post-Muzys: Jo's three questions

[Caleb]

Caleb, I share your dreams! But doesn't this describe the process Muzys already facilitates well?

It does to a large extent - which is why I thought that Muzys was the best sequencing tool on the planet.

It still didn't quite get there for me, but it was certainly better than what I saw elsewhere.

A few important enhancements to the KeyMap Player and the sequencer itself and you'd be damn close.

Caleb

[lepwisk]

I'm sure there are radical approaches that are still open. Maybe I should take more of a look at things like Csound before I spout off but I'm starting to get a vision of small, script-based modules with simple UIs and plugable framework. Something you could basically build to look like Muzys -- or whatever!

Would Supercollider (http://www.audiosynth.com/) be something along the lines of what you're thinking of?

Obviously, such an approach can lead to the creation of tools that have some very powerful creative possibilities. The only issue I have with this is that when you choose to go to more of a text-based interface you end up creating an app that is far more specialized, appealing to the user that is comfortable working in a text-based environment. An app like this also ends up having a far steeper learning curve than one with a more accesible GUI. As such, your potential market is pretty limited at first, until the developer (or the user community) create enough easy-to-use GUI modules to make the package attractive to your average computer user. The overall development path of such a platform (perhaps a better descriptor than application) would likely be much slower.

[shamann]

That I can think of, there are 5 basic host types/interfaces:

Linear multi-track:
http://www.museresearch.com/kvr/i/b/logic.jpg
http://www.museresearch.com/kvr/i/b/tracktion.jpg

Virtual studio:
http://www.museresearch.com/kvr/i/b/flstudio.jpg
http://www.museresearch.com/kvr/i/b/orion.gif

Modular workstation:
http://www.museresearch.com/kvr/i/b/audiomulch.gif

Tracker:
http://www.museresearch.com/kvr/i/b/berotracker.gif

Chainer/Rack:
http://www.museresearch.com/kvr/i/b/forte.jpg

Of course there are functional differences within each category, and others that mix two or more, to varying degrees of success.

Looking at the screenshots of the above examples, I can't help but feel that there is a lot of untapped potential for coming up with something outside the 5 categories, that only a limited few ways of thinking have been explored. We have only a handful of initial approaches to making music with computers, and they all seem conceptually to start from what you can do with studio hardware or pen and paper. Even the very software-only categories like the modular environments tend to extend from hardware (you have discreet modules, you connect them with wires, etc). Some of those ways have converged to such a point that they are virtually indistinguishable, but for variations on managing clutter. The 5 main forms have served us well, so why break the mould too much?

But, there are a lot of entries in most of those categories. Meaning, at this stage of the game, coming up with anything new that is only a variation on the 5 categories starts with fierce competition right from the start. And most people can find something out there to fill their specific needs that the 5 categories already cater to (except for those few who will always find something to bitch about).

A lot of the content in this thread, and other threads of its ilk, tends to be people's suggestions for software that thinks like them, except within the accepted 5 categories, and mostly lists to remove/add some minor personal dislikes/preferences.

Personally I'd rather see software that challenges my way of thinking, and offers something unexpected. I just can't see much room in the market for something that only varies what's already out there. There are a lot of good variations already.

The be-all-end-all, big sequencer/host doesn't make much sense, especially for a small business. Go in knowing that people already have a lot of software they frequently use, and make good software that is designed to work well with other software. I can't say what routes should be explored, that all comes down to personal interest, and what motivates the coder isn't necessarily what motivates me.

[pljones]

What bug do you mean?

I have a semi-predicatable crash when I mute a bunch of samples I've got mapping in sfz -- but only when hosted in Muzys. sfz doesn't crash when running standalone. So there apppears to be a certain degree of interaction involved between the plugin and the host. I'm guessing sfz may be sending maybe a short audio buffer or something like that.

[braj]

Non-linear multitrack: Ableton Live. Not sure if Project 5 fits in this category as well.

[pljones]

Would Supercollider (http://www.audiosynth.com/) be something along the lines of what you're thinking of?

That's the kind of thing, yes. And with a much heavier MIDI bias, there's http://www.nosuch.com/keykit, of course.

I do agree that getting a decent set of basic modules is essential -- I wouldn't expect to hit beta before you could ship something substantially useable (e.g. ship with a "muzys-like" preset). But at least part of the point is to keep it robust. Keeping each component discrete and well tested helps. The two big components - UI support and script engine - I could see being in constant development... Such an approach also makes platform-dependent code really, really easy to isolate. Ideally, the UI would be pluggable, too (so you could jump from Windows core UI to GTK+ to Qt to... etc... without changing the "application", just the UI wrapper).

Of course, I'm very biased - I'm a computer programmer, so such an approach is comfortable. But I'm not suggesting anyone would be required to do anything that didn't have UI support to be productive. It's just the power would be there. The appeal to third-party developers would also be there: "Here's my custom soundset, along with the custom UI to control it for MuTools users."

[shamann]

Non-linear multitrack: Ableton Live. Not sure if Project 5 fits in this category as well.

Looks pretty linear multitrack to me:



Live, FLStudio, Orion, P5, are all amalgamations of the 5 categories I've listed. I know Live users like to insist that its vastly different to everything else out there (for evidence, just search any host thread and look for the "...but Live..." posts). It may be great, perhaps the greatest audio app ever made, but it isn't so blindingly different from the rest of the field.

[SQ4]

Consider this :

- there's no such host that's ever gonna be "the next big thing".
- think also about Podium/Aero Studio : damn hard and long time programming work
- are you making a living out of this new host?
- is programming still fun?

anyway,

success!

[snooty30]

Jo, if it's a host you're going to develop then I would suggest to lose the routing from CMuzys. If you're talking workflow then this was possibly the one thing for me that destroyed it. Yes it's flexible but also time consuming. I also use Cubase (SE) & loading a VSTi in that automatically loads the relevant mixer channels. Lack of audio streaming was never a problem for me in CMuzys as I never used it for multi tracking, only for triggering VSTi & loops. I can see how some people would want that though.

[kovacs]

Non-linear multitrack: Ableton Live. Not sure if Project 5 fits in this category as well.

Looks pretty linear multitrack to me:

You're only seeing half the picture. Live's clip view lets you play the sequencer as an instrument.

[shamann]

But you're missing half of the explanation where I noted that hybrids abound, including Live. Still see nothing that fits outside multi-track and the virtual studio way of doing things. Clip triggering can be done in eXT, FL Studio, Orion, etc. Live's implementation may be an elegant solution, but conceptually it is still an extension of common ways of doing things.

It's not a criticism, those methods exist because we know they excel at certain tasks.

[blurk]

I am not aware of good alternatives to Energy XT for OSX.

Plogue Bidule? Not exactly the same, because Bidule is weak on sequencing, particularly compared to Energy XT. But I think they occupy similar ground. Basically, I decided not to buy an Energy XT license (at least for the moment) because I did buy a Bidule license, and I felt there was too much overlap for it to be worthwhile.

As an aside, I am another one who tried the CM edition of Muzys and never got into it. Not because I thought it was bad, more that I already was using FL Studio and Bidule, and didn't really want to put in a lot of effort learning a new host, which (being the CM edition) was somewhat stripped down and limited.

What I'd like to see made is another VST sequencer plugin, to make up for the lack in Bidule. That is, not a host (though I guess that maybe worthwhile to expand the potential market) but a VST plugin that outputs MIDI. The only other substantial one I can think of at the moment is Sonicbytes ERA. But I do realise this is probably only going to appeal to a narrow market.

[muzycian]

Jo, if it's a host you're going to develop then I would suggest to lose the routing from CMuzys. If you're talking workflow then this was possibly the one thing for me that destroyed it. Yes it's flexible but also time consuming. I also use Cubase (SE) & loading a VSTi in that automatically loads the relevant mixer channels.

Whether there comes a new host or not is still an open question. I'm reading the feedback in this forum, and thinking about things.

Lack of audio streaming was never a problem for me in CMuzys as I never used it for multi tracking, only for triggering VSTi & loops. I can see how some people would want that though.

IF there comes a new host, it will surely have audio streaming!

Cheers,

Jo

[Trapaki]

Jo please - a new host ! Muzys is still outstanding software, all what it needs is an audio streaming and OS X port... ok even only OS X port... I can't run OS 9 anymore and I miss Muzys so much... the best sequencer ever !

[sluggo]

Collaboration is the way to go. Look at what Fxpansion just did with Devine Machine (GURU).

I think there are a lot of developers that are saturating their market (but still doing cutting-edge development). Joining with them can accomplish new ideas and concepts, along with sharing the load of programming. Different developers bring different assets, the whole can be bigger than the sum of the parts.

s