automated live performance mixing

[MurphS]

Hi,

Pardon if this is a boneheaded question...I've been out of the music scene for 15 years or so, but am about to dive back in. I would like to find out about the state of automated mixing in a live context... software or hardware. Specifically, I'd like to 'sequence' audio parameters such as pan and volume, maybe eq parameters and such also. This might sound like a ridiculous thing to want to do in a live context, but I am a solo performer and will be playing a variety of instruments along with a sequenced drum track, and would like to pre-program things such as fade-outs since I won't have a sound-man handy. Is this trivial, or possible, or could someone please point me in the direction of information pertaining to equipment which might do this type of thing?

Thanks, Scott

[Antimon]

Possibly someone will chime in with solutions to your exact situation. If not, this is perfectly feasible in any of the various programming environments, in the style of Max/MSP (graphical) or ChucK (text programming language). You could also do it on a virtual modular platform like the Nord Modular (not sure about the alternatives here). The latter is hardware, in case you don't want to lug a laptop and audio interface around.

/Stefan

[MurphS]

Stefan,

Thanks for the advice. I hadn't even thought of doing it that way...Glad I asked. I am actually writing a JAVA program to be used with what I am wanting to do...I haven't even heard of the languages you mentioned, but that's about to change! At this point, if I could have the proper things automated I wouldn't care if it were software or hardware (and so far I haven't found any hardware that will do it -I'm still looking. Haven't heard of the Nord Modulator though) . I'll probably need a rack mounted CPU to run the JAVA program I mentioned, anyway.

Thanks, Scott

[Antimon]

I think if you're already looking into programming, the Nord Modular isn't the best solution. I mentioned ChucK since I've used myself a lot, there's also SuperCollider (we have subforums here for both of those, full with people happy to answer specific questions), CSound and others. If you feel like integrating with Java, all programming languages I mentioned (as well as Max etc) speak OSC, which is a UDP-based communications protocol for which there are .jar libraries to be found, and it's easy enough to roll your own if you want.

If you want to be able to control the mixing/fading in some way, it's easy to integrate MIDI pedals, drumpads or other controllers. You could do that with DAW solutions like Ableton Live too, if you want to. Also, I mentioned Max/MSP (which costs money), but there's also PureData, which is it's free predecessor.

There's lots and lots of stuff!

/Stefan

[MurphS]

Thanks for your suggestions, I will definitely start investigating all that you have mentioned...I can't wait to get home and start checking in to it. Fun at last.

The JAVA program I mentioned is something I have have tinkered with for years...I have gotten initial versions of it to work well enough, except that the garbage collector slightly interferes with the MIDI timing. If I explain what I am trying to do, might you be able to give me suggestions as to a potentially better language or approach to use for my audio programming?

What types of things are you working on?

Thanks again, Scott

[kkissinger]

I do this routinely with the VST automation in Cubase 5.

[MurphS]

Stephen,

I will dig through forums and look at the languages and software you mentioned...I should probably not have asked about my JAVA program in this forum...I'll seek out the proper venue for that question. Now on to the fun stuff...

Scott

[MurphS]

Kevin,

I've never used CUEBASE, so I don't know what exactly it can do...please allow me to be sure I have asked my question properly...

In my experience with DAW software (long ago), you had to have the parts already recorded, then you could automate the mix any way you would wish. I am hoping to have the 'mix' preprogrammed, but with no audio data recorded at all - I don't want to record audio at any point, this is only for live performance. My goal is to be able to play and essentially (my apologies to the soundmen out there) have an automated soundman... for things like fadeouts. My plans are more complicated than that, for otherwise it is a trivial problem, but that's the general idea.

So do I understand correctly that you can get CUEBASE 5 with VST automation to "sequence a mix", as it were, without having anything already recorded (other than the sequence, of course, with some type of beat to play along with)? If so, can you also automate things such as eq. and pan as well?

I hope I've explained what I am trying to accomplish well enough.


Thanks, Scott

[Antimon]

I think most DAWs are able to monitor inputs and then set automated envelopes (control values that change over time) on the monitoring. That would amount to automated mixing, I think.

I think there are a couple of Java programmers around, myself included. My experiments with Java have faultered on poor timing - I detect similar issues in your experiments? Java wasn't designed for the kind of exact timing that you need in music - SuperCollider and ChucK were. Especially has a timing implementation that hooks up to the sound card's audio clock, making it very exact and consistent in time calculation.

If you have any questions or comments regarding stuff like this I think you would be able to get at least a few responses if you posted them in an appropriate subforum here at electro-music.com. I for one am interested in these topics.

/Stefan

[MurphS]

Thanks, I sure didn't know that. Are the modern DAWs able to manipulate EQ and pan (an so on) on the incoming signals as well? I need to spend some time rediscovering the state of the art.

I have downloaded the MAX/MSP demo and am quite impressed and excited, and it looks like I can implement the program I've been working on with it instead - but I haven't looked at SuperCollider and ChucK yet. Umm, yes, I have had trouble with timing, which I could reduce but not eliminate entirely by using different garbage collector settings.I'll post to the appropriate forum to discuss programming issues any further.

Yes, very interesting stuff.

Thanks, Scott

[RFBB]

Yes, this can certainly be done with most if not all modern DAW such as Cubase or Ableton. I use Ableton (since version 5) and know that it can be done with or without any pre-recorded audio.

MAX and Chuck may be more suited to you since you have a grip on coding and language, I simply don't have the time nor desire for that but I know they have a well supported community and a lot of people on this board use them extensively.

There is an electronic music show in Johnson City, TN (just an hour from you) May 22nd. You should come out and we could discuss those automation functions with Ableton. I'd be more than happy to show you and I know there will be 3 other people performing with it as well that could answer any/all questions you may have. Good luck.

[kkissinger]

[MurphS]

Thanks guys, for letting me know. Time for me to take out my brain and beat the dust out of it, to better to be able to appreciate what has happened with music technology in the past 15 or 20 years.

RFBB, I would be delighted to attend the event you spoke of. That sounds absolutely great, and I really appreciate your offer. I'll be in touch with you when the time gets closer to work out the details. I've fooled around with Ableton, and have had it recommended to me, but I never got comfortable enough with it to see if it could do what I wished, so I'll be there. Thanks!

Kevin, I'm very impressed with your work. In most respects, that is exactly the type of thing I am endeavoring to do. BTW, I live about 4 miles, as the crow flies, from the MOOG mfg. plant here in Asheville. I'll go into a little detail here because you might be interested in what I am trying to do with the JAVA program I mentioned earlier. When completed, it will essentially will do the same as your rig (as far as record and playback are concerned), but there are differences ...imagine if your performance were being recorded into quarter note chunks (for explanation purposes only, it wouldn't really work like this) and so could be played back in any order or in any combination possible. For example, you could play a quarter-note scale, low to high, then have it play back in reverse order, high to low, while you are still playing and being recorded of course. Or, given the same scenario, you could simultaneously play back beat 1, 3, and 5 and have a chord which could repeat, change slightly..1,4 and 6, or whatever. The possibilities are mind boggling...well, for my mind anyway. It is not to be a looper, really, but you could sure use it as one. (It might get tedious to program it to play beat one, then play beat one again, ..... I'll probably need to provide for a simpler way to do that!).
I am currently having issues with the timing using JAVA, and I understand from Stefan that it might not be just me. So with the help and advice of these folks 'round here, I'm going to try to implement this program in a more suitable language (but I'll have to learn it first... ).

Thanks, Scott