Better Drum Programming

I remember some:

You can do a lot of this stuff with the OT.

A recent one, that’s quite nice :slight_smile:

And of course everything by @nedrush for crazy break stuff :slight_smile:
(see @Bwax post above)

This thread might be interesting to you as well:


This might seem really simple but I’d start by making beats with just 4 or 5 sounds max. It’s really easy to create kits with loads of interesting stuff but you may find yourself adding sounds just to compensate for a lack of a decent structure.

But then on the flip side maybe chuck a load of sounds together, go wild and pick out the interesting bits & dissect them.


Another great piece of advice! Sprinkling a ”busy” rimshot pattern on top of a very uninteresting beat, then add yet another glitchy sounding pattern on top, yep definitely done that. And I have a tendency for maximalism but I’m working on it! :grinning:

Thanks so much everybody so far and keep them suggestions coming! Will take a good look at the links in the coming days but start with the practising TONIGHT.

1 Like

I’d like to echo people’s recommendations for dissecting other’s work. When first trying to wrap my head around making good drum patterns something i did a lot of was finding songs that had melodies that felt in line with what i lean towards in my own writing and doing note for note recreations of the drums to get a detailed view of how they were made. I wouldn’t recommend using these recreations in your own songs but approaching it as an exercise like transcribing someone else’s work on guitar.

That being said nothing beats just making as much music as possible and experimenting as much as possible.

1 Like

I’m struggling with the same problems as the OP. Lots of good links here. Beats Dissected looks very promising. Thanks for the tipps

@korpinen I took the liberty to merge your thread with a related discussion.
Don’t hesitate to check what has been said above, and the following ones as well:


Listen to Jaki Liebezeit.


i THIS this.


Cheers! (As long as it’s ok we also talk about tips that are not specific to Digitakt.)

1 Like

Damn good point. Fixed.


Increasing the sequencer resolution by changing the track scale is great for creating more complex rhythms or just getting more intricate timings.


Electro, hip hop. Listen to that and twist it up.
Crazify it.

Training assignment: recreate beats from Syro at 16rpm then increase speed x2 :slight_smile:
Then come back and tell us what you’ve learned.


With your Elektrons at your disposal, definitely take advantage of probabilities. For example, just set up a basic groove with kick, snare and hats then set the kicks and hats to 90%. You’ll get lots of happy accidents that way. You can then record them and pick out the best bits. Or else get your beat going and resample it as a whole as if you were using a breakbeat (OT is perfect for this) then play it back using some random trigs across 16 steps, go into each trig and p-lock the start points randomly. You don’t need to be in total control of everything is my point. Experiment!


I would not be too dismissive of “genres” when it comes to grooves. Much of traditional music (e.g , Indian, Persian, Afro-cuban, but also European) builds on codified/categorized rhythmic patterns, often related to ritual and/or collective dance, which require coordination, and also to some extent build ups, drops, etc. Artistry can also be to vary within an established pattern.


100% agree! The OT took my drum programming to the next level, since it adds a nice layer of “mess”, and extensive use of probability on the Rytm was the first step to that delightful mess.

In the last year after posting this original thread, I spent time learning genre-specific drum patterns. I also got over the idea of needing each drum groove to be definitely unique. Drums offer a strong foundation to the rest of the track, and not every song needs the drums to play lead like Keith Moon.

By accepting some genre conventions/cliches, I was able to quickly start new tracks while gradually adding unique flavor to the patterns.

But, a glitched out drums-as-lead track is also a great project and fun… just takes practice!


Yes! I love probability on Rytm so much that I just leave them in the final tracks and fine tune the probabilities to taste so that some are way more likely and others are only once every few bars. It does suit my style of headphone hip hop tho. I’m not sure people down the club would be so into it when they’re trying to dance and beats keep getting missed, making them look silly hahahaha


Very good points in this thread!

I add one: Always count the bars in your head.

Have a break for each 8 bars (except some tension tweaking) and change something small after each 8 bars. Every 4 rotations change a structural element (new percussion, other pattern etc). This can go on forever, during TV or in the club :grin:


that sounds cool!

you know…after deep diving the Ae SysEx…its seems like probability isn’t really a thing. sure it may be used to find ideas. but when it comes down to the actual tracks…its just many versions of the same pattern in succession with slight variance.

I think random, tho neat and something I was wanting to do LOTS of, is actually less interesting to the brain. it prefers [so I’ve read] to find patterns, even in chaos. so perceived random is more interesting than actual random, where you can tend to lose interest.

just shit I was reading about making tunes. and shit I’m noticing with AFX and Ae etc…