Linux Audio Studio

One of my new favorite Bolger site users is http://homestudiotalk.blogspot.com/ and had I known about his site I may have not ever started this one. Today there is a new post about Ardour, which is a free open source digital music studio software for Linux and the Mac. I have been really into Linux a few years back and still see the potential for Linux. Several years ago I had setup a Linux box and attempted to turn it into a digital music studio. This was when I was in school and studying hardware and software and though it would be a good opportunity to use what I know about computers and music and put them together. I even went so far as turned it into my teach for extra credit, but really I did it just to say I could do it.

Wow how things have changed now there there are a bunch of different sound audio applications and a whole slew of different distributions just for audio production and recording.

Since the post about Ardour on http://homestudiotalk.blogspot.com/ I decided that I would dual boot my computer and really test out some of the software. As I right this I am downloading Ubuntu Linux 7.10 Desktop. I wanted to take this opportunity to talk about what it was like several years ago. I’m not sure what the exact date was but it was when Mandrake 10 was out and I’m guessing about 3 years ago.

At the time, there were some Linux issues; first was being latency and the second was a new sound driver build into the kernel called Alsa.

There were other audio freaks out there and they had created a low latency kernel that you could install. The problem with the latency in the kernel was pretty basic. Linux is based on Unix and is a server based operating system. It wants to server requests, but also it felt that a request for sound was less important that anything else the CPU might be doing and it would hold off on processing audio. This is what caused the latency. I’m under the impression that this has been fixed in more recent version of linux.

The second issue was with the sound driver called Alsa. Before Alsa there was a set of calls in Linux called OSS. The problem with that is it could only process one sound at a time. Fore example if you were listening to a CD and checking your email. The cd would play like normal but when ever your email program would chime that you have an email it would for wait to play the chime until the music was done. But Alsa fixed this but at the time it was a very new technology.

With my experiment I installed Mandrake 10 and then proceeded to re-compile the low latency kernel and alsa modules. Then other programs. In the end it took about 5 hours not including initial installation.

When I was doing this I was also comparing it to windows using Magic Music Studio Generation 6. I hated to admit that Windows performed better, was easer to use and configure.

The bottome line in creating a music studio is really about building a creative enroment. I want to make music, record my ideas, and produce and process my craft. I do not want to fight with Kernels, sound drivers, compiling source code, and shutting off services. Just let me play.

I’m going to give it another shot and will post my feeling and thoughts.