Linux in the Studio

Linux-Penguinby Gord Williams

For the most part when you mention an operating system in relationship to recording, you probably think Mac OS X or perhaps Windows 7. Windows 8 probably is a bit too new for many, though it does seem to be able to handle the chore of driving the processor very well, which is encouraging. For some it’s just not proven yet. Ideally, proven should actually be read like hardened or shone to perfection.

No one wants after the 32nd take from the bass player or the 30th vocal take to find out the that computer your recording on just doesn’t get it. Does Linux operate in the studio to the level that’s required for most professionals? The answer is yes. There are many professionals using Linux around the world in their studio.

Some of the reasons would include:

1. It can be very security conscious, and depending on the project can be more secure than OS X or Windows. Basically because the average Windows or Mac user isn’t allowed into the substructure of the operating system as readily. Also there are security conscious flavours or distributions to build from if desired.

2. Many performance based artists such as club DJs or similar performers are building their performance boxes with Linux, because smaller versions of Linux are powering many small devices. Taking the time to create a ‘performance box’ with Linux can pay off by creating a performance environment that is unique to the individual’s requirements. Rather than the performer remembering commands, the commands that come naturally to the performer can be programmed or stored.

3. Linux now requires less help in handling most audio hardware, particularly Firewire and USB devices. Linux can be deployed in the studio with less research into what devices can be used without restriction. If you have tried Linux before, more devices are supported and most major brands are represented. It would be best as with any other operating system to be thorough in researching, because sometimes the drivers aren’t available, especially for devices that are very new.

4. Linux does have a tendency to be a little behind the curve in terms of device drivers. This can be an opportunity to deploy older devices obtained through barter or at a reduced cost which were state of the art or bleeding edge as little as five years ago. This depends on the device and how proprietary they were to begin with, but now most proprietary device drivers are dealt with in what for Linux is a relatively short period of time.

5. Testing and development happens at an admittedly sporadic pace, but most major distributions have an audio flavour and a large enough community working on it for support. Some are combined with graphics, or photography into a more multimedia field, to help process everything from .raw still camera images to creating 3D animation. Some of the major studios have been using specific versions of Linux on major movies to create CGI.

6. Linux begets Linux, several machines can be joined together in a cooperative venture to create more power for a single process, share processing load, or distribute processing load.

a. CGI work for major studios was done on a Linux server farm including Avatar, to help speed up the rendering without losing detail. Several machines working on bits of a bit of a screen, can work faster than one, and with a server farm failover is possible. Rendering can be problematic. There can be similar rendering at mixdown in most Daws.

b. A recording studio can take to virtual space and work in time with other recording studios via streaming in sync, or via the cloud not in sync. Advantages of this can be that a studio does not need to have huge bandwidth to create geographically diverse projects. A bass player in Switzerland could be working with a drummer in LA and the back and forth can be very rapid. Assuming that both are technically competent or have the assistance to record to specification, the studio provides the interface for standards and sub mixing. The overall mix can be distributed a variety of ways. It can be streamed out so all project members can listen to mixes live at geographically diverse locations or via the cloud at their first opportunity.

c. Radio Stations can work in Pod type environments where producers work in groups to create a content feed, and as with the recording studio, it can be live or posted to the cloud for later in the broadcast schedule. The option you may find interesting is that field reporters can be set up to rely less on phone patches and record closer to studio quality audio in the field, and VoIP could be used in place of phones from computer to computer. Depending on bandwidth available this can actually be an improvement on the tinny phone patch.

While many of these items above can be deployed in OS X or Windows, most Windows or Mac users will not be able to easily adjust VoIP or streaming settings so that a last minute recording session can be facilitated. With Linux the answer often is to adjust a text file, as most settings are controlled through it.

In some cases a modification of the operating kernel is required and a reasonably technologically proficient person can handle most of it. If some help is required, much of it can be obtained without waiting in queue for customer service. Live chat on IRC is still very popular with Linux were you can find someone who has done something very similar.

In the studio the biggest advantage would be failover. In a session where you have limited time and the artist gets the right take you can more easily have two or three machines rolling rather than just one. The economics are encouraging to do this.

You can find a USB or Firewire interface for less than $1000 and for many studios you might peg that number at less than $500. Computers do not cost what they used to, and in fact everything doesn’t compare with even five years ago.

It is reasonable then based on this model, to suggest that in terms of hardware you could recycle an older computer as a secondary unit, or even simply purchase a mainboard and find a way to slot mount it. Several of these can serve as a farm or work as blade servers.

But let’s not get too far into that. Let’s assume you have a box or you want to buy a box and not get too far along with a blade server or self-made device.

All you need to buy is the best you can afford and no fluff. On these units is nothing more than a hard drive, and inputs to get the sound in and out, and a video card. The studio will have a monitor for each of its units. So far you could be looking at around $500 or less. We are not driving graphics work in the studio; the assumption is the mainboard video will do just fine.

Add another $400 on to the cost for two reasonable sized monitors. If you’re using the units near an open mic, like in a radio production studio/live studio, then silencing the units might cost $400 more. So far the tab is at $1800 (or even less if your frugal) because there isn’t the cost of software.

With most digital recording studios much of the overall cost can be software. Five hundred per seems about right for editors like Adobe Audition or some versions of Pro Tools. Then there can be specialty versions of the OS like Windows professional or Ultimate around the same price.

The hard part of a decision to go with Linux all the way is loss of user software programs already invested in. This could represent loss of a large investment.

The precise software programs you may do your work in are probably is not available for Linux -- not in the strictest sense of that statement. So there will be a learning curve for similar software. Ardour has been compared to Pro Tools for instance, but it just isn’t Pro Tools.

If you use a DAW like Pro Tools or Adobe Audition, the replacement Linux program would likely be something like Ardour/Mixbus, EnergyXT, or the newer Tacktion, which is being developed for all platforms, by some folks that used to be at Apple.

The tools are available. Deciding to move to Linux is a matter of deciding to learn the tools and keeping an eye on development of newer software. Tracktion for instance just recently picked up its pace of development after being around for a number of years.

Softening the transitioning blow (due to lack of native versions of the most popular DAWs) is the fact that you can run many of your beloved VST plugins and VSTi instruments. There are interfaces that will allow you to run them in software like Ardour and Mixbus, even wave editors like Audacity can benefit from this. Considering that thousands of dollars could be represented in plugins alone, this could be enough -- particularly if you require MIDI clock driven recording, Linux could achieve it.

If you had to start over with the insurance check or the needs of the many just are driving you forward, Linux could very well help you invest in areas you have been thinking about for a while like bandwidth or other aspects of the recording business. Encouraging you, perhaps enticing you are the opportunities for greater failover, streaming sessions within grasp, and software revisions without economic penalty.

To be realistic one would have to expect that at some point there may be some development required as there are more rough and ready edges on Linux than Windows or OS X. Windows in some versions has been reputed to have millions of lines of code strewn about waiting for hackers to exploit. Mac OS X has the same parent as Linux, and could be considered the closest comparative operating system. While that is true, the comparison does not fly from the economic perspective. Mac has always been very proprietary, and economically you are held to one domain, Mac.

Linux through emulators does run software from the other two natively (without Windows or OS X) or you can, like with Mac and Windows, run the entire OS via an emulator such as VirtualBox or VMWare. This can be a complex world but it can be done. Emulators have a life of their own at times, but they work and often they work well. Through virtualizing a studio, you can keep a foot in the past and work with familiar programs.

Virtualization could be a way to go for some studios depending on what their business model is. The model discussed throughout has been with the precept that most studios would want to access hardware via the OS without waiting for interpretation of an emulator or virtualization.

Conclusion

Linux has been ready for several years with a multitasking kernel that can be modified to purpose, or more recently be left to its own to administer the system in a way that is every bit as effective as its counterparts.

Linux excels at the networking, and can handle more than one machine at a time. A reasonably adept person can fairly easily make Linux into a studio recording device handler that offers failover on more than one machine operating simultaneously for the purpose of recording, much the way the good engineers would roll more than one device to capture a recording.

Linux is very compatible with other operating systems. The entire operating system can be ported over to one or more computers in the studio for many purposes, among them high file transfer reliability. Other software such as Office or specific internet software and similar are not lost in the process.

Linux can free a studio operator from some economic constraints of proprietary software facilitating opportunity to invest in a more diverse fashion, perhaps spurring new business opportunities.

Comments (0)

There are no comments posted here yet

Leave your comments

  1. Posting comment as a guest. Your post will be moderated. Your email address will not be shown or linked. (If you have an account, log in for real time posting and other options.)
0 Characters
Attachments (0 / 3)
Share Your Location