Jump to content

Think Linux Will Ever Be Viable for Music? Should It Be?


Anderton

Recommended Posts

  • Members

In the latest Sound on Sound, Paul White's editorial is about the complications of our computer-based music creation being dependent on two large corporations for whom music is not a big part of their bottom line. He further mentions how much time companies lose trying to keep up with OS changes. And although he didn't mention it, there's also the issue of testing for a variety of operating systems. He thinks it would be good if the DAW companies got together and just created an OS for music.

 

But...why not Linux? I don't know much about it so there's probably a good reason "why not." Maybe because it's open source it changes even more than The Big Two operating systems? Do you have to anything special with the hardware to make it run? Enquiring minds want to know...

Link to comment
Share on other sites

  • Members

I consider it to be a prime solution, with potential to be ideal.

 

The hitches are the same as for supporting any OS: Device drivers and Proprietary software

 

If manufacturers write device drivers properly (using virtualization to make them OS agnostic) it would be pretty simple to port drivers to new OS systems. If they stick with the traditional 'quick and dirty' method of coding (writing a separate work of code for each OS), they have to rewrite the driver from scratch for every different OS, and maybe even rewrite it for every different OS version.

 

To write virtuallized versions of drivers makes the upfront development cost slightly higher, but pays big dividends in the long haul.

 

It allows porting the same device driver code to different OS's, and when a new OS version (I'm talking Apple & Microsoft) comes out, it only requires minor code tweaks to fit the virtual driver to the new API are needed. This means it would be LOTS cheaper to support a product for a longer time, over multiple versions of an OS. (Maybe this isn't attractive to those manufacturers who count on OS version releases to force their products to go obsolete... But, for the 'good guys' in the industry, who continue support for products long after the sale, this gives them quite a leg up on the competition who don't...)

 

Proprietary software (such as an audio interface that, in addition to needing a device driver, also has a mixer app that allows low-latency monitoring and DSP effects in real time) - - this may be a whole different kettle of fish... While there may be opportunity to leverage virtuallization software coding techniques, it might be somewhat harder to make the code indifferent to which operating system it's running on. Probably a lot harder than doing so with a device driver.

 

With all that said, I am eagerly awaiting the day I can dump Microsoft for good, and get a really reliable operating system to use for studio work....

Link to comment
Share on other sites

  • CMS Author

It's one of those circular things. There's little music production software because there's not much support for audio hardware in Linux, and hardware manufacturers aren't writing Linux drivers for their interfaces because there's too little software that would make use of them. Harrison has been using Linux in their big production consoles for years, but they designed the software and hardware together and didn't worry that you couldn't run their software with a $100 TASCAM two-channel interface. They were active in the development and support of Ardour and now they have Mixbus which is based on Ardour and runs on Mac, Windows, and Linux. But while Mixbus is perfectly capable of supporting things like the 64-channel Antelope interfaces, Antelope doesn't talk to Linux.

 

Last time I played around with Ardour, most of the interfaces that were supported were either under a generic driver for simple interfaces or for older hardware like the M-Audio Delta 1010. There are a couple of driver projects ongoing with people reverse engineering from the hardware, but when they come up with something, it's just for audio I/O, so you don't get full mileage out of the interfaces with built-in monitor mixing and routing. The most capable multi-channel setup I was able to come up with using what was laying around the shop was Ardour running with an original Mackie mixer with the Firewire option card for which there was a driver from the FFADO project. The hardware manufacturers aren't inclined to share design documentation with volunteer open source software writers, and there isn't a "class compliant" standard that hardware manufacturers could write to if they wanted to bother.

 

Here's a link to the current list of "full supported" FFADO drivers (for Firewire interfaces). Not much new here. The ALSA project has pretty spotty support for USB audio interfaces.

 

So go tell all of your industry buddies about how they'll sell a pile more interfaces if they only supplied them with Linux drivers.

 

And along the same line, we users of cheap Android mobile devices are pretty much left out of the serious audio business. While Andrioid Version 5 and above will nominally support audio I/O through the USB port, without USB-OTG support, which seems to be left out of phones costing less than about $200, you can't even talk to an interface. And only Samsung seems to have expressed any enthusiasm for pro audoi support, though most of that is probably going into chasing virtual reality applications.

 

 

 

 

Link to comment
Share on other sites

  • Members

As has been said above, Linux is more than capable of doing the job, but it will take some time and money by some dedicated individuals before this will happen. I use Linux for everything except music recording, where I still rely on Windows and Sonar (and keep Windows off line). Like was said above, it will be a joyous occasion to completely dump Windows.

 

One has to wonder if the audio companies want to continue to rely on an OS that can break their software, and often does so on a regular basis. Not too long ago there was a thread in the Cakewalk SONAR forums where some of the users of SONAR were experiencing issues, and there was some nasty words directed to the folks at Cakewalk. User's wanted to know why they should continue to pay for 'buggy' software and when was Cakewalk going to issue a fix. Well, it turns out that the issue wasn't Cakewalk, but rather the latest update to Windows 10, and folks had to wait for Microsoft to make the fix.

 

And it took some digging by some dedicated folks to figure this out. It wasn't like Microsoft came out and said it was their fault. Instead, they let the software developers take the hit. Pretty nice, huh?

 

There is also the issue of Windows now becoming an advertising platform (see https://arstechnica.com/information-technology/2017/03/microsoft-put-gross-ads-in-windows-explorer-and-i-dont-have-the-energy-to-be-angry/). Think about how this might impact your ability to record and mix music, as the OS is more interested in pushing ads then making sure your music studio is running OK. And if this continues and becomes more 'adaptable' to your use of the OS, how happy will you be if it starts showing ads for a rival studio or DAW software?

 

 

Link to comment
Share on other sites

  • CMS Author

If you want something that works to keep working, avoid updates!!!!! This is more important than ever with Linux and audio applications. Updates are for common folks, not audio engineers. Most of my computers, for audio applications and othewise, are running Windows XP. I still can't smoothly get through the protections in Windows 7 and I'm always finding (and not being able to fix) folders that I as the adminsitrator can't access, both locally and on networked computers. Linux has its annoyances, too.

 

 

 

Link to comment
Share on other sites

  • Members

Hi Craig,

 

I don't see it happening. If it would have happened, it would have been 15 years ago when Linux had more momentum. Independent of it's merits, the music software companies don't have the resources to develop or maintain another platform. I think if they are contemplating another platform, it's going to be a tablet-based solution (as impractical as that is for editing, given the limited screen real estate).

 

My system is really stable with Windows 7, but I'm running software that is a bit older (Cubase 6.5 and Wavelab 7), and I keep it off the internet completely. I'd love to see Microsoft to an ultra-stripped down version of Windows for folks like us, but I don't see that happening either.

 

 

 

 

Link to comment
Share on other sites

  • Members

What might be useful at some point in the future is the ReactOS (https://www.reactos.org/). It is an operating system that is designed to be Windows compatible. They are working with the folks at WINE and have made some good progress,although it has been a while since I have played with it. One day this might be the best alternative to Windows instead of a Linux system.

Link to comment
Share on other sites

  • Members

I imagine that changing policy to support Linux users who not only have some more in-depth OS knowledge (or at least they should have) but also run applications which are free of charge (what an awful idea!) will be difficult. So unless hardware developers will see an opportunity to expand their markets within the wide range of Linux (or other OS) users nothing will change and we will have to deal with post-update problems.

 

However the advantages of building a music-dedicated OS are obvious:

- Its development could be sponsored by DAW developers. The costs could be shared, therefore acceptable.

- The users feedback would be just awesome. See some of Linux communities trying their best to deal with what they have to at least have an impression of good workflow in music production.

- The price of such OS could be low, or it could be even free as quite a big work would be done by volunteers just for the sake of their comfort.

 

Assuming above, expected popularity of such solution could be great. BUT it's still utopia as long as companies prefer to compete instead of cooperate :(

Link to comment
Share on other sites

  • CMS Author

However the advantages of building a music-dedicated OS are obvious:

- Its development could be sponsored by DAW developers. The costs could be shared, therefore acceptable.

- The users feedback would be just awesome. See some of Linux communities trying their best to deal with what they have to at least have an impression of good workflow in music production.

- The price of such OS could be low, or it could be even free as quite a big work would be done by volunteers just for the sake of their comfort.

 

Well, it didn't work for BeOS. The short-lived TASCAM SX-1 used it as its base, some Edirol (Roland) video editors used it, and RADAR recorders up through RADAR 5 used it. But those were all closed systems used their own hardware.

 

There are actually quite a few small (2 channel I/O) USB audio interraces that are supported by the ALSA project under Linux, so it's perfectly reasonable to set up a Linux-based singer-songwriter or a narration studio, but it's hard to get past that with the current level of audio hardware support. It's also hard to see that this will improve since audio hardware manufacturers are already doing just fine supporting Windows and Macintosh systems. And I don't see any Linux software developer thinking that he'll make money selling software support (for example, the Linux equivalent of an ASIO low latency driver) for a popular piece of audio hardware. One reason is that in the time it would take to develop and test the software, that popular piece of hardware would become last year's model. This is why there's partial Linux support for 10+ year old pieces like the M-Audio Delta 1010, but not this year's Focusrite Scarletts beyond two channels.

 

 

 

Link to comment
Share on other sites

  • Members

Sad but true.

 

But still I think that with appropriate amount of resources and programmers there is plenty of space in Linux market for such OS to be established and developed. The only condition is the first major "player" to appear and show others "the way". It will take long time and it will not pay out soon, but in long terms...

 

Linux becomes more and more popular and hardware manufacturers/software developers like to choose it as a base for their front-ends. Let's take Android for an example.

Link to comment
Share on other sites

  • CMS Author
Sad but true.

 

But still I think that with appropriate amount of resources and programmers there is plenty of space in Linux market for such OS to be established and developed. The only condition is the first major "player" to appear and show others "the way". It will take long time and it will not pay out soon, but in long terms...

 

There's a problem with making it pay, though. The big Linux users like it because it's robust and can be customized to specific applications. Those users don't care that the core operating system is free, they pay plenty for their support and feel that they're getting a good deal and a good working system. The problem is with the people who want to use Linux for other reasons, primarily:

 

(1) They won't be bound by Microsoft

(2) They won't be mount by Apple

(3) It's free! Free! Free!

 

That won't support development of musicial applications because, as you certainly know, many people expect that music should be free, too.

 

Where I think Linux can make some headway is for a manufacturer to come up with an integrated hardware/software system that is attractive enough to musicians, producers, and studios to be worth not doing it yourself. But we've become so accustomed systems that grow with add-ons of our choice. It's not likely that Waves or Sound Toys will be able to make a single version of a plug-in that will work with Roland's and Yamaha's and Avid's custom Linux based workstations.

 

 

Link to comment
Share on other sites

  • Members

A quick search showed a number of interfaces with support:

http://wiki.linuxaudio.org/wiki/hardware_support

 

Overall, Mike is right - - most of the interfaces there are 4 in/ 4 out or smaller, and more than a few are quite old.

(Side note: This might offer a way around the problem of having your hardware become useless because Win or Apple updates their OS. Run it under Linux instead)

 

Still, it looks like it has improved a quite a bit since I first tried Linux for audio some years back.

 

It's not something I'm ready to put time into again just yet. I killed several weeks last time I tried it. I don't have that kind of time these days.

 

This is an example of what I'm not willing to do to get something working:

https://ubuntuforums.org/showthread.php?t=2213941

Life is too short...

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...