Why Ardour don't work in windows?

Community Ideas for Ardour
#41

I definitely agree with you both.

I think windows support needs to fit in to the long term strategy of the development team. I think the core developers need to find out amongst them selves what should be Ardours stands on windows. Should windows be supported in 3.0 or 3.1 or 3.4 or should Ardour never support windows.

If a long term strategy is implemented then we can gradually start to work on windows support. Perhaps by attracting developers to a closed alpha or beta version so that all the installation and other windows specific issues can be solved before we get a flood of “Ardour wouldn’t run at all until I’d launched this dastardly thing called ‘X’” questions.

A solution might be to start very soft by creating a forum under Installation & Configuration called “windows development & beta testing” (or something better) to start attracting windows developers and give them a chance to communicate.

Sorry if my replies seem a little inconsistent. I’m just trying to get a feel for what peoples opinions are.

#42

That certainly ties in with my experience Paul. I found nothing to indicate that Ardour might be flaky or less reliable under Windows. In fact on my PC (which houses both Windows and Linux) Ardour ran exactly the same under them both. The issues were more peripheral and mostly to do with the fact that Ardour didn’t “behave like” a Windows program. For example, Ardour wouldn’t run at all until I’d launched this dastardly thing called ‘X’ (of course, I understood this but I think it would have puzzled most Windows users). Furthermore, I could only launch Ardour reliably from a command line. And it only “looked good” on my left-hand monitor. Dragging it to my right-hand monitor made the fonts look very “fuzzy”. None of these problems were in any way due to Ardour but I’m quite sure they’d generate a barrage of questions from confused Windows users.

Given enough developer time, I’ve no doubt these issues could be overcome (e.g. changing the menu font might have solved the font problem). But as you say, this kind of thing isn’t within the remit of Ardour’s development team.

So, to come full circle, my advice to Cinmay is that if he still wants to press forward, he needs to accept that “we” shouldn’t be assumed to include the existing developers.

#43

John - the issues with a Windows port are far from limited to bug reports about Ardour. Every project I know that creates software for Windows, whether proprietary or open source, inevitably faces a large number of support issues that ultimately resolve to something entirely unrelated to the software in question. Given the current nature of the Ardour community, we have the expertise and even (in general) the willingness to try to tackle such questions when they arise on Linux, in part because we don’t believe they will arise to the same extent and/or in quite the same way. However, I seriously doubt that our community would be capable of handling the myriad of Windows-naive user questions that we would face, and I equally doubt that our community would expand fast enough with Windows-centered knowledgeable and helpful members. I read quite a few audio tech related forums, and they almost as full as the same kind of “basic machine setup” issues as the ardour.org forums. I do not want myself or any potential Ardour developers (of any level, from design through extensive coding) to be spending time on issues like this. Its bad enough that distributions’ mistakes and unwillingness to provide for our use cases means that we have to do this on Linux. Adding Windows to the mix would just be a disaster.

#44

What’s becoming clear here is that a lot of people seem to be reading Bryce’s article but failing to understand it. And I don’t really blame them. IMHO it isn’t a well-written article. So here, in a nutshell, is the point that I think Bryce is trying (unsuccessfully) to make:-

The main thrust of Bryce’s argument is that porting a small project to Windows has the effect of exposing it to a much wider audience. More eyes looking at the software results in a big increase in bug reports. But it doesn’t necessarily bring in more developers to fix the bugs. Therefore it increases the burden of support for the existing developers. I suspect this is what Paul means when he refers to the social issues of porting Ardour to Windows. Unfortunately, Bryce’s article is cluttered with too many (more spurious) arguments and one has to read the article very carefully (and several times) before that message crystalizes. To the casual reader, he simply seems to be arguing that it’s better to sacrifice popularity and remain in blissful ignorance of the bugs. Or perhaps (to put it more charitably) it’s better to hope that the bugs won’t affect (or won’t get noticed by) a smaller audience. To my mind that sounds like a bad case of wishful thinking so I can fully understand why people aren’t persuaded by Bryce’s article.

However, as one of the few people who HAS successfully ported Ardour to Windows I think I can speak authoritatively about the likely problems - and lack of POSIX compatibility is NOT one of them. In point of fact, Microsoft offers an implementation of POSIX (for Windows) that’s reputedly every bit as good as anything you’ll find from Apple. Furthermore there’s a well-established open source project called Cygwin. Not only does it offer an exceptionally good POSIX environment for Windows but (with the sole exception of ALSA) everything needed to build Ardour is readily available both in source code and in binary form. I personally have compiled Ardour under Cygwin with almost no major changes needed. MIDI was my only significant area of concern as (I assume) it must also have been for the OS-X port.

But at the end of the day, Ardour is NOT a Windows program and therefore a Windows port wouldn’t offer the things that Windows users take for granted (such as simple installation). This is where my experience differs from Bryce’s. I agree with him that a Windows port would bring an increased burden of support but I saw no evidence that Ardour was any more buggy under Windows than under Linux or OS-X.

In my view the most likely issues would be installation issues. Ardour’s dev team would inevitably find itself swamped with questions about Cygwin (or MinGW, or whatever). I support the view that this isn’t (and shouldn’t be) in their remit. In fact, to give just one example, I never found a way that Ardour could be reliably launched from a desktop icon. Can you imagine the uproar that’d be created in the Windows community by that problem alone?

Paul is right to say (at this point in time) that a wholesale launch of “Ardour for Windows” would be counter productive. There aren’t enough developers to support it and there’s no obvious reason to believe that new devs would come forward. Having said that Cinmay, I personally agree with you that a small scale “trial” could only be instructive. Nonetheless, the key to success would depend on finding the developers FIRST and starting the project SECOND. I don’t think you’ll get very far if you want to start a project, then simply “hope” that developers will join it.

#45

No. I am not a Ardour developer. When I was talking about we I was talking about the community in general. I don’t like to front ideas without being ready to put my money where my mouth is and do the actual work too but I would much rather spend my time sending in patches on the new Ardour3 midi system witch is what I am really interested in.

That being said I do understand the value of community for the good of the software, support and even financial. So I would be willing to dig up some windows pc and putting a beta version on a newly created web page. But only if the rest of us (The community) and the main developers thought it was a good idea and maby someone would be willing to help me? :slight_smile: It’s been many years since I last did any sort of programing on windows and I dread having to learn a lot of stuff that I really don’t want to know.

#46

@tryken: i’d appreciate it if you would read the article that i linked to, and then show some signs of tackling the experience it reports and the argument it makes.

#47

Cinmay - please forgive me if you’re already an Ardour developer. You mentioned being a developer but I don’t recognise your name so I assume you must mean for some other projects.

It would help if you could clarify who you mean by “we”. Are you offering to take on this extra work or are you implying that the existing dev team should undertake it? Because I think that is really at the heart of Paul’s argument.

Towards the end of your last post you acknowledged that this would be a burdensome extra load. You also stated that it shouldn’t be piled onto the (already overworked) dev team. So do you have someone else in mind when referring to “we”?

#48

cinmay-

Read the link that Paul posted, I believe that sums up his views on 2 and 3 of your post rather well.

  Seablade
#49

I personally don’t think there is much comparison of the multitrack capabilities of Audition 1.5(Haven’t used the newer ones) and Ardour. That is what I was referring to. Audition, and before it Cool Edit was an amazing standalone destructive file editor(To be honest probably the best, and certainly my favorite), but the multitrack capabilities were a bit lacking to be honest when it comes to real time non-destructive mixing.

   Seablade
#50

Yes, apples and oranges.

Both are great sources of vitamin C, even if the ‘user interfaces’ in the manner of consuming them are different.

I can be just about as productive for the types of multitrack and touchup work I do in Audition as I can in Ardour; I can peel and eat an apple just about as quickly as I can peel and eat an orange, even though different tools and techniques are required. Both give me vitamin C… uh… good multitrack editing and touchup capabilities for song production from multiple, layered, sources. Even though the taste, texture, and varieties are different. I could comment on the details, but those aren’t really important, except that with Ardour I have to add a good soundfile editor in the mix, where with Audition it’s integrated.

Yes, apples and oranges. But both produce the same desired end result for my use cases.

#51

For the record, Audition 1.5 and Ardour really are apples and oranges;)

Seablade

#52
*mode rant=true* Windows users have Logic, Cubase, ProTools, Audition, and all the rest. Ardour at the moment is the biggest DAW fish in the Linux DAW pond. It would be a small fish in the Windows pond, and ANYTHING that distracts the Ardour developers from making Ardour the best Linux DAW out there is BAD for Ardour and BAD for Ardour's users, potential increased income for Paul from Windows users notwithstanding (and, Paul has been more than clear enough about his stance on this).

If I wanted to run Windows I’d use my Adobe Audition license on Windows. I choose to use Linux, and Ardour on Linux bests Audition (1.5 is what I own and used for quite a while) on Windows up one side and down the other.

And Bryce’s article is spot-on. There are a few opensoure Unix-centric projects with enough developers to make a Windows port worthwhile (PostgreSQL, for instance, works quite well with little issue on Windows, but getting the developers to agree to do it and getting dedicated developers to support it was a multiyear rigamarole; I was there, packaging PostgreSQL RPMs at the time), but Ardour is duplication of effort in the Windows space since there are plenty of Windows DAW’s out there. There is a paucity of really good DAWs in the Linux space; let’s not dilute the development effort any more than it already is! And if you’re looking for free-of-cost you shouldn’t really be looking at Ardour anyway, IMHO.

You want to use Windows? Fine, but please leave the best (IMHO) Linux DAW out of it. Or port and support it yourself.

mode rant=false musician/producer/Ardour subscriber=true

Sorry, but speaking as a producer/engineer/musician (and someone who puts in a monthly donation via subscription to Ardour development), I get tired of the idea that everything should work on Windows. Windows is not the be-all end-all, and it is OK if the developer of a program chooses to not port that program to Windows; there are certainly plenty of Windows developers who are unwilling to support Linux for many of the same reasons Paul won’t support Windows (support nightmare, few developers, low developer time, etc)!

Hrmph, perhaps when we can get MOTU, Cubase, ProTools, and Logic support on Linux, then I might change my tune…

But Paul likely won’t. And that’s his call anyway.

#53

If those who want a windows port are just looking for something free of cost there are options, apparently even an opensource option.

The closed source option is well known, reaper.

I only recently stumbled upon the open source option: http://www.experimentalscene.com/software/darkwave-studio/
I don’t know anything about the quality of this software and I strongly suspect that this project does have legal problems, mainly due to it’s use of Steinberg technologies and the resulting license issues.

Since ardour tries to stay in the legal area I suspect that it wouldn’t provide what windows users expect, which would be mainly VST and maybe even rewire or ASIO support (whereas jack is superior to the later two, afaik).
I suspect a lack of VST support alone would be reason enough for lots of windows users not to use ardour.

It seems that in contrast to VST, AU support is legally possible, which might have been another reason why ardour was ported to that platform.

This is just another argument that hasn’t been made yet, I added it for completeness’s sake, feel free to correct me if I’m wrong on any of this.

#54

Not Found

The requested URL /drupal/foss-win-paradox was not found on this server.

#55

I am a brand new subscriber today. From what I can tell I am the exact problem that is keeping development Ardour for Windows.

I didn’t realize that I would not be given access on my current OS. I didn’t read everything about Ardour. I just wanted it for what looks like it is excellent in the areas I wanted to address. I will go back to Audacity and make it and Avidemux work do the work I wanted Ardour to do for me. I will subscribe here while I wait for the windows version to come out.

Please help me.

#56

See my answer to the thread you created. You can read Bryce Harrington’s text at this URL: http://tu.linuxaudiosystems.com/BryceOnFossWin The original has long since vanished from the net, unfortunately.