The future of DJ controllers.

The future of DJ controllers.

You mean I drop $800 on a control surface, and I still have to use my keyboard and mouse?

Seriously, one day some company will truly understand what a controllerist needs from a MIDI controller, and that day the standard will be set and we will have the “steering wheel, three pedals and a gearstick” interface that everyone else will be measured by. Right now everyone’s thrashing around trying to work out what the minimum set of controls are, and it turns out there’s a lot more needed than originally thought.

Making a MIDI-only controller that maps the “mixing desk” idiom to the DJing problem misses the point - each channel needs way more controls than just a play button and a column of POTs. How do you drop a loop point on that channel? The same goes for the “two decks and a mixer” idiom where the idea of launching more than two sounds at the same time requires you to “special case” the other decks with modal keys (i.e. If the foot switch is down, decks controls are piped to C & D).

There has to be a better way that doesn’t try to recreate an old interface and repurpose it to the problem of controlling sound sources, and I think the key is that we only have two hands - there are only a few things that you can do to a sound source at any time, and there are only a few points in the sound chain where you would want to insert effects or interact with a device (a deck, a sampler or an FX unit). Map those operations to an interface and we might have something!

I have a feeling that future DJ tutorials will start to sound more like:

“Press Play to launch a new soundsource into group A, using nudge to get it into sync, then switch your attention to FX Send Group B to tweak the highpass filter on everything kicking off there. Use the crossfader to gently fade from group A to group D and then switch back to the Global FX…”

Look at it another way: you’re willing to pay $800 for a control surface when your laptop essentially has 100 MIDI buttons already?

But how do you define a “minimum” amount? I mean, why not use the computer also? I want my control surface to control everything my computer CAN’T, which are mainly the feel of pots and faders. Why do I need my control surface to drop loop points when my $1000 computer can do it? I almost never touch my mouse, and I don’t think it’s really needed beyond selecting a track, if you don’t set up hotkeys to select your search field.

But what’s the point of owning a computer at that point? Why not just use CD decks that have a loop function built in?

I mean, why avoid using the powerful tool you’ve already bought?

I hear you. I understand that the computer keyboard is a multi-purpose input device, it’s cheap, it’s ubiquitous and just about everyone can use it. Heck, just look at the lengths ProTools and video editing people go through to make their keyboard more closely tied to their workflow:

http://www.sweetwater.com/store/closeup/PTKeyWin--Main
http://www.datacal.com/video-editing.htm

My main aim of my rant is not to say that keyboards are crappy, I just think that the interfaces we seem to be getting from the hardware people are not really tackling the controllerism workflow. They’re taking an interface that supports a traditional workflow and asks you to map the controllerism problem to that interface. I was arguing for new interfaces, not remappings of new problems to old interfaces.

The key insight is that performing with sound is more about manipulating the controls on sound sources and effects than it is about “spinning decks”. Make an abstract model of mixing and see how far you can push the idea (grouping sounds, layers of effects, tying FX controls together on demand, flexible sound routing that is real-time reconfigurable, attach your control interface to the thing that needs to be manipulated then detach when you’re done), then make a device that supports that model rather than the two-decks or mixing-desk ideas.

Just my 2c. Not sure about the attach-detach concept as I really like the GUI rule of “all options are visibly available at all times”, but hey, it’s something different to try.

Otus Eks is one i myself am waiting for.
will be grand baby!

I don’t necessarily think this is possible. I mean, what is the standard “Controllerism” workflow? You can do all of this, you just need to re-alter a different form. Pick up a good MIDI controller keyboard and a trigger finger style controller and you can do almost anything with a creative use of commands. The Kontrol49 comes to mind.

I don’t think any manufacturer will ever be able to make a controller that will conform to what Ean does, what I do, what Bento does and what Chilly does, and then to what you do. Two decks and a mixer is a good place to start for all of us, and, in my mind, whatever makes it even easier for us to branch out and do something interesting from there is great.

:slight_smile:

Naysayers, already! Can’t be done! Bah, I say. BAH! :slight_smile:

In the science of Workflow Analysis there is the concept of “state” and “transitions” (“finite state machines” if you know the jargon). Let’s say you are at a classic “2 decks & a mixer” workstation and have two tracks running in sync. At any moment in time there are a finite set of operations you can do next: from where you are you can crossfade, tweak FX controls, stop a track, pitch up or down, etc. Many of those operations are short events that leave you pretty much where you started ready to do another tweak - as soon as you change a filter value, you are free to change it again. A few of those operations are more involved, like stopping a deck and taking the record off. Once you’ve committed to that operation your workflow is pretty linear - you’ve got a certain amount of time to put away the old record, find a new one and get it cued up. During that operation there is a slight chance you may want to tweak a filter on the running track, but you will definitely never going to be altering the pitch control on the silent deck while you’re diving into your record bag. Why would you bother? This is the concept of the “Two Hands” rule - you are never going to be using all of the knobs at the same time, there is a physically limited number of events that you can be operating simultaneously. (Computer help you add more events to that list by looking after things for you, like running Deck C, and this is the beauty of machines.)

The sounds coming out of your setup are called the “states” and the operations you have to go through to get to the next state are called the “transitions”. A good controllers job (and the software too) is to make these transitions as painless as possible so you can get down to the job of making and shaping sounds. These kinds of things are often diagrammed as dots connected by directed arrows, the dots are states, the arrows are transitions. RULES OF PLAY: Once you take a transition you can’t go back unless there’s a different arrow going in that direction (a “directed graph”).

Looking at the problem from that point of view, having a TriggerFinger set up beside your decks is great for the short, open ended events - you have many more of them at any moment! These are short events are diagrammed as little circles that that loop back to the current state. The TriggerFinger does nothing to help you do the more involved linear operations, which is also the place where the software often forces you to reach for your mouse and go interact with your DJ software - get mouse, move to the finder, select a song, right-click, select “load on Deck B”, wait, done. (Ableton Live helps by allowing you to preprocess some of this stuff into sample groups that you can preload into the grid, but these advantages are offset by it’s painful file finder, so it ends up about equal IMO).

I am betting that if you did the analysis with a good number of DJs, once they’re into their flow I would expect to see similar work patterns for their major tasks: lots of short transitions, a few set piece bits they’ve rehearsed and a lot of basic shuffling through the linear tasks. Yes, each one will have their own flavor of embellishment (Moldova’s sample & reorder controls for instance), but my thinking is that the extra controls on your controller above and beyond the basics is where you map these personalized embellishments. In my view it’s essential that controllers have more buttons, sliders and wibbly gee-gaws than strictly necessary, otherwise you will have nowhere to stamp your style on a set (which is what leads people to add TriggerFingers and knob boxes to their setups!) The Otus guy understood that in the video.

So, a bit handwavy for a grand unified theory of Controllerism, but I don’t subscribe to the “You can’t please everyone” camp. Maybe not immediately, but I think we can do a lot better than we are right now by thoughtfully applying Workflow Analysis.

Wow, that was long. I’ll shut up now.

this kind of conversation is why i like this website…good stuff.

Yes, if you did an analysis of most DJs today you would find similar work flows. But we aren’t talking about regular DJs here, who stand over turntables and a mixer. We’re talking about Controllerists, people who have customized and altered software and workflows into something that the art has never seen before.

Lining those people up will result in completely different workflows and answers. I don’t think there is a universal solution yet. Can there be? Well, of course. I think the VCI-100 (and its type) is pretty damn close to a universal solution. But look at what Ean did. He took that and created a MORE universal solution for him, but others wouldn’t want their Cues in that location, they want them somewhere else. So now it isn’t universal. It’s personal and individualistic.

I think this style is too new to have anything resembling a universal solution, or really any true way of approaching it. People like Moldover, Ean, and Bento are going to be pushing this into having solutions, but there will be many and they will then be tweaked and changed for whoever is using it. We are no longer restricted to two turntables and a mixer, and therefore that solution is not universal anymore.

I hope that makes sense. I got about 4 hours of a sleep and my head feels really weird… heh
:slight_smile:

I really don’t think that a universal system is the right way to go with these. I am reminded of a conversation I had with a DJ friend of mine. I had asked him how do I “make it”. He told me that I had to find and do something new. Something that no one was doing, something unique. It had to be the whole package, not just the way I mix and the tracks I select, but the way I present myself both to promoters and other DJs but to the crowds. Its the whole package.
Having a universal system is not going to help DJs accomplish this. You need to find the combination of hardware, software and workflow that not only makes you unique in the presentation, but also in equipment.

I’m not advocating a Universal-System-That-Everyone-Must-Use, I’m just asking for a new starting point that isn’t 2-decks or a set of mixing-strips. A starting point from which you can innovate, instead of spending all your energy trying to map the basics of controllerism to your device before you can even get around to doing anything interesting.

I’m essentially very lazy, which is why I do all this work. Not logical, but what can you do.

i agree.

i don’t think u need hundreds of knobs, buttons, faders, fx pads, and whirly-gigs. I think a slightly larger, more sophiscated vci-100-esque controller could do the job.. just have one button that cycles through layouts.. kinda like how the se has 3 different modes, but on a grander scale.