Creating a DJ set

I have some early beats/songs in a snapshot on my MD and have been struggling for years to figure out a good way to create songs on which I perform with/on the MD.

I think It’s about time that I ask for some help now.

I have a MegaCommandLive, Ableton, an APC Mini, a PadKontrol and a BCR-2000. MD lives in a flight case with an ESI-2000 sampler, a SE-50 FX unit, basic mixer, MIDI and audio interface.

Mainly I am stuck on how to divide up the MD’s tracks, and how/if I should utilize Ableton or MCL to wrangle everything.

What I would like:

  • To set up the BCR-2000 to control all 24 parameters of any track. It has 32 presets so using up 16 of them isn’t unreasonable. I have done a few performances on the MD but what I don’t like is how params are accessed, and how cramped the knobs are.
  • A consistent track layout or at least reduce the number that I need to worry about
  • To figure out I am going to (re-)arrange my songs. On MD? MCL? Ableton? (If Ableton, I believe I would prefer if it was a combo of audio clips and tweakable MD patterns, synced to Ableton or the other way around.) I would prefer a progression of patterns to jamming on a single pattern, though having some jamming sections might be fun.
  • Figure out which devices it makes sense to use right now. I’d rank the non-sound-generators this way, from most to least useful in this context: BCR-2000, MCL, APC Mini, and PadKontrol.

I would also like to involve my trusty Model:Cycles since I created a really nice sound combining the two in previous performances. The thing is it was too much to manage, mutes and pattern changes. I wonder if the thing to do is beforehand set up the patterns on both devices to match up.

Or if arranging audio clips in Ableton and lifting the burden of the machines so I can jam freely over the songs would be better.

The preference is to not require a laptop. One I don’t like having to bring my $1000 gaming machine to a crowded place, and I would like to avoid the troubleshooting that can be required or glitches that can happen. But I will do it if the pros outweigh the cons.

That’s all I can think of right now - I have tried many different setups and still feel like I don’t know anything.

1 Like

For the mutes, maybe you can find a midi controller that would take care of this.

As for the live thing, I have decided that song mode was not my thing, so I’m personally left with improvisation and a lot of rehearsal I guess.
That would be my advice, I think with practice you’ll get to find your way with two machines.


The BCR-2000 has 16 buttons so they are perfect for muting the MD.

I think I like song mode? The downside to song mode and pattern…matching?.. is a loss of control … like sound changes on the Cycles, getting reset with the pattern change … unless I use song mutes to create progressions most of the time. At least sounds will retain when patterns are revisited, so it might not be so bad.

Using a laptop would give me much more compositional freedom while also making improvisation a lot easier on me with my still rough skills. But I already mentioned the reasons for my reluctance. A friend suggested I get into iPad…

I think I need a starting place, a new workflow. Something that I develop hand-in-hand with my performance. Bring elements of the existing patterns in gradually and finish this set. MCL is nice … but there is that tradeoff of sounds resetting on each new pattern, now with the MD too and in a way where param changes are never remembered. Lame. But, trig conditions, and the ability to mix-and-match track sequences and sounds, might make up for it …

Oh, I just want to keep it simple!

Maybe get yourself a cheap mixer and make use of MD separate outs?
You can also hard pan M:C tracks to get a separation between e.g. a kick and some low end sound…
Using a mixer to mute tracks is easier, I guess, and you would have control over the volumes as well. Might be a good lead.


less procrastination, more: just do it


Just do what though??? I keep trying stuff and painting myself into a corner :frowning:

Keep in mind we’re all at least 2x as dumb when performing in front of a crowd. I’m probably 3x as dumb. Simplify your setup. Couple of tricks that I’ve found helpful when playing with Elektron boxes:

  • Each track is n patterns. In my setup, it’s four patterns per track…always. This way, in my 3x as dumb live brain, I don’t have to think “is that next track on pattern 12 or pattern 2?”
  • Each track lives in the same place in the bank/pattern structure, to the greatest extent it’s feasible (e.g., A1 on my RYTM corresponds to A1 on my A4).
  • A single master controlling program changes will restrict your flexibility to “jam”…I prefer doing pattern changes manually.
  • If you’re not performing atonal techno, group your tracks by key in a logical structure. In my setup, each bank is restricted to a key, and I walk around the “camelot wheel” by bank. This way, in my extra dumb live brain, I can overlay patterns flexibly (no track-list needed!) without making dumb mistakes that sound bad.
  • Don’t be afraid to let things stretch out, time is moving different in your brain vs. the crowd’s brains
  • Whatever can go wrong will go wrong. Again, simplify the setup.

Here’s what I’m going with for the moment:

  • MD
  • MCL

I pulled some knobs off of the BCR and swapped the MD’s for them and it’s much better. It revealed how low-resolution the encoders on the MD are…

I’ve found that mix mode lends itself to altering params 4 tracks at a time so I’ll be organizing my tracks in groups of four… well they kind of already were due to the standard track labels but I’ll be keeping it in mind…

I plan on adding the M:C, after I’ve figured out where my MD sequences are gonna live and if I’m gonna mess with song mode again (tried it but still on the fence). M:C will be synced and partially sequenced by the MCL depending on if I want live tweaking without sacrificing sequencing freedom, or p-locking / pre-existing patterns.

I spent an hour creating 16 presets for the BCR but after trying it out I found that while it felt good, switching tracks was clunky and having to guess at the current value of a parameter made things more complicated, so I’m on the fence about it. It would be nice if it was always in sync with the current values and maybe follow the current track, then it would be the ideal MD control surface.

If nobody minds I’ll post updates as I progress.


After a couple days’ work tweaking the flight case (and still not finished with that), I’m finally spending some more time with MCL. It’s … pretty complicated! My least favorite aspect is the save/load paradigm … like how when you copy/clear/paste slots that it doesn’t reflect the working sequence. You’ve gotta reload the row and if you’ve edited anything, it’ll be gone! Also, having to go to the Grid page to save your work. It make sense but … You have to think more discretely I suppose…

Anyway, I’ll be using tracks 1-3 of MCL’s grid Y to control the ESI and 4-6 to control the M:C. ESI of course for expanded sampling. I came up with a way to add controllable FX (without having to track down the expansion board) by routing the right channel to my FX unit. That way, Pan can act as dry/wet.

Here’s a photo of the setup…

This is actually the second attempt at this, which originally didn’t have MCL because I didn’t know about it. As some of you might already know I was about to sell the MD, but I’ve had a big change of heart. So far, it seems like it could work, but I gotta say, I will jump to Ableton and not look back if I can’t get used to the MCL sequencer, and just use it for its other functions. ESI would stay even though Ableton can do sampling… it’s just simpler.

All of this work just to get around MD’s limitations. I keep having to remind myself that nothing sounds like it.

Alternatively I’ve thought about getting a Digitakt. That would give me consistency and self-containment, in exchange for giving up 8 tracks and a frightful number of insert effects.

Actually an Octatrack would probably make a lot more sense than a Digitakt… I’ve tried it though and let’s just say… I’d rather use Ableton

the legends were true… Screenshot 2022-03-20 at 06.58.24

looking good, got any examples of this combo in action?

1 Like

Only I know where the button cap is kept. Someday it’ll be worth a fortune …


Work on your pattern changes so that one track blends into the next. Nothing worse than that awkward silence


I thought I’d post a little update.

MCL is so intense. I thought I had a good idea of what it can do, but it actually does so much more than what I could do with MD alone or an external sequencer.

Because MCL’s enhancements are so extensive, I am trying to do everything outside the box. This means some things are not as simple as recording an audio clip… if I want pads, I am limited to the chord machine on M:C, sampling a synth into the ESI, or bringing a synth along and sequencing it with MCL. Constantly having to be aware of the cursor in the grid before saving changes is not the greatest and hopefully something that I can get used to. It’s easy to do the wrong thing and lose work, if you get load/save swapped or save over the wrong row. Not only that but saving must be done from the Grid page. There must be a better way…

I tried sampling and sequencing a sound with ESI, and that went smoother than I expected. The easy part is the sampling, the part that sucks is creating a preset. Hopefully that gets easier.

Anyway, it’s a powerful sampler, and I think I made a good choice here. For one thing it’s really nice to be able to instantly switch programs for a given part at the beginning of a pattern. Also it should be pretty easy to create VA sounds on thanks to its wide array of filter types. Wonder if a preset can stack voices to create thick pads…

I am thinking that maybe I should take my time learning MCL, split this into phases. I should fire up Ableton and just make my tracks, sequencing on MCL, but recording the output. Just to get ideas down, finish some tracks. Then maybe I’ll circle back and figure out the non-linear version. In Ableton it’s tempting to just record audio clips of anything but if I do too much of that I could be forced to integrate the laptop into the setup. I have a Max4Live device I made a long time ago to launch a scene from program change. So I know at least I’ll have that ready to go if I need it to follow along with MCL.

Progress is still slow because I’m still learning the MCL and I’m questioning the cost-vs-reward ratio. There’s always plan B… which is just getting basically everything into Ableton, creating patterns on the MD solely for the purpose of improvisation while the backbone is pretty fixed, instead of the MD doing double-duty. I guess going another route depends if plan A clicks.

I just threw things together in the Soundcloud set and winged it. People enjoyed the performance, but people I showed the recording to said it was too experimental and too repetitive. So I’m looking for a balance, something with more arranged elements and structure, and space for human performance and improvisation.

I would go a step further and suggest, if you have the funds, getting a DJ mixer. There are pros and cons to a DJ mixer versus a traditional “mixing desk” but if you mean to perform live a DJ mixer is much easier to use live IMO. I’d say you’d want three-four channels. Drums, bass, and the rest on a three channel mixer. This admittedly could be a more expensive option but maybe you can find a business that would do a long term rental or borrow one?

Maybe I’m getting things wrong by calling it a DJ set? To me it’s a DJ set but do people assume that that means I need an external means of mixing?

Currently I’m imagining that all the mixing will be done directly on the grooveboxes and flight case mixer. Maybe the BCR-2000 could help wrangle things if that’s too complicated / not fluid enough. It has 32 knobs and 20 buttons. I’m still just trying to figure out the workflow for creating the tracks and barely starting to figure out what my role is gonna be during a performance.

Maybe I don’t know enough about what a DJ actually does. :laughing:

1 Like

Honestly I’ve kind of been half following this thread and saw we were talking about mixers so I threw in on that lol

For myself though if I were using just the Machinedrum, grouping sounds to the audio outputs and sending them to mixer would simplify a lot of things for me (transitions, mutes, live EQ). I would prefer patterns on the MD be fairly static–I would tweak FX on the MD patterns live and use the DJ mixer to handle live filter/EQ stuff. I’m not sure if I would use a song mode or not.

Very strictly speaking, to me, a DJ set is someone seamlessly blending pre-recorded tracks together using turntables/CDJs and a DJ mixer. I am a DJ who performs like this hence my bias in my suggestion.

1 Like

To me, it’s a DJ set if I’m playing other peoples’ music. If I’m playing my own, it’s a “live performance”. The lines between the two in terms of techniques and gear used have blurred so much recently that I think the only distinction is content.

(And if you wanna pick me up on this and point at scratch DJs as blurring the lines even more, I wont argue)


I guess it’s just a distinction I make for myself. Because I make music I consider to be “pop” where I mostly sing, whereas this project is meant to be dance music played in a bar and I’m “DJing” my own stuff.

Also because when I play stuff like this people ask me how long I’ve been DJing. :sweat_smile:

To me it’s also useful as a reference point on how much musicality I have to put on display.

I can’t call it a “dance set” though because all my music is dance music.

1 Like

Definitely this! Tried it with a delay in aux 1 and reverb in aux 2, it’s ace.

Last summer I played a live set with just a MD UW x.04 (or was it x.05? I forget).

At the time I used MD’s “Song” mode to build the whole set into one really really long song. Utilizing mutes and loops I could make entire songs with parts that progressed automatically, and loops where I could live mute / perform. To transition tracks I’d often sample the end of one track into a UW buffer, and then mix out of that and into the next track (or glitch it to heck and back and then drop the next part).

I’ve since bought an MCL and… haven’t learnt it yet. I want to get into it again but it sucks going back to “reading the manual” after reaching a solid understanding of how to use the MD by itself.

My suggestion is to keep it simple. One of the reasons I ended up performing like this was because I found a progression from jamming out an idea -> performing it on twitch -> take the best parts and build a set. Also it was useful that by building up a “song” for the structure, I could incrementally work on the live set and come out after a week or two with more minutes put in.