Home |
Search |
Today's Posts |
#81
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
On Dec 23, 12:13 pm, Moshe Goldfarb wrote: Maybe this sounds like a cop out but all I want to do is make music. Then why are you using computers? What is it that leads you to expect that a general purpose operating system (windows/OSX/linux) on a general purpose CPU on a general purpose motherboard is a sensible way to build a tool that will let you "just make music" ? What leads ME to expect it is that I've been able to do just that for a couple of decades using DOS then Windows systems, all on the professional level. So, if one CAN'T do that with Linux, what does it tell you? -- Best, Neil --- news://freenews.netfront.net/ - complaints: --- |
#82
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
On Dec 23, 12:13 pm, Moshe Goldfarb wrote: However, you raise the point which is my personal reason why audio under Linux is a mess. Too many sound systems, and too much needed interaction by the user. one person's "too much interaction" is another person's "not enough control". maybe we should pay more attention to the fact that there are probably more people in the first situation than the second, or maybe we're interested in a niche market of audio professionals who actually want control. There really needs to be somebody doing value added, to cover up the "control" from those who don't need it. Where Windows was a win was the various phases of the multimedia API, which gave people pretty finely-grained control. It's a *sad* API, but by the time N-Track 2.0 was available, you could get real work done. With Windows this stuff is almost totally transparent. Choose ASIO or WDM and that's it. Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? Does that mean it no longer works, or that it's not actively supported any more? Got link? That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That before WDM, there was no reasonable "out of the box" low latency solution on that platform? WDM isn't that low latency anyway... latency is a nonstarter... er, it's nice if you can get it, but it's pretty easy to design workflows that go around it. That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? The situation is certainly cleaner than Linux, but it hardly gets close to the cleanliness (for the user) of CoreAudio. With Linux it's a mess. Also, OSS, Arts, eSound, and now Pulse Audio the current prom queen of the Linux world, comes along to confuse things even more. What's a mess is that so many people, including you, have read so much stuff that has led them to completely misunderstand the role of these things in the Linux "audio stack". For a report on my perspective, this is useful: http://lwn.net/Articles/355542/ Thanks for that. Wow, 1998, huh? Yes Jack, which is kind of like Rewire coupled with Steinberg's Control Room, allows you to do a lot of things but it assumes you know what you are doing, or are willing to scour the net to figure out how. If I could control the entire Linux "audio experience" of every JACK user, you would never have this perspective. But Linux is not a company. If you want a smooth experience with Linux audio, do you randomly pick some distro, some machine, some audio interface and put them together and expect that it will all just work? It appears that many existing or potential Linux users do indeed expect this to be possible. Sorry, its not. But that's a defining difference between Dozers and Penguins. Its unfortunate that so many people believe that it is, or even more irritatingly, believe that it should be. Its not possible on Windows (you just have a higher success rate with random selections of (windows-version,hardware,audio-interface) - many audio forums for Windows DAWs are full of testimonies to problems that people have with particularly bad combinations of choices. True! Various Firewire chips have been discussed here ad nauseum. If you want that kind of experience, you need to get your system from a company that controls everything end-to-end, which means either Apple or a company that specializes in building machines for media work that run Linux. Unfortunately, I can't recommend any of them at this particular point in time. Or you can selectively buy known-good stuff for Doze. Maybe this sounds like a cop out but all I want to do is make music. Then why are you using computers? What is it that leads you to expect that a general purpose operating system (windows/OSX/linux) on a general purpose CPU on a general purpose motherboard is a sensible way to build a tool that will let you "just make music" ? FWIW, Mike frequently touts using units like the Alesis HD2424 plus a good, analog console for getting real work done. But DAW stuff has been more-or-less possible for ten years now, and we all pretty much know those will go by the wayside. http://www.alesis.com/hd24 And what I am seeing is that virtualization seems to be able to make the point moot - one may be able to run a Doze DAW on an otherwise Linux box. I'm still researching this myself. -- Les Cargill |
#83
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 14:20:18 -0800 (PST), dawhead wrote:
one person's "too much interaction" is another person's "not enough control". maybe we should pay more attention to the fact that there are probably more people in the first situation than the second, or maybe we're interested in a niche market of audio professionals who actually want control. They want control of their music and the applications that record their music. They most certainly do not want to be playing with operating systems like Linux forces the user to do. With Windows this stuff is almost totally transparent. Choose ASIO or WDM and that's it. Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That before WDM, there was no reasonable "out of the box" low latency solution on that platform? That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? The situation is certainly cleaner than Linux, but it hardly gets close to the cleanliness (for the user) of CoreAudio. You're kidding right? What I *do know* is that I plug in my audio interface, under Windows 7 x64 BTW, my control surface, select ASIO (my choice) and it all works. I never have to touch a single thing other than latency. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. All my plugins, well most of my plugins work. Some, like Toontrack are having some growing pains with Windows 7 x64, but that's coming shortly. Linux doesn't even support this stuff. And you are deluding yourself if you think for one instant that a professional, like myself or Mike or Scott or Hank or Fletcher will for even an instant claim that the Linux solution is "cleaner". Maybe it's cleaner from a programmer perspective, a developer perspective and maybe in theory ie: crunchy code and all that gibberish. However I can tell you that if Linux were the only software choice for a DAW we would all still be using tape and a razor blade. Try that under Linux. Oh yea, if you have SoundBlaster, it will probably work fine. With Linux it's a mess. Also, OSS, Arts, eSound, and now Pulse Audio the current prom queen of the Linux world, comes along to confuse things even more. What's a mess is that so many people, including you, have read so much stuff that has led them to completely misunderstand the role of these things in the Linux "audio stack". Why should I have to even read? The reason is none of this crap works correctly. Mike's venture into this is a prime example. You Linux developers are all doing your own thing and apparently nobody is talking to the other people. Example: Pulseaudio. Yea, I know Ardour etc doesn't deal with it because it's ALSA based. So why is that? If ALSA works fine, and it does IMHO, why all the others? Oh yea, "choice"... No, it's not choice. It's a cluster****. Take a look a the Ubuntu forums and see how many Pulseaudio problems you will find. For a report on my perspective, this is useful: http://lwn.net/Articles/355542/ Fair enough... My personal feeling is ALSA should be tightly integrated with the kernel (I'm not a programmer so please excuse the simple language) and the others should disappear. Yes Jack, which is kind of like Rewire coupled with Steinberg's Control Room, allows you to do a lot of things but it assumes you know what you are doing, or are willing to scour the net to figure out how. If I could control the entire Linux "audio experience" of every JACK user, you would never have this perspective. But Linux is not a company. If you want a smooth experience with Linux audio, do you randomly pick some distro, some machine, some audio interface and put them together and expect that it will all just work? It appears that many existing or potential Linux users do indeed expect this to be possible. Sorry, its not. Its unfortunate that so many people believe that it is, or even more irritatingly, believe that it should be. Its not possible on Windows (you just have a higher success rate with random selections of (windows-version,hardware,audio-interface) - many audio forums for Windows DAWs are full of testimonies to problems that people have with particularly bad combinations of choices. If you want that kind of experience, you need to get your system from a company that controls everything end-to-end, which means either Apple or a company that specializes in building machines for media work that run Linux. Unfortunately, I can't recommend any of them at this particular point in time. Bit that's exactly the problem! Too many hands in the pot and too many "solutions". Put it this way, if I had you in my studio for a couple of hours, I am certain you could show me things that Jack/ALSA etc can do, that I could not even in my wildest imagination conceive. But I don't have you and documentation is poor so I am left to mostly fend on my own, like most people. The closest example I can give is Linux Konqueror. How many people really realize how powerful that application is? It goes far, far beyond a browser. Maybe this sounds like a cop out but all I want to do is make music. Then why are you using computers? What is it that leads you to expect that a general purpose operating system (windows/OSX/linux) on a general purpose CPU on a general purpose motherboard is a sensible way to build a tool that will let you "just make music" ? Ivory. Garritan Addictive Drums UAD Sonnex Nuendo. ProTools. Do they run under Linux? That's where Linux loses, at least for me. You see, you look at the computer from a programming POV. I look at it from a musician's POV. Big difference. For me, the computer is a gigantic tape drive and the editing is the mother of all splicing blocks. The effects are the racks of equipment I used to have (I still have some BTW) or the units I could not afford back in the 70's. Ivory is the Steinway, Bosendorfer, Yammy etc I cannot fit in my living room as well as the superb mics needed to record such fine instruments I could not afford. Actually I have a Steinway B now and some very nice mics, but the acoustics in my living room where the piano is are marginal. Ivory or Garritan is a much better choice in MOST situations. It goes on and on. I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? |
#84
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 18:18:02 -0500, Little****t wrote:
On Wed, 23 Dec 2009 14:20:18 -0800 (PST), dawhead wrote: I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? You need to try an audio tageted Linux distribution instead of using a generic one. Try UbuntuStudio which will eliminate many of the problems you are talking about. Just my 2 cents. |
#85
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 18:11:19 -0500, Neil Gould wrote:
What leads ME to expect it is that I've been able to do just that for a couple of decades using DOS then Windows systems, all on the professional level. So, if one CAN'T do that with Linux, what does it tell you? DAWs are just one example of speciality applications (CAD is another) where there'll only ever be a small population of users but the software itself is very complex. Consequently anybody investing development time on such software commercially is only going to go for the most widely used platform, or their already tiny market will become even smaller. Actually if Pro Tools, Nuendo or whatever were ported to Linux, I'm sure it would run like a champ, but there just isn't the commercial incentive to do it. -- Anahata -+- http://www.treewind.co.uk Home: 01638 720444 Mob: 07976 263827 |
#86
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Neil Gould wrote:
Meanwhile, "professional" means that one can make a living in their area of expertise, which rules out most applications under Linux. From what I've seen, Open Office is about as good as it gets, and it isn't very competitive with MS Office 2000 w/r/t user features. I've found some glaring errors in its documentation that could not have persisted in a commercial app. I can only conclude that the users don't care that it doesn't work as the documentation says it does. Open Office is interesting, because the technical documentation is better than the user documentation. Contrast that to MS Office, where the user documentation is excellent but the technical documentation is nonexistent. This is probably the best example of what I mean when I say that there is a lot of excellent open source stuff out there which is poorly documented. But then, I'm a guy who would say that what is -wrong- with MS Office is that Microsoft keeps burdening it down with more and more useless "user features." --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#87
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 6:18*pm, Little****t wrote:
They most certainly do not want to be playing with operating systems like Linux forces the user to do. Then for the time being, at least, they probably shouldn't bother. What I *do know* is that I plug in my audio interface, under Windows 7 x64 BTW, my control surface, select ASIO (my choice) and it all works. I never have to touch a single thing other than latency. What I do know is that I plug in my audio interface (an RME hdsp9652) under Linux (Fedora 11 x86_64), plugin in my control surfaces, start JACK and it all works. I never have to touch anything except hdspmixer to reset the initial matrix mixer settings for the hdsp, and JACK to adjust latency. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. I fire up Ardour and it all works. All my plugins, well most of my plugins work. All of my plugins, well most of my plugins work. Linux doesn't even support this stuff. Wrong way around. The developers of those plugins don't support Linux. Can you run Ardour on Windows? Same deal. Linux' job isn't to support pre-existing software, just like OS X's job isn't to support Windows software. Clear thinking about where responsibilities lie here would help. And you are deluding yourself if you think for one instant that a professional, like myself or Mike or Scott or Hank or Fletcher will for even an instant claim that the Linux solution is "cleaner". Fortunately, I am not deluding myself for one moment into thinking that any professionals like yourself, Mike or the other people you mention (the last 3 I do not know) will claim that the Linux solution is "cleaner". I'm not even sure that I am claiming that. However, I can point you in the direction of a roughly equal number of people, all audio engineering professionals, who find the Linux solution compelling for a variety of reasons. Whose vote wins? Who even cares? Certainly not me. Maybe it's cleaner from a programmer perspective, a developer perspective and maybe in theory ie: crunchy code and all that gibberish. However I can tell you that if Linux were the only software choice for a DAW we would all still be using tape and a razor blade. On this point, you're sadly delusional. There are many valid criticisms to be made of Linux as a platform for audio engineering professionals. Many. This is not even close to one of them. Try that under Linux. Oh yea, if you have *SoundBlaster, it will probably work fine. Or any RME interface except the Fireface's (coming soon), or any ice1712/1724 device, or any of the following firewire devices: ECHO AudioFire 2 ECHO AudioFire 4 ECHO AudioFire 8 ECHO AudioFire 12 Edirol FA-101 Edirol FA-66 ESI Quatafire 610 Focusrite Saffire Focusrite Saffire LE Focusrite Saffire Pro 26 I/O Focusrite Saffire Pro 10 I/O Mackie Onyx Mixer FireWire Option Terratec Producer Phase 88 Rack FireWire TerraTec Producer Phase24 TerraTec Producer Phase X24 TerraTec Producer MIC 2/MIC 8 FireWire Of course, that is subject to my previous comments about choosing the right linux distribution. And if that's confusing to you, that is because you don't understand that "Linux" isn't a unitary thing. If you can't deal with this idea, then give up. Mike's venture into this is a prime example. I followed Mike's venture into this and interacted with him a lot on r.a.p as it happened. Mike went into his venture with a particular set of expectations and prior experience. His goals and requirements were not met, that was clear. However, his prior experience and his model of how things should be made it very much harder for it to work. I've had that discussion once (I think I wrote at least 20 posts on this at the time) You Linux developers are all doing your own thing and apparently nobody is talking to the other people. You have no idea who I talk to, how or where. Why do you insist on speculating, no *asserting* stuff that you simply don't know? Example: Pulseaudio. PulseAudio is a consumer/desktop audio architecture targetting consumer media workflows and applications. It is also very focused on low power/mobile devices, a market where Linux is growing increasingly dominant, and the requirements for the audio infrastructure are entirely different. If you are attempting to do pro-audio or music creation using PulseAudio, then you've already done way down the wrong track. Is it a problem that this is not more obvious from the outset? Sure. Is it partly related to what I said earlier about Linux not being a unitary thing in the way that Windows or OS X is? Absolutely. Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So why is that? If ALSA works fine, and it does IMHO, why all the others? ALSA is primarily a set of device drivers and a thin abstraction over them. That doesn't make it (necessarily) the appropriate API for developers to use when writing applications. Oh yea, "choice"... No, it's not choice. It's a cluster****. Sure. That's why people who are using ALSA drivers they created for data collection (i.e. non-pro-audio, just massively high sample rates) are finding it a cluster****. Once again, you fail to grasp that linux isn't a unitary thing. Its a way for people to put together systems that can do certain tasks much better than most other operating systems. Does that mean that when you install J. Random Linux Distro on your J. Random Hardware that it will make a superb DAW? It most certainly does not. For a report on my perspective, this is useful:http://lwn.net/Articles/355542/ Fair enough... My personal feeling is ALSA should be tightly integrated with the kernel It is. (I'm not a programmer so please excuse the simple language) and the others should disappear. If you talk to any audio developer who has written applications for JACK, CoreAudio, ASIO, WDM and ALSA (to name a few), I am fairly confident that they will tell you that JACK provides the simplest, easiest to use and most powerful abstraction for audio of any of them. None of these other systems provides the same kinds of capabilities, or the same ease of development. They don't make it possible to make arbitrary applications do arbitrary things. Ergo, unless ALSA were to incorporate all these things, its not likely that JACK is going to go away, for the same reason that JACK doesn't "go away" on OS X - CoreAudio just doesn't provide the same stuff. Same with any Windows audio API. Put it this way, if I had you in my studio for a couple of hours, I am certain you could show me things that Jack/ALSA etc can do, that I could not even in my wildest imagination conceive. But I don't have you and documentation is poor so I am left to mostly fend on my own, like most people. Sure, the documentation is not good. But its improving, and unlike the code, is an area where actual audio engineers could contribute a lot. Do they? A few. Ivory. Garritan Addictive Drums UAD Sonnex Nuendo. ProTools. Do they run under Linux? Sample libraries that use Gigasampler and a few other formats: yes. Pianoteq (a physically modelled piano that to my ears is more powerful than any sample library): yes (and windows and OS X too) Ardour: yes many VST plugins: actually yes. i helped make this possible, but i can't say i recommend it. so, your chosen tools don't support Linux (see notes above). how is this linux' fault? does it make an unsuitable platform for you? probably. but so what? I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? Because most of them are as confused and ignorant about it as you. Their/your fault? Not entirely, no. |
#88
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That before WDM, there was no reasonable "out of the box" low latency solution on that platform? That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? The situation is certainly cleaner than Linux, but it hardly gets close to the cleanliness (for the user) of CoreAudio. Well WDM and ASIO are real and work, for the most part, just fine now and for the last 5 years. geoff |
#89
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 15:53:27 -0800 (PST), dawhead wrote:
On Dec 23, 6:18*pm, Little****t wrote: They most certainly do not want to be playing with operating systems like Linux forces the user to do. Then for the time being, at least, they probably shouldn't bother. Most don't, hence the popularity of the Mac. What I *do know* is that I plug in my audio interface, under Windows 7 x64 BTW, my control surface, select ASIO (my choice) and it all works. I never have to touch a single thing other than latency. What I do know is that I plug in my audio interface (an RME hdsp9652) under Linux (Fedora 11 x86_64), plugin in my control surfaces, start JACK and it all works. I never have to touch anything except hdspmixer to reset the initial matrix mixer settings for the hdsp, and JACK to adjust latency. You are cherry picking. Try a Euphonix. Try an Alesis. Try the Cakewalk/Roland V7xx systems. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. I fire up Ardour and it all works. My 10 year old Soundblaster Live! works with Linux as well. Your point? All my plugins, well most of my plugins work. All of my plugins, well most of my plugins work. Linux doesn't even support this stuff. Wrong way around. The developers of those plugins don't support Linux. Can you run Ardour on Windows? Same deal. Linux' job isn't to support pre-existing software, just like OS X's job isn't to support Windows software. Clear thinking about where responsibilities lie here would help. Doesn't matter to the end user. The end user wants the stuff he reads about in the trade rags to work. While some of this stuff is hyped junk, I agree, a lot of it is very useful and used every single day to produce what you hear in the media. And you are deluding yourself if you think for one instant that a professional, like myself or Mike or Scott or Hank or Fletcher will for even an instant claim that the Linux solution is "cleaner". Fortunately, I am not deluding myself for one moment into thinking that any professionals like yourself, Mike or the other people you mention (the last 3 I do not know) will claim that the Linux solution is "cleaner". I'm not even sure that I am claiming that. However, I can point you in the direction of a roughly equal number of people, all audio engineering professionals, who find the Linux solution compelling for a variety of reasons. Whose vote wins? Who even cares? Certainly not me. I've yet to work in a studio that is running Linux other than as a back up system/server or in a Receptor module which I believe (correct me if I am wrong) runs Linux as it's base system. I work in NYC mostly and it's all ProTools or Nuendo. Smaller studios, like jingle houses are running everything from Cubase to Logic to Reaper and even Sonar. Maybe it's cleaner from a programmer perspective, a developer perspective and maybe in theory ie: crunchy code and all that gibberish. However I can tell you that if Linux were the only software choice for a DAW we would all still be using tape and a razor blade. On this point, you're sadly delusional. There are many valid criticisms to be made of Linux as a platform for audio engineering professionals. Many. This is not even close to one of them. So then why aren't people flocking to Linux? The OP makes a valid point. Try that under Linux. Oh yea, if you have *SoundBlaster, it will probably work fine. Or any RME interface except the Fireface's (coming soon), or any ice1712/1724 device, or any of the following firewire devices: ECHO AudioFire 2 ECHO AudioFire 4 ECHO AudioFire 8 ECHO AudioFire 12 Edirol FA-101 Edirol FA-66 ESI Quatafire 610 Focusrite Saffire Focusrite Saffire LE Focusrite Saffire Pro 26 I/O Focusrite Saffire Pro 10 I/O Mackie Onyx Mixer FireWire Option Terratec Producer Phase 88 Rack FireWire TerraTec Producer Phase24 TerraTec Producer Phase X24 TerraTec Producer MIC 2/MIC 8 FireWire Of course, that is subject to my previous comments about choosing the right linux distribution. And if that's confusing to you, that is because you don't understand that "Linux" isn't a unitary thing. If you can't deal with this idea, then give up. That's a small subset of what is out there and again what level of support is given and how well is it documented? Do all of the features work? Mike's venture into this is a prime example. I followed Mike's venture into this and interacted with him a lot on r.a.p as it happened. Mike went into his venture with a particular set of expectations and prior experience. His goals and requirements were not met, that was clear. However, his prior experience and his model of how things should be made it very much harder for it to work. I've had that discussion once (I think I wrote at least 20 posts on this at the time) I read it in Google after the fact so some stuff might have been missing, but I would classify Mike as your poster child for Linux acceptance. He is very, very typical of the genre. You Linux developers are all doing your own thing and apparently nobody is talking to the other people. You have no idea who I talk to, how or where. Why do you insist on speculating, no *asserting* stuff that you simply don't know? He makes a point. Why so many versions of Linux? Why doesn't one work with the other? Why 5 different package manageers? Why do some systems use Xorg.conf and others do not? Why real time kernels? etc. It's a mess. And it's hindering Linux. You see only your little world, and that's ok but Joe user sees it a different way. Example: Pulseaudio. PulseAudio is a consumer/desktop audio architecture targetting consumer media workflows and applications. It is also very focused on low power/mobile devices, a market where Linux is growing increasingly dominant, and the requirements for the audio infrastructure are entirely different. But it's still there. When Joe user installs his Ubuntu, he sees Pulseaudio and I can almost assure you that he will have troubles. In fact, with Ubuntu, he will have troubles with 9.10 and ICE1712 right from the start because one of the config files is missing a line. I can't believe they even test that garbage and pleas don't get me going on Ubuntu because it sucks IMHO. Debian great. Redhat great. Fedora, not bad. SuSue pretty good. Ubuntu,,,sucks... If you are attempting to do pro-audio or music creation using PulseAudio, then you've already done way down the wrong track. Is it a problem that this is not more obvious from the outset? Sure. Is it partly related to what I said earlier about Linux not being a unitary thing in the way that Windows or OS X is? Absolutely. And how is Joe user supposed to know this? He installs "Linux" whatever Linux it may be, tries to run Ardour and is baffled. Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So it will work without ALSa installed? Will you get the same latency specs? Performance? etc? So why is that? If ALSA works fine, and it does IMHO, why all the others? ALSA is primarily a set of device drivers and a thin abstraction over them. That doesn't make it (necessarily) the appropriate API for developers to use when writing applications. From and end user perspective, ALSA works. Pulseaudio does not, in many cases. The net is full of problems with sound that disabling Pulseaudio fixes. Oh yea, "choice"... No, it's not choice. It's a cluster****. Sure. That's why people who are using ALSA drivers they created for data collection (i.e. non-pro-audio, just massively high sample rates) are finding it a cluster****. Once again, you fail to grasp that linux isn't a unitary thing. Its a way for people to put together systems that can do certain tasks much better than most other operating systems. Does that mean that when you install J. Random Linux Distro on your J. Random Hardware that it will make a superb DAW? It most certainly does not. I'm talking about professional audio/DAW work. IOW recording the next Aerosmith CD. Not some scientist in a lab recording the sounds of carpenter ants mating. Wonder if they can do "Close to You", hahaha! For a report on my perspective, this is useful:http://lwn.net/Articles/355542/ Fair enough... My personal feeling is ALSA should be tightly integrated with the kernel It is. Yes, but why RT kernels and normal kernels. More confusion. (I'm not a programmer so please excuse the simple language) and the others should disappear. If you talk to any audio developer who has written applications for JACK, CoreAudio, ASIO, WDM and ALSA (to name a few), I am fairly confident that they will tell you that JACK provides the simplest, easiest to use and most powerful abstraction for audio of any of them. None of these other systems provides the same kinds of capabilities, or the same ease of development. They don't make it possible to make arbitrary applications do arbitrary things. Ergo, unless ALSA were to incorporate all these things, its not likely that JACK is going to go away, for the same reason that JACK doesn't "go away" on OS X - CoreAudio just doesn't provide the same stuff. Same with any Windows audio API. The problem is the end user can't figure out how to make it work. Put it this way, if I had you in my studio for a couple of hours, I am certain you could show me things that Jack/ALSA etc can do, that I could not even in my wildest imagination conceive. But I don't have you and documentation is poor so I am left to mostly fend on my own, like most people. Sure, the documentation is not good. But its improving, and unlike the code, is an area where actual audio engineers could contribute a lot. Do they? A few. Point taken! Ivory. Garritan Addictive Drums UAD Sonnex Nuendo. ProTools. Do they run under Linux? Sample libraries that use Gigasampler and a few other formats: yes. Pianoteq (a physically modelled piano that to my ears is more powerful than any sample library): yes (and windows and OS X too) Ardour: yes many VST plugins: actually yes. i helped make this possible, but i can't say i recommend it. Giga is dead.... so, your chosen tools don't support Linux (see notes above). how is this linux' fault? does it make an unsuitable platform for you? probably. but so what? These are tools I can assure you are found in 99 percent of professional studios and only scratch the surface. I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? Because most of them are as confused and ignorant about it as you. Their/your fault? Not entirely, no. I'm not confused. I'm speaking from a layperson's POV. You are speaking from a programmer/developer's POV. It's obvious. I would suggest you spend some time talking to real musicians, engineers etc who are making the music you hear on the radio. Ask them what they need. Spend some time on Gearslutz, KVR and other sites. The Harrison Mixbus IMHO is a great start and is getting pretty decent reviews from the masses. This is great for you. Keep the momentum going, improve the doc and release a Windows version and if it catches on, you very well might have the next Presonous Studio One which is also surprising a lot of people with it's ease of use and quality. That's my 2 cents. |
#90
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Thu, 24 Dec 2009 13:19:57 +1300, geoff wrote:
dawhead wrote: Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That before WDM, there was no reasonable "out of the box" low latency solution on that platform? That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? The situation is certainly cleaner than Linux, but it hardly gets close to the cleanliness (for the user) of CoreAudio. Well WDM and ASIO are real and work, for the most part, just fine now and for the last 5 years. geoff It's more programming crap while ignoring the practical end of things. These people seem to ignore the fact that currently this stuff does work and probably will for some time to come. So what if WDM is depreciated? One mouse click and I am running ASIO. No big deal to me. |
#91
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Little****t wrote:
On Wed, 23 Dec 2009 14:20:18 -0800 (PST), dawhead wrote: Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That before WDM, there was no reasonable "out of the box" low latency solution on that platform? That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? The situation is certainly cleaner than Linux, but it hardly gets close to the cleanliness (for the user) of CoreAudio. You're kidding right? What I *do know* is that I plug in my audio interface, under Windows 7 x64 BTW, my control surface, select ASIO (my choice) and it all works. I never have to touch a single thing other than latency. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. All my plugins, well most of my plugins work. Some, like Toontrack are having some growing pains with Windows 7 x64, but that's coming shortly. Linux doesn't even support this stuff. And you are deluding yourself if you think for one instant that a professional, like myself or Mike or Scott or Hank or Fletcher will for even an instant claim that the Linux solution is "cleaner". Maybe it's cleaner from a programmer perspective, a developer perspective and maybe in theory ie: crunchy code and all that gibberish. However I can tell you that if Linux were the only software choice for a DAW we would all still be using tape and a razor blade. Try that under Linux. Oh yea, if you have SoundBlaster, it will probably work fine. SNIP Then why are you using computers? What is it that leads you to expect that a general purpose operating system (windows/OSX/linux) on a general purpose CPU on a general purpose motherboard is a sensible way to build a tool that will let you "just make music" ? Ivory. Garritan Addictive Drums UAD Sonnex Nuendo. ProTools. Do they run under Linux? That's where Linux loses, at least for me. You see, you look at the computer from a programming POV. I look at it from a musician's POV. Big difference. For me, the computer is a gigantic tape drive and the editing is the mother of all splicing blocks. The effects are the racks of equipment I used to have (I still have some BTW) or the units I could not afford back in the 70's. Ivory is the Steinway, Bosendorfer, Yammy etc I cannot fit in my living room as well as the superb mics needed to record such fine instruments I could not afford. Actually I have a Steinway B now and some very nice mics, but the acoustics in my living room where the piano is are marginal. Ivory or Garritan is a much better choice in MOST situations. It goes on and on. I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? Hello, Flatfish. -- HPT |
#92
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
This is not the case. Is this Ben? We worried a great deal about the reliance on JACK. It turns out to have been a boon to many of the initial Mixbus users because rather than grappling with the full DAW capabilities it has (because its based on Ardour), they were able to route directly out of Logic (or whatever other CoreAudio compatible software they were using) into Mixbus and just use it "as a console". That's sort of what I said, isn't it? The promo material describes it as a mixing console, though I do recognize Ardour in there, and that's just from the screen shots. (I don't have a Mac) However, this was purely a (positive) side effect of our use of JACK - Mixbus was not ever conceived as some kind of "addon" or "frontend" for other DAWs. Its just that the kind of modular environment that I've worked at developing for the last 10 years makes this particular workflow possible. Now you're talking like a Linux booster. What's wrong with the workflow that's been pretty well established over the past ten years for a DAW? People know how to do that. Why hand them a construction set when most potential users would just want to set it up like a conventional DAW anyway? Users who have bothered to read the documentation, or even more so, watch the video on Mixbus, have understood that its actually a full- featured DAW (without MIDI editing/recording/playback), not just a mix engine. The fact that it is usable as "just" the latter is a side effect of Ardour's design, not a design decision. I was actually looking for that, and I'm not surprised to find that it's the case. However, the PERCEPTION based on the introductory material is that it's a mixing console that's more intuitive to use than what's typically found in a DAW. I think that's a great idea, but it's really a pretty small step to add a recorder to it, particularly when it's already in the basic code. This is just another example of the confusion about Linux audio software that's a result of its flexibility and the lack of focus toward a functional task-oriented end product by the purveyors. there is no hardware supported on OS X that does not work with JACK on OS X. If it doesn't work well with JACK (e.g. digidesign I/O) then it doesn't work well with Logic (if at all - digi's coreaudio driver for their h/w is just almost laughably non-standard in how it does just about everything.) Maybe there's been some progress since I looked at Ardour nearly a year ago. At the time, I found a very limited number of multichannel interfaces that could be used. Surely you can use the large number of interfaces that are supported by Core Audio to record in Logic and then mix those tracks in Mixbus. My impression, and perhaps this is misguided, is that Jack, which talks to Logic and Mixbus, can route the audio stream coming into Logic out to Mixbus. If this is the case, it would allow me to use any audio hardware that can talk to Logic as a front end for the recording portion of Mixbus. Ardour doesn't support hardware. Ardour doesn't interact with hardware directly at all. ALSA is the HAL for audio devices on a platform (linux) that has seen many manufacturers deliberately refuse to make driver support possible. Oh, no! Here we go again. I suggest you read it again before passing your judgement. The parts that "look like linux documentation" probably refer to aggregate devices Nope, that's the part that I understood. I wish Windows had a similar capability. CEntrance tried to come up with a universal driver that could aggregate multiple hardware devices into a single hardware device but they couldn't deal with all the testing necessary to make it into a reliable commercial product. At least Apple seems to have something that works. a lamentable state of affairs on Intel OS X caused by Apple's curious refusal to provide duplex (simultaneous playback & capture) capabilities for the builtin audio device on these systems Is that what "aggregate" is all about? My impression, and the way I've seen it applied, is when you have a collection of external audio devices and want to use them together to get more channels. |
#93
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Neil Gould wrote:
Moshe Goldfarb wrote: however the time required to make some of them work properly is all on your own dime. I think the problem is managing people's expectations. Those who think they'll get the same "appliance" level of functionality by going the Linux route are likely to be disappointed. The problem is that when people take the attitude that a computer is an appliance that they don't have to look inside and they can just accept as a black box, they are on the road toward disaster. And it might not be a disaster today, it might be a disaster many years down the road, but it will happen. If you want an appliance, buy an appliance. There are plenty of standalone recording devices out there, from a 2" Ampex through a RADAR and on down to a bunch of little portastudios from Roland and Korg. They do what they're expected to do for the most part. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#94
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Little****t wrote:
On Wed, 23 Dec 2009 14:20:18 -0800 (PST), dawhead wrote: one person's "too much interaction" is another person's "not enough control". maybe we should pay more attention to the fact that there are probably more people in the first situation than the second, or maybe we're interested in a niche market of audio professionals who actually want control. They want control of their music and the applications that record their music. They most certainly do not want to be playing with operating systems like Linux forces the user to do. They should try an Ampex, then. It works great for me, and it's billable. For the most part it doesn't break, and when it does break the documentation is exceptionally complete. Why are so very few using it at a professional level? I'm using NetBSD in the studio. It runs all the billing systems. And when it all comes down to it, accounts payable is the most important studio system there is. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#95
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 7:29*pm, Moshe Goldfarb wrote:
You are cherry picking. Try a Euphonix. They were very enthusiastic about getting support for their protocol into Ardour. Then they just dropped the ball. It may get picked up again with the advent of mixbus. Try an Alesis. An Alesis *what* ? Try the Cakewalk/Roland V7xx systems. If it generates generic MIDI CC messages or Mackie/Logic Control, it works with Ardour. If not - then sure, someone has to implement support for it. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. I fire up Ardour and it all works. My 10 year old Soundblaster Live! works with Linux as well. Your point? The same as the OP's point - citing a list of things that "just work for me" is really just anecdotal and beside the point. I could cite a long list of equipment that will work with Ardour and other lInux apps. Its not the equipment you use? Oh dear ... what does that mean? Wrong way around. The developers of those plugins don't support Linux. Can you run Ardour on Windows? Same deal. Linux' job isn't to support pre-existing software, just like OS X's *job isn't to support Windows software. Clear thinking about where responsibilities lie here would help. Doesn't matter to the end user. The problem is that enduser is mis-informed about how this stuff works, and so they form expectations. "Linux won't run my plugins". That much is true. "Linux should run my plugins". That should perhaps be true. "The responsibility to fix this rests with Linux and the communist hippie software hackers types who use it since no audio guys do". False. I've yet to work in a studio that is running Linux other than as a back up system/server or in a Receptor module which I believe (correct me if I am wrong) runs Linux as it's base system. So once again, we have your anecdotes versus someone else's versus mine. Yes, *MASSIVELY* more studios run non-Linux DAWs. But so what? MASSIVELY more studios run ProTools over Logic ... does this mean that Logic is crap? That it can't do anything? That nobody in their right mind would use it? That it does everything that ProTools does? That it has no capabilities that ProTools is lacking? No, no, no, no and no. Now substitute a specific app like Ardour in ... the answers are the same. So then why aren't people flocking to Linux? The OP makes a valid point. As I noted at the end of my last post: most of them have faulty knowledge, shallow knowlege or both about Linux. Moreover, audio engineering professionals do not FLOCK to anything. They did't flock to Pyramix, they didn't flock to Paris, they didn't flock to the Mackie HDR, they didn't flock to Nuendo - they have technological inertia to a greater extent than almost any niche user community of computer software, mostly thanks to the abject failure of the audio tech industry to ever develop adequate open standards for data interchange and the laughable proliferation of plugin APIs. (list of support audio interfaces) That's a small subset of what is out there and again what level of support is given and how well is it documented? Do all of the features work? My list was for illustrative purposes, it was not exhaustive. Sometimes, you r.a.p folk complain that you can only use soundblasters on linux - not true. Are there devices that don't work? Sure. Should you know this before trying to use one on a linux system? Probably. Is it all adequately documented? Not really. I read it in Google after the fact so some stuff might have been missing, but I would classify Mike as your poster child for Linux acceptance. He is very, very typical of the genre. I do not wish to discuss Mike's thread again. I made a lot of commentary on it in that thread, and I think I stand by everything I said there. He makes a point. Why so many versions of Linux? Why doesn't one work with the other? Why 5 different package manageers? Why do some systems use Xorg.conf and others do not? Why real time kernels? Why do you care? You care because you are expecting a certain kind of product. Right now, nobody offers the kind of product that you think you want. There, said it. Does this mean that Linux is useless as a platform for a DAW? no. It's a mess. And it's hindering Linux. You see only your little world, and that's ok but Joe user sees it a different way. I'm afraid its the other way around. *You* see only the world of pro- audio, and when someone says you can do some cool stuff with Linux, you imagine that Linux is there to cater "off the shelf" to "your little world". The problem is that Linux isn't about just "this little world", its about a lot of things. The fact that its such a massively superior technology for audio technology has a lot to do with why its so good for mobile devices and data servers - it can be made to do all these tasks well because it can be *customized*. Can you buy the customized version you want for pro-audio off the shelf? Right now, you cannot. Does this mean it can't do the job? Does this mean that those of us who do, in fact, use it for such things are delusional? no. Does this help people who might want to "just use it" ? not really. Example: Pulseaudio. PulseAudio is a consumer/desktop audio architecture targetting consumer media workflows and applications. It is also very focused on low power/mobile devices, a market where Linux is growing increasingly dominant, and the requirements for the audio infrastructure are entirely different. But it's still there. When Joe user installs his Ubuntu, he sees Pulseaudio and I can almost assure you that he will have troubles. Right, and Ubuntu is not an appropriate distribution for people who want to do pro-audio, partly (but far from entirely) for this very reason. I can't believe they even test that garbage and pleas don't get me going on Ubuntu because it sucks IMHO. Debian great. Redhat great. Fedora, not bad. SuSue pretty good. Ubuntu,,,sucks... Not of the above are appropriate base platforms for pro-audio work. If you are attempting to do pro-audio or music creation using PulseAudio, then you've already done way down the wrong track. Is it a problem that this is not more obvious from the outset? Sure. Is it partly related to what *I said earlier about Linux not being a unitary thing in the way that Windows or OS X is? Absolutely. And how is Joe user supposed to know this? He installs "Linux" whatever Linux it may be, tries to run Ardour and is baffled. Are you complaining about the technology, or the documentation? Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So it will work without ALSa installed? Sure. Will you get the same latency specs? Depends on the hardware. All other things being equal, sure. Performance? Ditto. ALSA is primarily a set of device drivers and a thin abstraction over them. That doesn't make it (necessarily) the appropriate API for developers to use when writing applications. From and end user perspective, ALSA works. Pulseaudio does not, in many cases. The net is full of problems with sound that disabling Pulseaudio fixes. Did I suggest PulseAudio as an alternative API to ALSA? My personal feeling is ALSA should be tightly integrated with the kernel It is. Yes, but why RT kernels and normal kernels. More confusion. Only for people who have read the wrong documentation or not understood what they read. Can we stop the flood of misinformed user posts on forums about this sort of thing? Its rather hard. Even on this thread alone, I've read some fundamentally wrong claims about the way Linux audio works, and I'm sure have walked away believing them. The problem is the end user can't figure out how to make it work. The problem is that a particular class of end user can't figure out how to make it work. Thats a different (though important) issue. |
#96
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Anahata wrote:
DAWs are just one example of speciality applications (CAD is another) where there'll only ever be a small population of users but the software itself is very complex. Consequently anybody investing development time on such software commercially is only going to go for the most widely used platform, or their already tiny market will become even smaller. Actually if Pro Tools, Nuendo or whatever were ported to Linux, I'm sure it would run like a champ, but there just isn't the commercial incentive to do it. What's worse is that the systems that are really designed to do this kind of job are systems that are niche products even in the Unix world. What you really want is a realtime operating system where, when you make a call to the OS, you can tell the OS how long you're willing to wait before something is completed. A true realtime operating system (and RTLinux comes reasonably close) allows you to do proper realtime stuff in a reliable fashion, and in such a way that when you run out of resources you are immediately told there is a problem instead of later discovering insidious clicks in your file. The only time anyone is ever going to develop a DAW system based on a realtime base is going to be in an environment where they can design the system as a whole. Oh, did I mention RADAR? --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#97
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 7:29*pm, Moshe Goldfarb wrote:
I would suggest you spend some time talking to real musicians, engineers etc who are making the music you hear on the radio. Ask them what they need. Spend some time on Gearslutz, KVR and other sites. As noted above, I put it to you that you have little idea how I spend time communicating with (actual or potential) users. If you visit gearslutz and read the mixbus thread(s), check the poster nicks. compare. ditto on KVR. you might ponder what i'm even doing on this thread. The Harrison Mixbus IMHO is a great start and is getting pretty decent reviews from the masses. This is great for you. Keep the momentum going, improve the doc and release a Windows version and there are no plans for a windows version. if it catches on, you very well might have the next Presonous Studio One which is also surprising a lot of people with it's ease of use and quality. as is mixbus, without the windows masses. a problem for us? perhaps. we'll see. |
#98
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On 23 Dec 2009 20:46:21 -0500, Scott Dorsey wrote:
Neil Gould wrote: Moshe Goldfarb wrote: however the time required to make some of them work properly is all on your own dime. I think the problem is managing people's expectations. Those who think they'll get the same "appliance" level of functionality by going the Linux route are likely to be disappointed. The problem is that when people take the attitude that a computer is an appliance that they don't have to look inside and they can just accept as a black box, they are on the road toward disaster. And it might not be a disaster today, it might be a disaster many years down the road, but it will happen. Ten years ago I might have agreed, hell even now *my philosophy* is the same as yours Scott. Times have changed. For better or worse. It's called a throw away society. How many engineers really know what is going on behind their PTHD system? Very few. They are interested in tracking. How man artists know what is going on with autotune? They sing off key and it gets "fixed" that's all they care about. This can be extrapolated to just about anything, for better or worse. Who cares how an iPod works? We plug it in and it syncs...who really cares? Most people do not. If you want an appliance, buy an appliance. There are plenty of standalone recording devices out there, from a 2" Ampex through a RADAR and on down to a bunch of little portastudios from Roland and Korg. They do what they're expected to do for the most part. --scott The PC *IS* an appliance. It has reached that level, like it or not. The days of nuts and bolts are long gone. Example: I can build you a stellar computer for your DAW. Guess what? You can buy one already made for less. Sure you are getting lower quality components, say China sys board vs Asus, but in reality it will work as well as your custom job and in a couple of years, or less, both will be obsolete so who cares? Understand, I am a person who has built systems since 1981 whien the original IBM PC was released. Sure, my build is of higher quality, but in reality what does that really mean? Most hardware will last it's useful life. The computer has become an appliance, like it or not. People run applications not chips and bits and their off brand system will in most cases run them just as good as your custom built high end brand for the life of the hardware. |
#99
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On 23 Dec 2009 20:49:16 -0500, Scott Dorsey wrote:
Little****t wrote: On Wed, 23 Dec 2009 14:20:18 -0800 (PST), dawhead wrote: one person's "too much interaction" is another person's "not enough control". maybe we should pay more attention to the fact that there are probably more people in the first situation than the second, or maybe we're interested in a niche market of audio professionals who actually want control. They want control of their music and the applications that record their music. They most certainly do not want to be playing with operating systems like Linux forces the user to do. They should try an Ampex, then. It works great for me, and it's billable. For the most part it doesn't break, and when it does break the documentation is exceptionally complete. As long as it it isn't vintage 456 bg ! Unless they have an oven of course! Why are so very few using it at a professional level? I'm using NetBSD in the studio. It runs all the billing systems. And when it all comes down to it, accounts payable is the most important studio system there is. --scott And many people do. I run Linux in my studio as well as my back up system. My home systems are tied to a Linux server. My DAW runs Nuendo and PTHD. |
#100
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Moshe Goldfarb wrote:
However, Ivory is the "best" piano sound/playability out there IMHO although I hear the new Steinberg Grand 3 is supposed to be outstanding. Then there is my UAD card. So I'm kind of out of luck with Linux. I get great piano and organ sounds out of my GeneralMusic WK2HD Midi Arranger and Roland GW-7 keyboard. Why would I want to go UAD? -- HPT |
#101
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On 23 Dec 2009 20:53:02 -0500, Scott Dorsey wrote:
Anahata wrote: DAWs are just one example of speciality applications (CAD is another) where there'll only ever be a small population of users but the software itself is very complex. Consequently anybody investing development time on such software commercially is only going to go for the most widely used platform, or their already tiny market will become even smaller. Actually if Pro Tools, Nuendo or whatever were ported to Linux, I'm sure it would run like a champ, but there just isn't the commercial incentive to do it. What's worse is that the systems that are really designed to do this kind of job are systems that are niche products even in the Unix world. What you really want is a realtime operating system where, when you make a call to the OS, you can tell the OS how long you're willing to wait before something is completed. A true realtime operating system (and RTLinux comes reasonably close) allows you to do proper realtime stuff in a reliable fashion, and in such a way that when you run out of resources you are immediately told there is a problem instead of later discovering insidious clicks in your file. The only time anyone is ever going to develop a DAW system based on a realtime base is going to be in an environment where they can design the system as a whole. Oh, did I mention RADAR? --scott The problem is niche systems are yesterdays news. Even PTHD or other hardware based systems, Creamware etc. Let's face it, the hardware processing advantages of a PTHD system are being eroded by the day by faster processing, cheaper software etc. The only reason PTHD survives today is because of the installed base and the fact it is the "defacto" standard in the biz. Digi realizes their days are numbered henceforth all the trade up offers they are offering. |
#102
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
one person's "too much interaction" is another person's "not enough control". maybe we should pay more attention to the fact that there are probably more people in the first situation than the second, or maybe we're interested in a niche market of audio professionals who actually want control. There's nothing at all wrong with Group #2. But you seem to have at least initially grabbed the attention of those in Group #1. Even me, and I don't even have a Mac. Really. A fascinating assesment of the windows situation. You know that WDM is deprecated in Windows7, yes? "Deprecated" as in "stable, so soon to become obsolete?" But few pro DAW users are using WDM drivers. That ASIO was never provided by Microsoft and always relied on 3rd party drivers which didn't always track the latest version of Windows precisely? That, indeed, is a problem, but it's as much a Microsoft problem as with any other vendor. Why does Windows change in ways that affect the hooks around which ASIO is built? Steinberg recognized that there was a problem with WDM that stood in the way of their DAWs working as well as they wanted them to, and they came up with a solution, along with somewhat of a standard to which others could write drivers for their hardware. That before WDM, there was no reasonable "out of the box" low latency solution on that platform? WDM is "low latency?" It's better than MME, but not as good as ASIO, nor is it adjustable to take adavantage of a well tuned computer. That most windows consumer desktop applications used MME, not ASIO, for playback and capture, which could often conflict with ASIO use of the audio interface? We (the professional "we") recommend that a computer that's used as a DAW be devoted to DAW use and not get corrupted with consumer desktop audio applications. It's good advice, and a cheap solution to a lot of problems. What's a mess is that so many people, including you, have read so much stuff that has led them to completely misunderstand the role of these things in the Linux "audio stack". That's because nobody, not even you, has successfully related how Linux does audio with what those working with other software understand. I've tried, really I have, but I just don't get it. There's something very fundamental that I'm missing, and I suspect that the Linux world figures "everybody knows that." I'm no expert in Windows audio connections either, but at least the terminology and user interface has always been pretty clear. Not so with Linux. But Linux is not a company. And this is part of the problem with trying to build a commercial product around it. If you want a smooth experience with Linux audio, do you randomly pick some distro, some machine, some audio interface and put them together and expect that it will all just work? What other choice do you have? Maybe not random, but you need to learn a lot about how the pieces fit together to make the right choice. For better or worse, Microsoft and Apple have made some of those choices for us and limited the scope of our choices to things that are pretty likely to work together without too much fooling around. It appears that many existing or potential Linux users do indeed expect this to be possible. Sorry, its not. Its unfortunate that so many people believe that it is, or even more irritatingly, believe that it should be. Why not? Certainly a company who wants to sell a program can specify a certain version of Linux, a certain distribution, and as long as the user sets that up as the program vendor tells him, and doesn't muck with it, it'll work. It's just like Windows. If you have something working under XP, you may indeed expect to have some problems when moving to Vista or Seven, but that doesn't happen every few weeks. There's no configuration management with a user's version of Linux, however. This is why Linux will never be the operating system of choice for people who want to run an application and not fool around with things that nobody else is doing. audio forums for Windows DAWs are full of testimonies to problems that people have with particularly bad combinations of choices. If you want that kind of experience, you need to get your system from a company that controls everything end-to-end, which means either Apple or a company that specializes in building machines for media work that run Linux. Unfortunately, I can't recommend any of them at this particular point in time. So how is it that you can buy a retail copy of Windows, load it up on just about any computer that meets a certain minimum specification, and have it all work? The CPU works, the hard drives work, the optical drives work, the USB ports work, the networking works. Plug in a Firewire card and Windows knows what it is and how to talk to it. Connect a Firewire audio interface and, when you install the driver that's provided by the manufacturer (and you have the major version of Windows for which that driver is written) it'll work. The only part of that game that's really different in Linux is that, for example, when you buy a PreSounus interface, you can't stick the CD that comes with it into the drive and have the driver for that interface installed magically. The procedure is different - it can be learned, but it's not documented with the new purchase, you need to know where to find it in the Linux world. And then your hardware may or may not be supported depending on whether someone got around to it. And it may be only halfway supported - like it'll pass audio but you won't have a control panel for the built-in mixer. Maybe this sounds like a cop out but all I want to do is make music. Then why are you using computers? What is it that leads you to expect that a general purpose operating system (windows/OSX/linux) on a general purpose CPU on a general purpose motherboard is a sensible way to build a tool that will let you "just make music" ? I've tried that argument. The brutal fact of life is that today, recording has become an integral part of the work of a hobbyist and even professional musician. Musicians don't have a lot of money, and a computer is an inexpensive way to put together a functional recording system. If they could get it cheaper by using open source software that they didn't have to pay for in dollars, they would - but only if they could get the same results with about the same level of effort. Many times, I've recommended a simple hardware workstation (some of them cost less than a computer) but it's a hard sell. |
#103
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
On Dec 23, 7:29 pm, Moshe Goldfarb wrote: I would suggest you spend some time talking to real musicians, engineers etc who are making the music you hear on the radio. Ask them what they need. Spend some time on Gearslutz, KVR and other sites. As noted above, I put it to you that you have little idea how I spend time communicating with (actual or potential) users. If you visit gearslutz and read the mixbus thread(s), check the poster nicks. compare. ditto on KVR. you might ponder what i'm even doing on this thread. The Harrison Mixbus IMHO is a great start and is getting pretty decent reviews from the masses. This is great for you. Keep the momentum going, improve the doc and release a Windows version and there are no plans for a windows version. If it's any good you could be rich. Or if that's not important, have the satisfaction of many satisfied users. geoff |
#104
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Les Cargill wrote:
There really needs to be somebody doing value added, to cover up the "control" from those who don't need it. I don't see any reason why someone couldn't come up with a whole package dedicated to audio recording that was based on Linux. Just bring your own computer (minimum configuration specified). But because of all the variations in Linux that are around, in order for it to be guaranteed successful, it would have to be essentially a closed system. And not being a closed system is what sells people on Windows DAWs. |
#105
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 17:52:00 -0800 (PST), dawhead wrote:
On Dec 23, 7:29*pm, Moshe Goldfarb wrote: You are cherry picking. Try a Euphonix. They were very enthusiastic about getting support for their protocol into Ardour. Then they just dropped the ball. It may get picked up again with the advent of mixbus. Try an Alesis. An Alesis *what* ? Mastercontrol http://www.alesis.com/mastercontrol Try the Cakewalk/Roland V7xx systems. If it generates generic MIDI CC messages or Mackie/Logic Control, it works with Ardour. If not - then sure, someone has to implement support for it. Point is, can someone plug it in and have it work? That's the big question with Linux. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. I fire up Ardour and it all works. My 10 year old Soundblaster Live! works with Linux as well. Your point? The same as the OP's point - citing a list of things that "just work for me" is really just anecdotal and beside the point. I could cite a long list of equipment that will work with Ardour and other lInux apps. Its not the equipment you use? Oh dear ... what does that mean? The equipment, plugins etc I use are industry standards. Do they work with Linux? Most do not. Linux loses before it even starts. Wrong way around. The developers of those plugins don't support Linux. Can you run Ardour on Windows? Same deal. Linux' job isn't to support pre-existing software, just like OS X's *job isn't to support Windows software. Clear thinking about where responsibilities lie here would help. Doesn't matter to the end user. The problem is that enduser is mis-informed about how this stuff works, and so they form expectations. "Linux won't run my plugins". That much is true. "Linux should run my plugins". That should perhaps be true. "The responsibility to fix this rests with Linux and the communist hippie software hackers types who use it since no audio guys do". False. Who cares? The programmers? The users will go elsewhere. And they do. I see what you are saying but in reality it doesn't work that way for most people. I've yet to work in a studio that is running Linux other than as a back up system/server or in a Receptor module which I believe (correct me if I am wrong) runs Linux as it's base system. So once again, we have your anecdotes versus someone else's versus mine. Yes, *MASSIVELY* more studios run non-Linux DAWs. But so what? MASSIVELY more studios run ProTools over Logic ... does this mean that Logic is crap? That it can't do anything? That nobody in their right mind would use it? That it does everything that ProTools does? That it has no capabilities that ProTools is lacking? No, no, no, no and no. Now substitute a specific app like Ardour in ... the answers are the same. Logic can run most if not all of the popular plugins. So can Nuendo. So can Sonar. So can Reaper. etc.... Can Linux/ardour? If not, you lose....... Any program, even the bargain basement ones included with low end sound cards can record digital audio. It goes, far beyond that in a professional situation. So then why aren't people flocking to Linux? The OP makes a valid point. As I noted at the end of my last post: most of them have faulty knowledge, shallow knowlege or both about Linux. Moreover, audio engineering professionals do not FLOCK to anything. They did't flock to Pyramix, they didn't flock to Paris, they didn't flock to the Mackie HDR, they didn't flock to Nuendo - they have technological inertia to a greater extent than almost any niche user community of computer software, mostly thanks to the abject failure of the audio tech industry to ever develop adequate open standards for data interchange and the laughable proliferation of plugin APIs. (list of support audio interfaces) There is a tendency toward snobbery and most flock to Protools, but you also have the Prosumer, amateur muscian types with project studios who can't afford PTHD. Why are they ignoring Linux? Or better yet, why are they attempting to run Reaper under Wine/Linux? And it runs rather well from what I have read. Why not run ardour/audacity/Rosegarden natively? Think about it..... That's a small subset of what is out there and again what level of support is given and how well is it documented? Do all of the features work? My list was for illustrative purposes, it was not exhaustive. Sometimes, you r.a.p folk complain that you can only use soundblasters on linux - not true. Are there devices that don't work? Sure. Should you know this before trying to use one on a linux system? Probably. Is it all adequately documented? Not really. For the record my ICE1712 devices have had excellen Linux support. RME is another example etc. But support has been a long train running. These cards are released with Windows and OSX support. Linux always lags. The one exception may be M-Audio which is going down the tank lately and lagging badly in Windows 7 x64 support. See their forums for the horror stories. I will not buy another one of their products BTW. I read it in Google after the fact so some stuff might have been missing, but I would classify Mike as your poster child for Linux acceptance. He is very, very typical of the genre. I do not wish to discuss Mike's thread again. I made a lot of commentary on it in that thread, and I think I stand by everything I said there. Fair enough. I suspect the entire thread I read wasn't in Google. He makes a point. Why so many versions of Linux? Why doesn't one work with the other? Why 5 different package manageers? Why do some systems use Xorg.conf and others do not? Why real time kernels? Why do you care? You care because you are expecting a certain kind of product. Right now, nobody offers the kind of product that you think you want. There, said it. Does this mean that Linux is useless as a platform for a DAW? no. I care because I expect things to just work. I install Windows 7 and most everything just works. For the Mac user it's even easier. It's a mess. And it's hindering Linux. You see only your little world, and that's ok but Joe user sees it a different way. I'm afraid its the other way around. *You* see only the world of pro- audio, and when someone says you can do some cool stuff with Linux, you imagine that Linux is there to cater "off the shelf" to "your little world". The problem is that Linux isn't about just "this little world", its about a lot of things. The fact that its such a massively superior technology for audio technology has a lot to do with why its so good for mobile devices and data servers - it can be made to do all these tasks well because it can be *customized*. Can you buy the customized version you want for pro-audio off the shelf? Right now, you cannot. Does this mean it can't do the job? Does this mean that those of us who do, in fact, use it for such things are delusional? no. Does this help people who might want to "just use it" ? not really. When I think Pro-Audio the last thing that comes to mind is Linux. Sorry, but that's how it is. Millions of users, people who use DAW software every day, will tell you the same thing. Example: Pulseaudio. PulseAudio is a consumer/desktop audio architecture targetting consumer media workflows and applications. It is also very focused on low power/mobile devices, a market where Linux is growing increasingly dominant, and the requirements for the audio infrastructure are entirely different. But it's still there. When Joe user installs his Ubuntu, he sees Pulseaudio and I can almost assure you that he will have troubles. Right, and Ubuntu is not an appropriate distribution for people who want to do pro-audio, partly (but far from entirely) for this very reason. So why do they offer UbuntuStudio? I can't believe they even test that garbage and pleas don't get me going on Ubuntu because it sucks IMHO. Debian great. Redhat great. Fedora, not bad. SuSue pretty good. Ubuntu,,,sucks... Not of the above are appropriate base platforms for pro-audio work. So what is? If you are attempting to do pro-audio or music creation using PulseAudio, then you've already done way down the wrong track. Is it a problem that this is not more obvious from the outset? Sure. Is it partly related to what *I said earlier about Linux not being a unitary thing in the way that Windows or OS X is? Absolutely. And how is Joe user supposed to know this? He installs "Linux" whatever Linux it may be, tries to run Ardour and is baffled. Are you complaining about the technology, or the documentation? In theory the technology may work but Joe will never figure out how to make it work. Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So it will work without ALSa installed? Sure. So you can use Pulseaudio? I did not know that. Will you get the same latency specs? Depends on the hardware. All other things being equal, sure. Ok Performance? Ditto. Ok ALSA is primarily a set of device drivers and a thin abstraction over them. That doesn't make it (necessarily) the appropriate API for developers to use when writing applications. From and end user perspective, ALSA works. Pulseaudio does not, in many cases. The net is full of problems with sound that disabling Pulseaudio fixes. Did I suggest PulseAudio as an alternative API to ALSA? No. But that's what they are going to have installed when they install their favorite Linux distribution. My personal feeling is ALSA should be tightly integrated with the kernel It is. Yes, but why RT kernels and normal kernels. More confusion. Only for people who have read the wrong documentation or not understood what they read. Can we stop the flood of misinformed user posts on forums about this sort of thing? Its rather hard. Even on this thread alone, I've read some fundamentally wrong claims about the way Linux audio works, and I'm sure have walked away believing them. When people search for solutions to their DAW Linux problems this is what they will encounter. Like it or not. The problem is the end user can't figure out how to make it work. The problem is that a particular class of end user can't figure out how to make it work. Thats a different (though important) issue. Yet these same people can install Windows, Nuendo, Ozone, Tracks etc and it all just works for them. Doesn't make sense. |
#106
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 19:01:43 -0700, High Plains Thumper wrote:
Moshe Goldfarb wrote: However, Ivory is the "best" piano sound/playability out there IMHO although I hear the new Steinberg Grand 3 is supposed to be outstanding. Then there is my UAD card. So I'm kind of out of luck with Linux. I get great piano and organ sounds out of my GeneralMusic WK2HD Midi Arranger and Roland GW-7 keyboard. Why would I want to go UAD? UAD has nothing to do with piano sounds. Look he http://www.uaudio.com/ |
#107
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 9:31*pm, Moshe Goldfarb wrote:
Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So it will work without ALSa installed? Sure. So you can use Pulseaudio? I did not know that. This is just one example of what I mean when I talk about people becoming confused by misleading and inaccurate information posted by people who don't actually know what they are talking about. Did anything that you have read suggest that PulseAudio is an alternative to ALSA? Do you not understand the relationship between Rewire and CoreAudio? Or Rewire and WDM/ASIO? Or for that matter, JACK and CoreAudio? Programming interfaces come in layers. ALSA is at one layer (a very low one), just like WDM is. PulseAudio is at a higher layer. One is not an alternative to the other, no more than running JACK on OS X provides an alternative to CoreAudio. I see speculation and misinformed commentaries on the linux audio stack on r.a.p and forums everywhere. I wish I could stop it. I can't. Its dismaying. Did I suggest PulseAudio as an alternative API to ALSA? No. But that's what they are going to have installed when they install their favorite Linux distribution. As I've indicated above, you still fail to understand the basic layering. Let me try to make this more clear, from the bottom up: Driver level (inside the kernel): ALSA (alternatives include OSS (deprecated for years) or FFADO which is firewire specific) HAL(*) (for applications that want intimate relationships with the hardware devices): ALSA desktop/consumer media applications: PulseAudio pro-audio/music creation: JACK PulseAudio can talk to JACK for systems that really need to run both. But as Mike Rivers noted, this is not really how you set up a DAW. In addition, many consumer/desktop media apps use additional software libraries that have different backends, enabling these apps to connect to audio hardware either "directly" via ALSA, OSS etc., or via layers like JACK or PulseAudio. And for those wondering about PulseAudio and why it has to exist: the current audio systems on Windows and OS X are totally inappropriate for use in low power situations (the iPhone, for example, runs a heavily modified version of CoreAudio for this very reason, in part at least). They waste battery power by being too focused on latency, which is not important on most low power mobile devices. Like it or not, there is a LOT of money flowing into Linux development that is focused on platforms where this kind of design consideration is not optional. As a result, Pulse is going to grow in importance over time. Does this mean that you need to run Pulse on a DAW system? Certainly not. Does it mean that you have to? Certainly not. Does it mean that Linux distributions that are focused on desktop/consumer users (i.e. almost everyone) will continue to focus on Pulse as a key part of the audio infrastructure of their systems? Yes, it almost certainly does. (*) HAL stands for Hardware Abstraction Layer |
#108
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 9:09*pm, "geoff" wrote:
dawhead wrote: On Dec 23, 7:29 pm, Moshe Goldfarb wrote: I would suggest you spend some time talking to real musicians, engineers etc who are making the music you hear on the radio. Ask them what they need. Spend some time on Gearslutz, KVR and other sites. As noted above, I put it to you that you have little idea how I spend time communicating with (actual or potential) users. If you visit gearslutz and read the mixbus thread(s), check the poster nicks. compare. ditto on KVR. you might ponder what i'm even doing on this thread. The Harrison Mixbus IMHO is a great start and is getting pretty decent reviews from the masses. This is great for you. Keep the momentum going, improve the doc and release a Windows version and there are no plans for a windows version. If it's any good you could be rich. *Or if that's not important, have the satisfaction of many satisfied users. To understand why a Windows version is not of much interest: http://www2.bryceharrington.org:8080...ss-win-paradox I've been rich. The satisfaction of the kind of Windows users that have continued to verbally attack me for years for not developing my software for Windows, let alone declaring that because it only runs on Linux (which was true once) it must be ****, is not something that I'm frankly all that interested in. Satisfying a user community that understands why I value the kinds of technology that Linux represents (and to a lesser extent OS X) and wants software to get the job done on that kind of platform - thats important to me. |
#109
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 9:13*pm, Mike Rivers wrote:
Les Cargill wrote: There really needs to be somebody doing value added, to cover up the "control" from those who don't need it. I don't see any reason why someone couldn't come up with a whole package dedicated to audio recording that was based on Linux. http://www.64studio.com/ (there are other examples. its just the best of the lot. alas, in my opinion, not good enough (yet)) Just bring your own computer (minimum configuration specified). the configuration requirements for a DAW don't really consist of minimums. they consist of things that most users don't know anything about (and don't want to know anything about). IRQ sharing ... PCI(x| e) chipset ... USB host chipset ... firewire bridge chipset ... precise version of the video driver (yes, the last one was working, the new one is broken, the next one will work again) ... the list goes on. But because of all the variations in Linux that are around, in order for it to be guaranteed successful, it would have to be essentially a closed system. 64studio is no more "closed" than any other distro. It just happens to be focused on media production. |
#110
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 19:11:10 -0800 (PST), dawhead wrote:
On Dec 23, 9:31*pm, Moshe Goldfarb wrote: Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So it will work without ALSa installed? Sure. So you can use Pulseaudio? I did not know that. This is just one example of what I mean when I talk about people becoming confused by misleading and inaccurate information posted by people who don't actually know what they are talking about. Did anything that you have read suggest that PulseAudio is an alternative to ALSA? Just about everything I have read. It's the default system in many distributions. Do you not understand the relationship between Rewire and CoreAudio? Or Rewire and WDM/ASIO? Or for that matter, JACK and CoreAudio? Programming interfaces come in layers. ALSA is at one layer (a very low one), just like WDM is. PulseAudio is at a higher layer. One is not an alternative to the other, no more than running JACK on OS X provides an alternative to CoreAudio. I see speculation and misinformed commentaries on the linux audio stack on r.a.p and forums everywhere. I wish I could stop it. I can't. Its dismaying. I don't care. I want to boot the thing and have it work. The minute you or any other Linux pundit starts talking core audio, you've lost me and 99.99 percent of all musicians. Did I suggest PulseAudio as an alternative API to ALSA? No. But that's what they are going to have installed when they install their favorite Linux distribution. As I've indicated above, you still fail to understand the basic layering. Let me try to make this more clear, from the bottom up: Driver level (inside the kernel): ALSA (alternatives include OSS (deprecated for years) or FFADO which is firewire specific) HAL(*) (for applications that want intimate relationships with the hardware devices): ALSA desktop/consumer media applications: PulseAudio pro-audio/music creation: JACK PulseAudio can talk to JACK for systems that really need to run both. But as Mike Rivers noted, this is not really how you set up a DAW. In addition, many consumer/desktop media apps use additional software libraries that have different backends, enabling these apps to connect to audio hardware either "directly" via ALSA, OSS etc., or via layers like JACK or PulseAudio. And for those wondering about PulseAudio and why it has to exist: the current audio systems on Windows and OS X are totally inappropriate for use in low power situations (the iPhone, for example, runs a heavily modified version of CoreAudio for this very reason, in part at least). They waste battery power by being too focused on latency, which is not important on most low power mobile devices. Like it or not, there is a LOT of money flowing into Linux development that is focused on platforms where this kind of design consideration is not optional. As a result, Pulse is going to grow in importance over time. Does this mean that you need to run Pulse on a DAW system? Certainly not. Does it mean that you have to? Certainly not. Does it mean that Linux distributions that are focused on desktop/consumer users (i.e. almost everyone) will continue to focus on Pulse as a key part of the audio infrastructure of their systems? Yes, it almost certainly does. (*) HAL stands for Hardware Abstraction Layer I see, and appreciate what you are saying. However, like I stated above, musicians just want this stuff to work. We don't care about the details. Just make it work and make it work better than the other guy and you have our money. |
#111
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 9:05*pm, Mike Rivers wrote:
dawhead wrote: That, indeed, is a problem, but it's as much a Microsoft problem as with any other vendor. Why does Windows change in ways that affect the hooks around which ASIO is built? Steinberg recognized that there was a problem with WDM that stood in the way of their DAWs working as well as they wanted them to, and they came up with a solution, along with somewhat of a standard to which others could write drivers for their hardware. The difference on Linux is that they never need have done that. They could have fixed the actual driver design. Mentioning Steinberg is a segue into another important part of this story. They were one of a number of companies who got very excited about the promise of BeOS. Despite not really offering much more than the best that Linux could back in the mid-to-late 1990's, many audio/ media tech companies were salivating at the prospect of an OS tailored to their (and their users' needs). They had products lined up, new marketing strategies waiting ... and then BeOS turned themselves into an internet kiosk company and then folded. I've been told by more than one executive at some of these companies that they will not get "burned again" by such ideas. This is despite the fact that I was there in 1999 talking to these same people, promoting a vision of using an operating system that has outlived (and now dramatically outperforms) BeOS, and would have offered them complete control over their own destiny because of its open source development model. Sigh. If Steinberg had actually had the vision for Linux that they were developing for BeOS, they (and others) would still be following through on it today. But it appears that once burned, twice shy applies here, even though almost everything about the technology "committment" is different. But Linux is not a company. And this is part of the problem with trying to build a commercial product around it. There's absolute no evidence of this. There are lots of commercial products built around Linux. The actual "problem", such as it is, is people in threads like this that talk about "Linux" as though *IT* was a product. many existing or potential Linux users do indeed expect this to be possible. Sorry, its not. Its unfortunate that so many people believe that it is, or even more irritatingly, believe that it should be. Why not? Certainly a company who wants to sell a program can specify a certain version of Linux, a certain distribution, and as long as the user sets that up as the program vendor tells him, and doesn't muck with it, it'll work. and will attract even less of an audience that linux' already small user base. This is why Linux will never be the operating system of choice for people who want to run an application and not fool around with things that nobody else is doing. unfortunately, it already is the operating system of choice for quite a few people who want *precisely* that. as many as run windows? does it have to be that many to reveal how pointless your comment is? So how is it that you can buy a retail copy of Windows, load it up on just about any computer that meets a certain minimum specification, and have it all work? The CPU works, the hard drives work, the optical drives work, the USB ports work, the networking works. Plug in a Firewire card and Windows knows what it is and how to talk to it. Connect a Firewire audio interface and, when you install the driver that's provided by the manufacturer (and you have the major version of Windows for which that driver is written) it'll work. with 5msec latency? I can find you forum posts for windows software that will disprove almost every claim you've made in this paragraph, except for the ones that are true for Linux too (the genera computing stuff). Linux worked on x86_64 before (and always better than) windows; it runs more hardware than windows (because it continues to support legacy equipment), and it installs easier than windows, as agreed by almost recent side-by-side install reviews in major computing magazines. What it lacks, for this context, is out-of-the- box support for pro-audio work, and this is indeed a deficiency that is not trivially addressed. oh, and sure, some of the latest and greatest gadgets that emerge without manufacturer support for linux take 4-8 months to be supported (this is less and less true these days though). I've tried that argument. The brutal fact of life is that today, recording has become an integral part of the work of a hobbyist and even professional musician. Musicians don't have a lot of money, and a computer is an inexpensive way to put together a functional recording system. If they could get it cheaper by using open source software that they didn't have to pay for in dollars, they would - but only if they could get the same results with about the same level of effort. i'm not particularly interested in specifically catering to this crowd (though if it happens, all the better). I consider audio engineering a skill set that requires learning and dedication, and part of that process involves understanding the toolset in just the same way that you used to understand a studer. the fact that a lot more people are doing this is in no way different than when desktop publishing first emerged - the result was a gazillion people producing a gazillion and a half dreadfully typeset documents. making the tools easy to use doesn't lead to good results. and i've said hundreds of times before (on usenet, going back to 1992) that i consider computers to be tools more like woodworking tools than some kind of trivial button pressing device. we've had this discussion before though, not much point in rehashing it. |
#112
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Neil Gould wrote:
I think the problem is managing people's expectations. Those who think they'll get the same "appliance" level of functionality by going the Linux route are likely to be disappointed. Why can't we expect that? All it takes is someone who will manage the system for us, or rather pre-manage it. I know that's not the normal Linuxhead's view of how it should work, but someone could do it. Though other than saving a few bucks, I don't really see the point when we have affordable functionality right now. You can (I did) buy a refurbished ready-to-go computer with Windows installed, and a copy of Reaper for about what a retail copy of Windows costs, what it can reasonably expected to do is pretty clear, and it'll do it. |
#113
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 22:38:45 -0500, Mike Rivers wrote:
Neil Gould wrote: I think the problem is managing people's expectations. Those who think they'll get the same "appliance" level of functionality by going the Linux route are likely to be disappointed. Why can't we expect that? All it takes is someone who will manage the system for us, or rather pre-manage it. I know that's not the normal Linuxhead's view of how it should work, but someone could do it. It's actually happening already like dawhead mentioned. Distributions like 64studio really do all the nuts and bolts for you. Though other than saving a few bucks, I don't really see the point when we have affordable functionality right now. You can (I did) buy a refurbished ready-to-go computer with Windows installed, and a copy of Reaper for about what a retail copy of Windows costs, what it can reasonably expected to do is pretty clear, and it'll do it. And that is the problem that Linux faces. A person can load up Reaper for little or no cost and it will support all his VST plugins and he is ready to go. |
#114
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Dec 23, 10:53*pm, Moshe Goldfarb wrote:
And that is the problem that Linux faces. A person can load up Reaper for little or no cost and it will support all his VST plugins and he is ready to go. How is this a problem for Linux? Why do you assume that in order to be successful, "Linux" (which I reiterate is neither a product nor a company) that it has to dominate everything else? Is it a problem for Logic that most studios still use ProTools? Is it a problem for Pyramix that hardly anyone uses their system? My goals with Ardour are mostly centered on providing a high quality DAW that puts audio quality and robustness just slightly ahead of workflow perfection (that's a joke , and that leverages the capabilities of excellent operating platforms to provide performance, robustness and flexibility for its users. I believe that if these are one's goals then Windows makes no sense as a target platform, and it also means that the user requirements of the "i want to plugin and make music" crowd are not particularly critical. If those are your needs, then by all means use Reaper, Windows or whatever else floats your boat. I believe that original question in this thread was whether Linux could be a feasible platform for "professional DAW work". Since nobody has defined what that means precisely, I'll cite a point made by Ben Loftis at Harrison: if you're doing contemporary pop production, with its emphasis on FX processing, looping, sample-driven songs etc, then Ardour is probably not the best tool for you. If you're tracking musicians playing instruments with their bodies, and the finished product relies less on production gloss and more on audio quality and performance, then its hard to see what features Ardour-on-Linux is lacking that will make your life easier. |
#115
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On Wed, 23 Dec 2009 20:09:30 -0800 (PST), dawhead wrote:
On Dec 23, 10:53*pm, Moshe Goldfarb wrote: And that is the problem that Linux faces. A person can load up Reaper for little or no cost and it will support all his VST plugins and he is ready to go. How is this a problem for Linux? Why do you assume that in order to be successful, "Linux" (which I reiterate is neither a product nor a company) that it has to dominate everything else? Is it a problem for Logic that most studios still use ProTools? Is it a problem for Pyramix that hardly anyone uses their system? What I am saying, is like it or not, there are some certain "defacto standards" that the vast majority,and the number is not small, are using. Linux does not support these. Reaper does. It's not a problem for platforms unless they reach a point where they realize people are not using their platform because of lack of support for some of these products. I think UAD, Sonnex, Sony Waves and many others qualify. Walk into any professional studio and you will see one or all those used. My goals with Ardour are mostly centered on providing a high quality DAW that puts audio quality and robustness just slightly ahead of workflow perfection (that's a joke , and that leverages the capabilities of excellent operating platforms to provide performance, robustness and flexibility for its users. I believe that if these are one's goals then Windows makes no sense as a target platform, and it also means that the user requirements of the "i want to plugin and make music" crowd are not particularly critical. If those are your needs, then by all means use Reaper, Windows or whatever else floats your boat. I believe that original question in this thread was whether Linux could be a feasible platform for "professional DAW work". Since nobody has defined what that means precisely, I'll cite a point made by Ben Loftis at Harrison: if you're doing contemporary pop production, with its emphasis on FX processing, looping, sample-driven songs etc, then Ardour is probably not the best tool for you. If you're tracking musicians playing instruments with their bodies, and the finished product relies less on production gloss and more on audio quality and performance, then its hard to see what features Ardour-on-Linux is lacking that will make your life easier. Well you have hit a particular nerve with me wrt to loopers, cut and pasters and other "semi talents" as I call them. It does seem like the current crop of DAW software manufacturers are catering to them. Sonar in particular. I do agree with you about tracking real musicians, but that is what ProTools and Reaper accel at. In fact just about any "try till you buy" software included even with low end soundcards can do this. I'm sitting here trying to get into the Xmas spirit listening to Manilow's "In The Swing Of Christmas" and sorry but the whole thing sounds highly sequenced to me. Anyone know for sure? I don't have the album notes as this is one of those "cough cough" mp3 rips Too stiff on most tunes.... Maybe a combination. I like Michael Buble's album although I swear he is auto tuning on a couple of tracks. Excellent piano playing and arrangements though. OT...just thought I would share.... And BTW I like you dawhead...You do a lot of good.... |
#116
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
But Linux is not a company. And this is part of the problem with trying to build a commercial product around it. There's absolute no evidence of this. There are lots of commercial products built around Linux. There are indeed commercial products built AROUND Linux. Harrison has some. But when you buy a Harrison product recorder, you don't get a package of software, you get a turnkey system. And when you need support, you don't go to a newsgroup, you go to Harrison, and you get your support, in the language you can understand (or else they don't make the next sale to you). The actual "problem", such as it is, is people in threads like this that talk about "Linux" as though *IT* was a product. No, I don't think they do. It's an operating system, just like Windows. Now, nobody ever bragged about support from Microsoft, but for the most part, end users don't need much support, because the system pretty much configures itself when it's installed - and in fact most people who use Windows are using a pre-installed version running on off-the-shelf hardware. I know that you can buy Linux systems that way, too. I wonder how well they work, and adapt to, say, audio applications and interface hardware. unfortunately, it already is the operating system of choice for quite a few people who want *precisely* that. as many as run windows? does it have to be that many to reveal how pointless your comment is? Well, yes. "Quite a few" doesn't include anyone I know when it comes to audio applications, and that's what we're discussing here. So how is it that you can buy a retail copy of Windows, load it up on just about any computer that meets a certain minimum specification, and have it all work? The CPU works, the hard drives work, the optical drives work, the USB ports work, the networking works. Plug in a Firewire card and Windows knows what it is and how to talk to it. Connect a Firewire audio interface and, when you install the driver that's provided by the manufacturer (and you have the major version of Windows for which that driver is written) it'll work. with 5msec latency? Is that all you have to say? Some hardware will give you 5 ms latency (mic in to monitor out) on Windows without too much fooling around. Besides, that's not the only criteria for a workstation. In fact (and I'm writing an article about it right now) 1 ms latency is too much. DAWs that have to run audio through software can't possibly work right. People put up with it because they choose not to fix the problem with hardware. But my point is that you can take the disks out of the box, plug them in, punch a few keys, and start recording. It's intuitive. I have not seen a Linux based audio application that's intuitive to get started with. And if you can't get started quickly, you lose your audience. I can find you forum posts for windows software that will disprove almost every claim you've made in this paragraph, except for the ones that are true for Linux too Why bother? You can find forum posts proving that bees can't fly, too. Sure, some people can never get anything to work, and some things, by design, never will work. But your chances of getting off the ground without specialized knowledge (bear in mind that few people "know" Windows) are better with Windows than Linux. it runs more hardware than windows (because it continues to support legacy equipment), and it installs easier than windows, as agreed by almost recent side-by-side install reviews in major computing magazines. I can argue about it installing easier than Windows. You can't start a Windows installation and ignore it for a couple of hours like you can with Linux, but you also get feedback along the way as to what it's doing. The first time I tried to install Linux, it didn't, and I didn't have any idea why not. I tried it again and then it did. I think maybe the problem was that I didn't have it connected to the Internet the first time around and it wanted something and couldn't get it. But I didn't know I had to do that. As far as supporting legacy equipment, that's not such a big deal these days. In fact, most of the problems that people are having with Windows and audio gear today is that they buy a new computer, it has Windows 7 installed, and they don't have Win7 drivers for the audio interface that they bought last year. But Linux may not support that hardware at all. The brutal fact of life is that today, recording has become an integral part of the work of a hobbyist and even professional musician. i'm not particularly interested in specifically catering to this crowd (though if it happens, all the better). Well, that's the largest potential customer base. But I guess as long as you're not dealing a commercial product it doesn't matter if you have 20 or 200,000 customers. You make the same money either way. I consider audio engineering a skill set that requires learning and dedication, and part of that process involves understanding the toolset in just the same way that you used to understand a studer. If I could understand Linux the same way I understand a tape deck, I'd be happy. I can relate a schematic to wires, circuit board traces, and voltmeter readings, and make adjustments and fix things that break. I can't approach a software troubleshooting problem the same way. I don't relate source code to anything physical. You don't have to know how to design a tape deck in order to adjust or repair it. You need some understanding of software design in order to troubleshoot software. And you need an instruction manual - which even a lot of Windows applications doesn't have a good one of. the fact that a lot more people are doing this is in no way different than when desktop publishing first emerged - the result was a gazillion people producing a gazillion and a half dreadfully typeset documents. making the tools easy to use doesn't lead to good results. I say the same thing, and people always argue that SOME manage to do it pretty well, and those may not have ever had the opportunity to do it if they didn't have inexpensive tools. I don't think this is a legitimate point to argue. I wish it were. I've often said that if it cost $100,000 to equip a facility for recording, we'd have less bad music, but that's really not for me to say. |
#117
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
Moshe Goldfarb wrote:
On Wed, 23 Dec 2009 22:38:45 -0500, Mike Rivers wrote: It's actually happening already like dawhead mentioned. Distributions like 64studio really do all the nuts and bolts for you. Is 64Studio suitable for "professional" audio production? Or is it just a cool package to surround the sound card built into your computer? That's a real question. I haven't looked at it. I never even heard of it before about five minutes agao. And that is the problem that Linux faces. A person can load up Reaper for little or no cost and it will support all his VST plugins and he is ready to go. So 64Studio won't? And I suppose it won't support my Mackie 1200F either? |
#118
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead wrote:
I don't see any reason why someone couldn't come up with a whole package dedicated to audio recording that was based on Linux. http://www.64studio.com/ Another fine example. I went to that web page and I didn't see anything that actually told me what it is and what it will do. It mentions a lot of names that I've heard before (Ubuntu, Debian, and the like) but where's the audio? |
#119
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
dawhead writes:
On Dec 23, 6:18Â*pm, Little****t wrote: They most certainly do not want to be playing with operating systems like Linux forces the user to do. Then for the time being, at least, they probably shouldn't bother. What I *do know* is that I plug in my audio interface, under Windows 7 x64 BTW, my control surface, select ASIO (my choice) and it all works. I never have to touch a single thing other than latency. What I do know is that I plug in my audio interface (an RME hdsp9652) under Linux (Fedora 11 x86_64), plugin in my control surfaces, start JACK and it all works. I never have to touch anything except hdspmixer to reset the initial matrix mixer settings for the hdsp, and JACK to adjust latency. I fire up Nuendo 4, Reaper, Sonar 8.5 and it all works. I fire up Ardour and it all works. All my plugins, well most of my plugins work. All of my plugins, well most of my plugins work. Linux doesn't even support this stuff. Wrong way around. The developers of those plugins don't support Linux. Can you run Ardour on Windows? Same deal. Linux' job isn't to support pre-existing software, just like OS X's job isn't to support Windows software. Clear thinking about where responsibilities lie here would help. And you are deluding yourself if you think for one instant that a professional, like myself or Mike or Scott or Hank or Fletcher will for even an instant claim that the Linux solution is "cleaner". Fortunately, I am not deluding myself for one moment into thinking that any professionals like yourself, Mike or the other people you mention (the last 3 I do not know) will claim that the Linux solution is "cleaner". I'm not even sure that I am claiming that. However, I can point you in the direction of a roughly equal number of people, all audio engineering professionals, who find the Linux solution compelling for a variety of reasons. Whose vote wins? Who even cares? Certainly not me. Maybe it's cleaner from a programmer perspective, a developer perspective and maybe in theory ie: crunchy code and all that gibberish. However I can tell you that if Linux were the only software choice for a DAW we would all still be using tape and a razor blade. On this point, you're sadly delusional. There are many valid criticisms to be made of Linux as a platform for audio engineering professionals. Many. This is not even close to one of them. Try that under Linux. Oh yea, if you have Â*SoundBlaster, it will probably work fine. Or any RME interface except the Fireface's (coming soon), or any ice1712/1724 device, or any of the following firewire devices: ECHO AudioFire 2 ECHO AudioFire 4 ECHO AudioFire 8 ECHO AudioFire 12 Edirol FA-101 Edirol FA-66 ESI Quatafire 610 Focusrite Saffire Focusrite Saffire LE Focusrite Saffire Pro 26 I/O Focusrite Saffire Pro 10 I/O Mackie Onyx Mixer FireWire Option Terratec Producer Phase 88 Rack FireWire TerraTec Producer Phase24 TerraTec Producer Phase X24 TerraTec Producer MIC 2/MIC 8 FireWire Of course, that is subject to my previous comments about choosing the right linux distribution. And if that's confusing to you, that is because you don't understand that "Linux" isn't a unitary thing. If you can't deal with this idea, then give up. Mike's venture into this is a prime example. I followed Mike's venture into this and interacted with him a lot on r.a.p as it happened. Mike went into his venture with a particular set of expectations and prior experience. His goals and requirements were not met, that was clear. However, his prior experience and his model of how things should be made it very much harder for it to work. I've had that discussion once (I think I wrote at least 20 posts on this at the time) You Linux developers are all doing your own thing and apparently nobody is talking to the other people. You have no idea who I talk to, how or where. Why do you insist on speculating, no *asserting* stuff that you simply don't know? Example: Pulseaudio. PulseAudio is a consumer/desktop audio architecture targetting consumer media workflows and applications. It is also very focused on low power/mobile devices, a market where Linux is growing increasingly dominant, and the requirements for the audio infrastructure are entirely different. If you are attempting to do pro-audio or music creation using PulseAudio, then you've already done way down the wrong track. Is it a problem that this is not more obvious from the outset? Sure. Is it partly related to what I said earlier about Linux not being a unitary thing in the way that Windows or OS X is? Absolutely. Yea, I know Ardour etc doesn't deal with it because it's ALSA based. Ardour doesn't use anything except JACK. Which is why it works on OS X (and has, historically been built and run on Windows). So why is that? If ALSA works fine, and it does IMHO, why all the others? ALSA is primarily a set of device drivers and a thin abstraction over them. That doesn't make it (necessarily) the appropriate API for developers to use when writing applications. Oh yea, "choice"... No, it's not choice. It's a cluster****. Sure. That's why people who are using ALSA drivers they created for data collection (i.e. non-pro-audio, just massively high sample rates) are finding it a cluster****. Once again, you fail to grasp that linux isn't a unitary thing. Its a way for people to put together systems that can do certain tasks much better than most other operating systems. Does that mean that when you install J. Random Linux Distro on your J. Random Hardware that it will make a superb DAW? It most certainly does not. For a report on my perspective, this is useful:http://lwn.net/Articles/355542/ Fair enough... My personal feeling is ALSA should be tightly integrated with the kernel It is. (I'm not a programmer so please excuse the simple language) and the others should disappear. If you talk to any audio developer who has written applications for JACK, CoreAudio, ASIO, WDM and ALSA (to name a few), I am fairly confident that they will tell you that JACK provides the simplest, easiest to use and most powerful abstraction for audio of any of them. None of these other systems provides the same kinds of capabilities, or the same ease of development. They don't make it possible to make arbitrary applications do arbitrary things. Ergo, unless ALSA were to incorporate all these things, its not likely that JACK is going to go away, for the same reason that JACK doesn't "go away" on OS X - CoreAudio just doesn't provide the same stuff. Same with any Windows audio API. Put it this way, if I had you in my studio for a couple of hours, I am certain you could show me things that Jack/ALSA etc can do, that I could not even in my wildest imagination conceive. But I don't have you and documentation is poor so I am left to mostly fend on my own, like most people. Sure, the documentation is not good. But its improving, and unlike the code, is an area where actual audio engineers could contribute a lot. Do they? A few. Ivory. Garritan Addictive Drums UAD Sonnex Nuendo. ProTools. Do they run under Linux? Sample libraries that use Gigasampler and a few other formats: yes. Pianoteq (a physically modelled piano that to my ears is more powerful than any sample library): yes (and windows and OS X too) Ardour: yes many VST plugins: actually yes. i helped make this possible, but i can't say i recommend it. so, your chosen tools don't support Linux (see notes above). how is this linux' fault? does it make an unsuitable platform for you? probably. but so what? I leave you with one thought, Linux is free and a lot of musicians and engineers know about it. Why are so very few using it at a professional level? Because most of them are as confused and ignorant about it as you. Their/your fault? Not entirely, no. Dear Dawhead: Thanks for your detailed replies. I know you're putting effort into this. We may have corresponded in the past too, I'm not sure. Anwway, I use both Linux and Windows, so maybe I can offer some perspective. What I've noticed is Windows programs come with all the drivers and configuration. I can do a clean Windows install (XP for now), get the updates, then install my drivers for Presonus Firepod. Then I can install Cubase or whatever DAW program I like. After that, I'm done. I don't know to know what ASIO, MME, or whatever is. I just go ahead and start working. Contrast that to Linux. I have to decide which distribution. OK, let's stick with Debian because it has been around for a long time and I've used it elsewhere. Then I try to install Ardour. When I install it says something "do I want realtime kernel priority on Jack". WTF is Jack? I don't know. I hear it is something great "under the hood" but I don't care really. Then I try to get my Presonus gear working. I read something about needing firewire drivers. So I load those. Then I'm not sure it works right. By this time I go back to my Windows machine. OK, I realize that Windows is a dog's breakfast under the hood. But thanks to commerical vendors (Steinberg, Presonus), they've made Cubase work for me. If it didn't work, noone would buy their software or hardware! I think Linux is great for many things. I use it for servers. I also use it for my research machine. In both of these cases Linux provides the reliability and security I require. But I still rely on commercial software, in my case, Matlab. I think if someone comes out with Cubase (or Adobe Audition or Wavelab or ...) for Linux it would catch on real fast. But, as others have said, the market is not there. The DAW market is small enough already. Intersect DAW and Linux and you've got an even smaller audience. Richard |
#120
Posted to rec.audio.pro
|
|||
|
|||
Is Linux A Feasible Platofrm For Professional DAW work ?
On 24 Dec 2009 00:45:29 -0500, Richard Mann wrote:
Dear Dawhead: Thanks for your detailed replies. I know you're putting effort into this. We may have corresponded in the past too, I'm not sure. Anwway, I use both Linux and Windows, so maybe I can offer some perspective. What I've noticed is Windows programs come with all the drivers and configuration. I can do a clean Windows install (XP for now), get the updates, then install my drivers for Presonus Firepod. Then I can install Cubase or whatever DAW program I like. After that, I'm done. I don't know to know what ASIO, MME, or whatever is. I just go ahead and start working. Contrast that to Linux. I have to decide which distribution. OK, let's stick with Debian because it has been around for a long time and I've used it elsewhere. Then I try to install Ardour. When I install it says something "do I want realtime kernel priority on Jack". WTF is Jack? I don't know. I hear it is something great "under the hood" but I don't care really. Then I try to get my Presonus gear working. I read something about needing firewire drivers. So I load those. Then I'm not sure it works right. By this time I go back to my Windows machine. OK, I realize that Windows is a dog's breakfast under the hood. But thanks to commerical vendors (Steinberg, Presonus), they've made Cubase work for me. If it didn't work, noone would buy their software or hardware! I think Linux is great for many things. I use it for servers. I also use it for my research machine. In both of these cases Linux provides the reliability and security I require. But I still rely on commercial software, in my case, Matlab. I think if someone comes out with Cubase (or Adobe Audition or Wavelab or ...) for Linux it would catch on real fast. But, as others have said, the market is not there. The DAW market is small enough already. Intersect DAW and Linux and you've got an even smaller audience. Richard Well stated Richard!! Yes dawhead is a good person and despite our differences I respect his opinion and I truly appreciate all his hard work in the open source world. There is no denying he is making an impact. I really have nothing to add to what you have already stated Richard. I agree with you. |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Linux is More than Ready For Professional Audio. Here is Proof! | Pro Audio | |||
Linux audio applications ARE PROFESSIONAL! | Pro Audio | |||
Linux and PROFESSIONAL AUDIO?? "I have no professional training" | Pro Audio | |||
Linux Used In a Professional Setting. Here is an Example!!!!!!!!!!! | Pro Audio |