
Urpmi will also give you the choice of an smp kernel. Choose the smp kernel if you have a dual processor machine. (If you have to ask, you don't)
On sound systems and sound daemons
When you install Mandrake, it should autodetect your card (assuming it's a PCI card), and ask you which driver to use. This is where linux audio gets really confusing. First I'll try to explain the madness, then I'll give you a suggestion to keep you life simple. If you're not into technical details, skip down to the 'simple solution'.
The original sound card drivers for linux are referred to as OSS/Free, OSS/Lite, or OSS/Kernel. These are all the same thing. They are part of the linux kernel, and have been for years. The most common cards are well supported, and virtually all audio software knows how to interface with these drivers. There is also a commercial version of these drivers with some extra features created by www.opensound.com. They support more cards than OSS/Free, and the company offers support for getting them installed and working.
The main limitation with OSS is that it was developed to send one audio stream to your sound card. This is quite simple in linux. Your sound application simply begins writing data to /dev/dsp, and OSS takes care of making your sound card play it. If two applications begin writing to /dev/dsp, the kernel tells the second application to wait until the first one is finished. This can be quite strange for a new user. I remember using ICQ and listening to MP3's at the same time. While an MP3 was playing, GnomeICU would try to chime every time I received a message, but /dev/dsp was busy with my music. As soon as the song was over, GnomeICU would be allowed access to /dev/dsp, and I'd hear twenty chimes all at once, and then the next MP3 would start. It was quite funny.
To avoid this problem, sound daemons were developed. (A daemon is just a background process that remains idle until a certain condition arises, like sound playing, or printing being sent to the printer, etc.) Sound daemons were originally developed to mix sound from two applications, and write the combination to /dev/dsp, allowing you to hear two applications at once. In traditional linux style, several daemons have evolved in parallel, and are not compatible. They have vastly different capabilities and ease of use. GNOME uses esd (the 'enlightened' sound daemon, or esound), KDE uses aRts. Esd has very few features, and thus is currently being replaced by asd (the 'advanced' sound daemon, asd.sourceforge.net), while arts has almost too many features (which are rarely used), and sometimes it gets a bit intrusive. Other desktops including IceWM have their own sound daemons too. It all gets very confusing for the pro audio user, but there is a simple solution.
More recently, a second free sound system has developed. The ALSA (advanced linux sound architecture) project stated to develop advanced drivers for the Gravis Ultrasound series of cards, but now supports roughly as many cards as OSS/Free does. Currently it is a separate project from the kernel, but as of linux 2.6, the ALSA drivers will be distributed with the kernel itself. ALSA offers many advantages over the OSS system. The main advantage is in setting up complex system with multiple sound cards, running multiple sound applications, or running real-time audio applications. There are all thing we're interested in.
Since ALSA is a totally new system of sound drivers, all applications would have to be rewritten to know how to communicate with the ALSA system. To avoid this problem, there is a compatibility layer which allows ALSA to receive instructions intended for OSS. Of course you lose some of the fancy features of ALSA doing things this way, but you're assured that older linux applications can still use your sound card.
ALSA is capable of writing multiple sound streams to your card if your card supports this. You can check for this ability by executing:
$ cat /proc/asound/card0/pcm0p/info | grep 'subdevices'
You'll see something like:
subdevices_count: 1
subdevices_avail: 1
This means my card can accept one stream at a time, and I have one stream currently available. If my card had multiple subdevices, I could listen to sound from two applications simultaneously without a sound daemon, by using the ALSA drivers alone. Since this is not the case, I still want a sound daemon for day-to-day operations.
Arts and esd have a different strategy too. Esd just sits around waiting for audio to be piped to it; it then mixes the sound and pipes it to your sound card. So as long as you're not piping anything to esd, it will sit inactive and allow you direct access to your card. Arts is a bit more intrusive though. It remains active all the time, and any audio being directed to your card, arts tries to process. This is a problem when you want direct access to your card. You can susupend arts using kickarts (a panel applet), or on the command line with:
$ artsshell suspend
ALSA has recently developed its own sound daemon called the Jack Audio Connection Kit (jack or jackit). Unlike esd and arts, it is intended mainly for audio production. Thus is allows for more configurations and low-latency performance. On the downside, it's trickier to configure, isn't setup automatically by Mandrake, doesn't emulate OSS, and isn't supported by many audio applications (yet).
Most simple audio applications use OSS by default. This is fine if you're using ALSA with OSS emulation, or arts, as it grabs that audio and processes it however you want. If you're using esd, the application will get direct access, and esd will be put aside.
Many applications have multiple output options. For example, you can install xmms plugins to pipe your music to alsa (xmms-alsa), esd (xmms-esd), arts (xmms-arts), or jack (xmms-jack). Which one to choose? Luckily, it doesn't really matter for simple applications. You won't notice any performance difference. Use whatever works, or whatever makes you feel good.
Now, if your head is spinning after that barrage of linux history and technical facts, go ahead and forget it all.
Simple Solution
Currently, this is the easiest way to have a full time or part time audio workstation. Mandrake will setup a fully functional desktop for day-to-day work, so we'll take advantage of that, yet make it as easy as possible to switch to pro-audio mode.
First of all, use ALSA! This is easy to setup during the installation or later through the Mandrake Control Centre or by running draksound. ALSA drivers are all named snd-. So if you are currently using a driver called 'cmipci', it's the OSS driver. Switch to the 'snd-cmipci' ALSA driver using draksound. This isn't Windows, so you don't have to reboot. :-)
Gnome Users
You can tell esd to start when GNOME starts if you want. Use the Mandrake menu and go to Configuration>GNOME>Sound and see 'Enable sound server startup'. If your workstation is only going to be used for production, you'll want to it disabled for simplicity. If you use your desktop for day-to-day operations, you'll probably want esd running all the time.
Esd is quite harmless, and most serious audio apps will just bypass it. If you're using GNOME, you can just ignore esd when you're doing audio production work.
KDE Users
In KDE, arts can be turned on or off by default. Use the KDE Control Centre to adjust the default setting. If your workstation is only going to be used for production, you'll want to leave it disabled for simplicity.
If you use your desktop for day-to-day operations, you'll probably want arts running by default. You can then use the kickarts panel applet to control arts. Install it with:
# urpmi kickarts
then add it to your kicker panel. When you want to do some serious audio production though, arts may get in the way, use up processing power, and get confusing. Use the kickarts applet to disable arts before you do any serious audio production work.
Everyone
Now 99% of simple linux audio applications will run without any fancy configuration necessary.
You'll need jack later. To install the low-latency version of jack, execute:
# urpmi jackit-realtime
Note that you won't be able to use the realtime option without running the multimedia kernel at boot-time. Otherwise you can still run jack, just not with realtime.
Production Time
When you want to do serious audio work, definitely choose ALSA or jack output if your application has an option to do so. KDE users, remember to diable arts.
If (and only if) your application has the ability to communicate with jack, start jack with:
$ jackstart -d alsa -d hw:0
See the jack and ALSA documentation for information on more complex system (multiple cards, or multi-input cards, etc.).
Applications for every need
There are tons of cool applications included with Mandrake linux. In the menu, select Configuration>Packaging>Install Software, view packages by group, and look in the 'Sound' category. Try a few or install them all.
The two audio workstation's you'll use the most are ardour and audacity. Install them along with some helper tools with:
# urpmi ardour audacity alsamixergui swh-plugins cmt
You'll probably want to use alsamixergui to control your sound card. It's in the menu, but you can run it from the command line of course. swh-plugins and cmt are collections of amazing audio effect plugins. You can use them, and any other ladspa compatible plugins with most linux audio tools simply by installing the rpm.
Ardour
Ardour is the pro audio system for linux. The author (Paul Davis) is also the author of jack, and a contributor to the ALSA project. So the ardour/jack/alsa combination works together seamlessly, and allows you to use any ALSA supported sound card (from $20 to $1000) to record multitrack pro-audio with.
VERY IMPORTANT: Please note that ardour is not finished. Everyone who uses it knows this, including the author. It is very usable, but some features are missing, and it is known to crash when something goes wrong. PLEASE do not pester the author or any other ardour developers with this. The developers know what the bugs are. Ardour changes every day, usually for the better. If you have problems getting it working, contact a Mandrake support group. If there are bugs with the menu/icons/file locations/etc., contact the Mandrake development team. Once again, this is free software, so the ardour developers do NOT want to hear you complain about bugs or missing features.
Ardour is modeled after a self-contained audio workstation like a multitrack tape recorder or a DAT or hard disk multitracker. If you're a computer geek and not an audio geek, the controls will be somewhat foreign, but they are easy to learn, and ardour has a lot of great documentation. Most of it you can find in /usr/share/doc, including some tutorials.
Ardour requires two scratch directories. These default to /tape1 and /tape2. With our setup, you'll want to remap these to you home partition or scratch partition which has lots for free space and uses an XFS filesystem. Create some scratch directories:
$ mkdir /scratch/{tape1,tape2}
If you are the only person using the workstation, you can simply redirect the default location to your scratch space location:
# ln -s /scratch/tape1 /tape1
# ln -s /scratch/tape2 /tape2
NB: If /tape1 and /tape2 already exist (installing ardour automatically creates them), you will have to remove them before you can make links to another directory:
# rm -fr /tape{1,2}
Alternatively, you can use the Preferences option in ardour to select any directories you want as the scratch space.
Starting ardour is easy. Start jack as explained previously, and run ardour from the menu or the command line. Before you can do any work, or even experiment with the program, you'll have to create a new session from the File menu.
After running it for the first time, you'll find the file 'ardour.rc' in your home directory. If you want to do anything fancy, like control ardour with a MIDI device, you'll have to edit ardour.rc. See the ardour documentation.
Note that ardour is designed with non-scaling pixmaps, and is designed for a very high resolution display. Thus, if you use it on a standard 1024x768 display, you won't be able to see the whole workstation on one screen. This is useable, but unpleasant. If you want the ideal case, use the Mandrake Control Centre to resize X to 1280x960. You'll need a decent monitor to do that.
Again, be warned that ardour is not mature software. It may crash from time to time. You can simply remember what made it crash, and don't do that anymore. ;-) Luckily, you can often save your session, even after the program has become very messed up. Watch for updates on MandrakeClub.
Audacity
Audacity is another multitrack recorder and wave editor, written by Dominic Mazzoni.
Audacity and ardour have many of the same intentions. Multitrack recording, mixing, effects, and editing. Audacity looks more like a traditional GUI, so if you're not an audio engineer, you may find it easier to use. It currently has a lot fewer features than ardour, the main one being that it uses OSS audio interface, so performance and sound quality will not be equal to ardour. Don't run jack; you'll want audacity and ALSA to communicate directly. Start audacity using the menu or the command line. The default scratch directory will be .audacity in your home directory. If you have a scratch partition, use the preferences menu to point to it.
NOTE: Version 1.1.1 of audacity has a lot of problems. You may not encounter any of them. If so, you're lucky. Check for version 1.3.0 on www.mandrakeclub.com soon.
Other Multitrackers
You may want to try sweep, gnusound, tkeca, and others. If you're actually recording multitrack audio though, you're likely to stick with ardour or audacity.
Other Audio Software
Mandrake Linux supplies you with easy to install versions of almost any tool you will need.Score Editing: try noteedit, brahms, or lilypond
Audio Players: try xmms, alsaplayer, or snackAmp
Software Synthesizers: try amSynth, iiwusynth, or spiralsynthmodular
Sequences: try muse, or rosegarden
Sample Editors: try dap, ecawave, mhwaveedit, rezound, snd, or wavesurfer
Drum Machines/Sample Loops: try hydrogen, freebirth, or SpiralLoops
Mandrake's urpmi takes care of dependency checking, as long as you have your contributions repository properly setup. Use urpmi.setup to do so. To install new software, use Configuration>Packaging>Install Software from the menu, or just execute:
# urpmi
and urpmi will take care of the rest.
Remember, for any applications that deal with live audio, choose the jack output option if it's available. If not, choose alsa if it's available. If not, go the the default (OSS). All will work, but of course you want artifact-free audio on the first take if possible.
Updates and New Packages
Between Mandrake releases, packages are not officially updated. The MandrakeClub has a team of volunteers to provide this service for you. If you need an application updated (audacity and ardour have frequent bug-fixing releases), you can request an update on MandrakeClub, and usually a volunteer will provide it for you, and everyone else. Also, if you hear of a new, free application you're interested in, a MandrakeClub volunteer is likely to package it for you if you ask.
The MandrakeClub rpms are available to everyone on any Mandrake-devel server, under the /unsupported/MandrakeClub directory. This directory is also urpmi-capable. Use Configuration>Packaging>Software Sources Manager in the menu to set it up.
Being a member for MandrakeClub is a great way to help the distribution continue, because virtually 100% of your membership price goes to the company. If you can't afford the small fee, you can easily become a volunteer and get a free VIP membership that way.
Non-free software
Mandrake only distributes software which is free and freely distributable, so you'll have to find some software yourself due to licensing, distribution, or patent regulations.
The best source for free but not open-source software (RealPlayer, Flash), is MandrakeClub. They provide RPMs specifically for your distro, and you can configure urpmi to get them automatically. Otherwise you can look elsewhere on the web. Try to use RPMs built for Mandrake 9.1. If there are none, try RPMs built for the latest RedHat version. These may not work as expected though.
The one thing you may want is an mp3 encoder. This is not free. MP3 is a patented technology, and anyone distributing an mp3 encoder is supposed to pay royalties. Luckily, there is a free alternative called ogg. Ogg/vorbis files are supported by most linux applications, and many Windows applications including WinAmp. LAME is the most common mp3 encoder and is supported by audacity.
You can find packages of LAME and other restricted software, built for Mandrake 9.1, on the web. Be sure to read the license before you use them! PCLinuxOnline.com has a great list of places to find unofficial Mandrake packages.
Monitoring Your Performance and Improving It
As mentioned earlier, the enemy is latency. This is a combination of hardware and software being too busy, or the kernel scheduler not realizing that you're doing realtime work. Install the multimedia kernel. This will solve most problems by itself.
To test your latency, use ardour and jack, as this is the most efficient combination you can have. Run jackstart in a terminal. Open ardour and do some recording. You can even try doing other things at the same time (open other applications, browse the web, etc). Every time your system experiences latency, jack will show an xrun in its terminal window. You can ignore xruns under a millisecond or two, as these won't be audible. If you are experiencing xruns over 10 ms, you will notice them in your recordings. Consider a faster processor, more ram, a faster RPM hard drive, or an XFS partition.
Getting help and helping others
There are many many sources of help on the net, but if you want a good response, be sure you:- try to solve the problem yourself first... read the documentation (usually in /usr/share/doc)
- browse the web for similar problems
- make sure you're asking for help in the right place
- don't expect miracles unless you plan on paying for it
- be kind and respectful; most of this software is written by volunteers
If your problem is with Mandrake itself (installing or configuring the OS), you have two options. If you simply are confused and need assistance, go to www.mandrakeexpert.com. If you know what you're doing and you think you've found a bug in the software, go to qa.mandrakesoft.com.
The same is true for the audio applications. Most have a users' mailing list where you can chat and get advice. See:- http://lists.sourceforge.net/lists/listinfo/audacity-users for audacity
- http://sourceforge.net/mailarchive/forum.php?forum_id=11845 for ardour
Both of these projects also have developers' mailing lists too, but don't bother them unless you really know what you're talking about. As I mentioned earlier, they know about most bugs already. If you really want to see a new feature, it's best to subtly suggest it on the users' list. I'm sure it will get back to the authors soon enough.
An amazing resource is the Linux Audio Users' Group. They also have a mailing list with very knowledgeable people. This is a great place to ask about hardware, software, and production. It's not recommended to get new software here though, as it won't be as easy to use as the rpms designed specifically for your Mandrake version.
http://www.linuxdj.com/audio/lad/subscribelau.php3 for the mailing list
http://www.djcj.org/LAU/guide/index.php for the homepage and FAQ's
If you're a developer and want to help out, start by joining the Mandrake Cooker List. Cooker is the developers' work in progress to improve Mandrake. See http://www.mandrakelinux.com/en/cookerdevel.php3 for more information. You may also consider becoming a volunteer on www.mandrakeclub.com.
If you can't contribute in the form of bug-testing, coding, or writing documentation, you can help any of these projects out with cash donations. Joining MandrakeClub is the first step, but Ardour, Audacity, ALSA, and jack are all willing to accept donations of any size as well.
Disclaimer
Here come the ubiquitous disclaimers.
Firstly, this document and related software are licensed under the GNU Public License, and thus carry no warranty, and the authors carry no liability regarding your actions as a result of this document, or the actions of any software employed as a result of this document.
Secondly, while much of this information crosses distributions, and even platforms, much of it applies only to Mandrake Linux. Do not expect to follow these instructions on other distros with any sort of success.
Thirdly, multimedia applications are probably the most rapidly changing segment of linux software. This document applies only to the time period when it was written or updated. This is an unfortunate situation for both users and developers, but it is necessary. Rest assured that standards will emerge, software will mature, and people will agree – not totally, but more than they do now.
Conventions
Please realize that any words encased in angle brackets are not to be taken literally. They are to be replaced with another appropriate word.
Also, commands which are to be executed in a terminal are written after a ($) dollar sign. Simply open a terminal (like gnome-terminal, kconsole, or RXVT), and type the command, without the dollar sign. Commands preceded by a (#) pound sign have to be executed by the super-user because they are slightly dangerous. To become the super-user, open a terminal, type 'su', and you will be asked for the root password. Then execute your super-user commands. Then type 'exit' to become a regular user again.
Acknowledgments
This project would not have been possible without:
Olivier Tauvin, author of urpmi.setup, Götz Waschk, packager of many of the multimedia applications, Fernando Lopez-Lezcano of Planet CCRMS and the members of the Linux Audio Users Group for ideas and inspiration and most of all, Danny Tholen, packager of kernel-multimedia, Buchan Milne has been a contributor to this document.
About the principal author: Austin Acton is a synthetic organic chemistry student and teaching assistant at York University, Toronto, Ontario, Canada. He is also a member of the Mandrake Club Volunteers and the Mandrake Contributors. He is also a contributor and maintainer of many scientific and multimedia applications for Linux on Mandrake.
Talk back! Do you have comments or questions about this story? talkback here
(Click here for further information)
|
news feed