“Give it All, Zero For Rules!” shouts Mattin. For some, free music software isn’t just a way of saving money. It’s a way of life.

At the center of the debate is a semi-rivalry between two similar programs: Pure Data (aka “Pd”), and Max/MSP. Both Pd and Max/MSP use a similar interface, allowing you to build custom music, video, and multimedia applications by visually interconnecting objects. It’s not quite programming (though in each you can code your own objects if you like), but it offers many of the benefits of programming visually, and in real-time. Pd is free/open source; Max/MSP costs a few hundred dollars.

So why are there two programs? Here’s a VERY long story made short: [read more]

Max was originally the creation of Miller Puckette while at the IRCAM research center. Max grew into various generations and flavors from the early system, but musicians have gravitated toward one of two variants: Max/MSP (and Jitter) is maintained as commercial software by Cycling `74, an offshoot of development work that began at Opcode in the 90s with David Zicarelli, while Pd is an open-source, free project that grew out of continued development work by Miller Puckette. The two can coexist, and often even share externals (add-on objects).

For many, both are simply useful tools. For others, though, the fact that Pd — and other programs — are open source has heavy political implications. Mattin, for one, interprets “computer music” literally, “playing” the hard drive and bowing the computer case. If you want a heavy dose of radical theory, see the extensive manifesto stay free (which is openly anti-Max/MSP), or the Give it All, Zero for Rules! article — the latter being more useful, with links to people actually using this stuff.

Commercial vs. open source argument aside, I have to observe that some (not all) Pd die-hards are making arguments that are misleading. The stay free article, for instance, implies that Pd is more extensible than Max/MSP — theoretically true because of its open source nature, yes — but then claims the reason is the use of externals, which Max/MSP has, too. Max is plenty extensible, so while you can argue against its price, I think Pd advocates are going to need a different argument to suggest Pd is fundamentally better for any other reason. (Note: one significant advantage of Pd not mentioned is that it runs on Linux; Max/MSP does not.)

I’m not as interested in the free vs. commercial discussion, so to me, both of these are fantastic tools. Pd’s free nature is important: not everyone can afford Max/MSP, particularly students (important to us teachers). Each has some unique capabilities, so if you’re serious about this kind of work, you really should have both on your hard drive. And if you can invest in Cycling ’74’s software, it really is worth the money in terms of the payoff in maturity and usability. For me, the answer has been to use both.

I am sympathetic to the ‘stay free’ argument, though I can’t agree with it. I’m not sure why we’re so hung up on the idea of free software — I’d like to start with the idea of free health care first, and move from there. (Yes, people in the rest of the world behind the US, I know YOU have national health care, but I’m here in America, so I can’t get as emotional over free software!) Commercial software enables those privelaged by extra disposable income to support the endeavors of programmers, thinkers, and an industry for music, and that can coexist (rather than compete) with open source software.

But, of course, that’s the point: this is about choice. Choose a tool — or choose both, or neither — and make music. Even just using free software, you have more musical choices than you could ever use. Zero for rules, musically speaking.

  • Guest

    One thing PD and Max/MSP share though is their crappy sound quality…

  • admin

    Pd and Max/MSP can sound like whatever you want, dependent on your available processor speed and how you configure things like FFT sample window. Real-time native operation requires sacrifices on any system — meaning its up to the user to figure out ways of compensating musically (always true, regardless of sound quality).

    Of course, I think we are witnessing only the beginning of the tech here — it's clear that, in video especially, we're waiting on faster processors, etc.

    But to simply say "they sound crappy" isn't even remotely fair. I've heard people do things that sound fantastic. Depends on how you use it.


  • Guest

    You should include SuperCollider (www.audiosynth.com) in this discussion.

  • Guest

    'Depends on how you use it.'
    -More a matter how sensitive your ears are… 

  • Guest

    ' SuperCollider (www.audiosynth.com) '
    -….is designed for sensitive ears & compositional minds…

  • admin

    All sorts of applications and hardware tools claims to be "designed for compositional minds" as if other things are not. A piano is designed for sensitive ears and compositional minds, too — and it has all sorts of flaws. (And I'm a pianist.) Is it better than a trombone? I don't think so.

    I haven't seen any evidence that SuperCollider (or anything else) is somehow inherently superior to Max or Pd. (In fact, SC's design owes something to the original Max and Max Mathews' own Csound.)

    Sensitive ears / compositional mind / etc. — again, depends on how you use your mind and ears, and if (hopefully) you're able to develop the chops you need for music that's satisfying to you. If you're more satisfied working in another tool, then great, be adventurous, find the tool that makes you happy!


  • Guest

    'SC's design owes something to the original Max'
    -Presumingly referring to the 'pyrite-object'
    in Max…
    The SC-language and audio-engine as far as design goes,
    has nothing to do with Max/MSP of PD.

    ' Max Mathews' own Csound''
    -Csound author is Richard Boulanger…

  • admin

    Not directly related — SC is based on similar conceptual research.

    As for CSound, got my wires crossed and left out a step — meant CSound *by way of* Music I – IV; CSound is a derivative of Max Mathews' original Music I (Mathews and others then contributed to subsequent versions . . .)

    Richard Boulanger did NOT write CSound. If you had to credit it to one person, you'd probably pick Berry Vercoe who contributed more than anyone, but again, CSound wouldn't exist without Mathews and others at Bell, and the whole family (CSound, Max, Music I – xx) came out of a lot of sharing of work.

    All these tools are related, some strongly (like CSound to Music IV), some via more weak conceptual connections. This is easily the subject of a BOOK, not a comment(!!), but I may do a quick overview of the history in a future article — which would certainly require research on my part, as it all gets rather convoluted!

    Anyway, bottom line: to say any one of these tools is inherently superior is silly. Fundamentally, they all have an identical purpose and function. The fact that there are so many variations just demonstrates that people are driven by highly individualized tastes — and that's fine.

  • admin

    . . . I see what may have tripped this off:

    I'm using the Pd vs. Max debate as an indication of the open source vs. commercial camps. It is by NO means inclusive of all the possible tools on each side (Ardour vs. Pro Tools, for instance).

    If we wanted to just look at the major free tools for interactive software design (jMax, Pd, SuperCollider, CSound, etc.), that would be do-able — though I didn't do that here. (Actually, those four would probably be fine to get people started, unless anyone has any additional favorites? Anyone for JSyn? RTcmix?)

    Oh, and to any fans of Dr. Boulanger: of course, a huge influence on Csound, and the community at large . . . though even he defers to Vercoe for the pioneering work on Csound.

    Now, if you'll excuse me, I have to get back to . . . Logic Pro. 😉

  • Guest

    'to say any one of these tools is inherently superior is silly'
    -Don't think so. Each of them have their own strengths in
    their own area: SC is excellent in scripting audio
    structures and the Max'es (PD, jMax, Max) are
    wonderful in controlling these.
    Csound is perfect for non-realtime rendering.

  • admin

    . . . each has strengths. I'm responding to the notion that SuperCollider is 'more musical' — don't see it — or that (from the link) Pd is 'more extensible' than Max/MSP because of externals — simply not true.

    Max/PD/etc. and real-time pathing environments do tend to be limited in setting up linear musical structures in time, that's true, and there SuperCollider and Csound have an edge (working in different ways). Csound is very mature, but I'm not personally convinced that it's kept pace with the times. (though, of course, it still has a very loyal following)

    Anyway, all of these fall into the 'people use them' and 'people do interesting things with them' category, so they're all noteworthy, and do have things to differentiate them from one another. Though I still can't find anyone using jMax — please stand up if you are.


  • Guest

    The Freedom of Free Software is the ability to read and modify the source code, price has nothing to do with it (in principle anyway).
    Being able to read and modify the code for creative tools is incredibly useful, your creativity doesn't have to stop at the UI or at a limited scripting interface, and your ability to work with both can benefit from understanding how they work.
    Here endeth the polemic. :grin

  • Guest

    'Though I still can't find anyone using jMax'
    -It was recently killed by its creators….

  • Guest

    ‘to say any one of these tools is inherently superior is silly’
    -Don’t think so. Each of them have their own strengths in
    their own area: SC is excellent in scripting audio
    structures and the Max’es (PD, jMax, Max) are
    wonderful in controlling these.
    Csound is perfect for non-realtime rendering.

  • iTom

    the correct link to the "stay free" article is