linux as a musician's os?

[prophetik music]

from slashdot:

"Keyboard Magazine has an in-depth article about the state of music production on Linux. While it does introduce Linux to the average musician, the article does get into some of the available music applications and music-oriented Linux distributions out there. From the opening paragraph 'You might think there's no way a free operating system written by volunteers could compete when it comes to music production. But in the past couple of years, all the tools you need to make music have arrived on Linux.'"

link: http://www.keyboardmag.com/story.asp?sectioncode=29&storycode=17973

discuss. i still think it's too limited to use as a serious platform for music creation...but...

[djpretzel]

If it were somehow possible to create a wrapper for VSTs... that would seriously change the ballgame. I'm not sure what exactly would be entailed to enable that - whether bits of WINE would be helpful, if it's even possible to achieve with any acceptable degree of latency, etc., but that'd be a major catalyst.

Without VST, DXI, DirectX, AU, RTAS, or TDM support of any kind, the OS can still go far for multitrack audio recording and editing, which is a totally viable use, but is severely stunted for anything else.

A wrapper like that would be paradigm-shifting... wonder if anyone's working on it. There's a Java wrapper that lets you write VSTs/VSTis in Java, so clearly some degree of interception is possible, even though it's not the same thing by a longshot...

[zircon]

The other big issue is hardware compatability. People aren't going to switch to Linux if it means throwing out their MBoxes, their EMU 0404s, their Firepods, their Audiophiles, their Powercores, etc... Linux really needs to have full support of a lot of these major brands.

[prophetik music]

i agree with you about the wrapper, dave. i know that a lot of musicians write using java-based music tools (neil voss, who wrote tetrisphere, comes to mind) but there simply wouldn't be any ability to sample really anything using previous equipment. i assume that linux compatibility for those are pretty far behind other stuff, like compatibility with a lot of the major games out there. it's a pity - i'd love to have all the functionality of linux and the cpu management and all that with something like fl or reason.

[Rellik]

Actually, it's even now possible to use Windows VST's in Linux - there are several wrappers out there that, as far as I know, allow a reasonable majority of synths to be used.

Plus there are plenty of synths written specifically for Linux - I think it's as viable a platform as any other for writing music. I mean, even a Commodore64 or Amiga will do the job - there's lots to like in Linux for any reasonably code-sensible musician. You just won't be able to use the same tools, necessarily.

Of course, I can't see any good reason for most people who are used to their Windows environments to switch - I know I'd never give up my comfortable workflow and hardware support if I didn't have to.

[prophetik music]

not only the hardware support, but i just like working off of a familiar host, you know? until they port something that i can figure out easily, i'm not going to switch. i thought that at one point fl was looking into putting out a linux version, but i don't know if that was for certain or just a rumor.

[OverCoat]

High quality digital music for $0.

Can't get much better than that!

[analoq]

this topic comes up in the Remixing forum here and there. what i usually say is:

if you want to make music in a nice Unix environment, get a Mac!

with that out of the way, the music tools on Linux would probably be enough for me.

but i am pretty minimalist when it comes to my artistic palette.

[djpretzel]

Actually, it's even now possible to use Windows VST's in Linux - there are several wrappers out there that, as far as I know, allow a reasonable majority of synths to be used.

Plus there are plenty of synths written specifically for Linux - I think it's as viable a platform as any other for writing music. I mean, even a Commodore64 or Amiga will do the job - there's lots to like in Linux for any reasonably code-sensible musician. You just won't be able to use the same tools, necessarily.

Whether you CAN use a linux box to make music isn't debatable; as you mention, numerous operating systems that are far more limiting can be used to compose. What we're really talking about here is anything approaching feature parity with the dominant commercial OS's, Windows and OSX. It's a matter of usability, convenience, and quality, and it's partisan *nix bullshit to claim otherwise: quite clearly, there are benefits to using professional audio software, and quite clearly, very few of the commercial plugin developers are targeting linux. It's not a question of whether it's possible, but whether you'd want to. This factors directly into viability, which isn't a boolean, but a spectrum.

I was unaware that anyone had successfully wrapped Windows VSTs for Linux; seeing as plenty of VSTi plugins are unstable enough on Windows as is, I think the "as far as I know" part of your comment is suspect. I didn't see compatibility lists of any kind, screenshots, or even what I'd call straightforward directions and instructions, so while it appears someone's accomplished this task to some extent, the extent itself remains shadowy at best. This is a common problem among linux advocates: someone asks whether something is possible, or whether X feature has been implemented, and the answer is always "yeah, sure, that's been done... it's still got some issues that are being worked out." Which is a fine answer for a labor of love that someone's using as a hobbyist in their spare time, but for anything involving professional work, where deadlines matter and "when it's ready" isn't good enough, that dog won't hunt. Furthermore, many promising linux projects tend to reach 60-70% completion levels and then stagnate indefinitely. For understandable reasons, of course - this is pro bono software development we're talking about, after all - but still problematically. You can't have your cake and eat it too - all too often, people herald the benefit of open, standards-based software, but turn a blind-eye to the support issues, lack of feature parity, and high abandonment and stagnation figures associated with it. Keep in mind, I'm not a Microsoft bigot by a longshot, and love my Firefox. I just calls 'em like I sees 'em.

I think this is a hugely important point because, unless a earth-shattering, groundbreaking, world-bending event occurs, linux is at best gaining desktop acceptance at a glacial pace, even slower in professional arenas where the use of tools like Photoshop, Logic, Pro Tools, Cubase, etc. is absolutely required. I think the best chance for adoption is a "can't beat 'em, join 'em" attitude, and wrapping VSTs effectively would be a huge step - all of the myriad freeware VST/VSTis could be part of readymade distributions, putting very formidable power in one's hands for free. And, if the wrappers were effective enough, allowing upgrade paths to full versions of commercial windows plugs that, like it or not, pretty much rock the socks off anything I've seen anyone put together for free.

Don't get me wrong, the people writing native linux audio tools are saints, geniuses, and deserve the highest praise. I just personally think it's still nowhere near as mature a platform, and that the best way of attaining viability is through compatibility, not creating new standards and implementations.

[prophetik music]

probably half the reason that those programs aren't getting wrapped into linux is just that it'd probably violate copyrights in some way. a wrapper for a free vst would still be violating the free distribution method of the vsts, wouldn't it? since it, to an extent, modifies the vst's functionality without the consent of the creator?