Studio Diary: August 16, 2023
No-input feedback loops as drum samples
In today’s studio diary entry, I’m exploring a technique that was kindly shared by user trytyke over on one of the internet’s loveliest and enjoyable corners, Lines. I think what they’ve uncovered - a method of using the octatrack to endlessly loop and modulate “no-input” noise and feedback - is brilliant, and I had to have a go with it.
I’ll follow my general, standard agenda for diary entries:
The full performance video
Some personal background/context/musing
Step-by-step demonstration of the process
Closing thoughts
Before continuing, let me just quickly welcome some of the new subscribers who have jumped on board the last couple weeks. I really appreciate your support and I don’t take it for granted. I have a blast writing this newsletter, and will continue to put full effort into making it worthwhile and interesting (I hope). Thanks again!
Onward…
The full video
So, what is going on here?
All of the percussion heard in this piece is made entirely from sampling the internal self-noise of the Octatrack, then zooming in on extremely short slices of that audio for playback.
For the uninitiated, this is “no-input mixing:” a method of extracting sound and noise from a mixer without any external sound sources plugged into the input of a device, then manipulating that sound in creative ways. I’ll come right out and say that I don’t think I’m the best person to give any kind of history lesson on this music, and I don’t want to sound didactic. Many musicians have utilized feedback loops in creative ways over the years, and many writers can probably explain more effectively and comprehensively all of the historical examples that are out there.
But for purposes of providing you with a reference point for this video, and to give you a sense of the type of sounds I was trying to make to then sample from, I tend to think of the onkyo movement that came out of Japan in the late 1990s and early 2000s as the best example of this method. Bandcamp did a really great retrospective on Toshimaru Nakamura’s career if you want a primer of what this music can sound like (directly from the master of it, no less - in my opinion of course).
Music history aside, I want to highlight something about making this video that I really enjoyed: the process itself is something I never would have thought to try were it not for hearing about it from a stranger on the internet who decided to share their idea.
I mentioned in my very first post on this Substack account that one of my motivations for making the site was to leave behind the din of traditional social media and create alternative spaces online that I actually enjoy being in, especially as a musician making non-mainstream music. Leaving Facebook and X (and mostly IG, which I only use to share my posts here), and focusing on this instead, has been tremendously beneficial for my mental health. I also wanted a place that allowed for longer, freer explanations of what’s actually happening in the music I share. I love understanding the context of a creation, and because I think we’ve all seen the nicely-staged videos on Instagram and YouTube with potted plants and thousands of dollars of gear with no explanations, I’d like to ensure that any videos I share are presented with a little more depth and a “pay-it-forward” mentality than a glorified product placement.
To that end, one of my favorite places on the internet, and one that inspired this whole line of thinking to begin with, is the Lines forum. The community there has its origins in monome gear and so it features plenty of discussion on new scripts under development for the flagship Norns device, but you don’t have to be a Norns user to have an enriching experience there. In fact, I think the overwhelming majority of discussion happening there is non-monome in nature. The place is filled with immensely creative, welcoming, and generous people sharing thoughts and ideas on a huge array of musical and non-musical topics. I learn something new every time I open the site.
This video was inspired by something that user trytyke shared there in a post in the Octatrack tips & tricks thread:
I highly encourage you to visit the original posts and listen to the sound samples they’ve shared, as well as the subsequent stream of clips and examples that emerged from other people excited about this technique and trying it for themselves. There are beautiful sounds there, full of movement and unexpectedly rich detail for such a simple process.
I was quite taken by how the results other people seemed to be much more dynamic than I would’ve expected, so I decided to try this for myself.
The process (as written)
If you aren’t a user of the Elektron Octatrack, it’s tough to succinctly explain what’s happening here, so I’ll try to speak as generally as possible. Basically, when one decides to record a sample to the internal device memory, one must decide which source they would like to sample from. In the Octatrack’s case, you’re allowed to sample from input pair A+B, input pair C+D, or from internal audio playback (this can be any of the individual tracks in isolation, or the master mix coming from the main or cue output pairs). You must also decide how the sample is recorded: you can set specific steps in the sequencer on which you’d like recording to begin, and for how long, and how for how many more times before stopping (just once, or endless, for instance).
If you’ll notice in the original example given, this particular user is letting tracks 1 through 4 record themselves ad infinitum, while also letting one of them record the main output of all tracks together…all while LFOs are modifying the sound of the Octatrack’s internal noise. You can imagine that this results in a loop that evolves and shifts pretty dramatically over long periods of time.
Basically, it works like this: by amplifying the volume a huge amount, the Octatrack’s internal noise is made audible. Then, by adjusting other effects like filter, comb filter, etc. with LFOs, this audible, mysterious machine noise is shaped and modified over time. Overdubbing the recordings over time will capture a series of evolving, subtle changes.
trytyke is letting the internal feedback loop go for indeterminate periods - in one case more than 7 minutes - while the internal self-noise and static of the machine is eventually modulated into something interesting.
This can result in very interesting and unexpected audio glitches, as evidenced in the recordings users are sharing. It’s unknown why some sounds are panned hard left and right while some aren’t, for instance, and it’s impossible to predict exactly what you’ll get over time. Even very subtle changes to any parameters result in drastic differences in the ultimate outcome.
Sounds fun, right?
The process (as I performed it)
I took a slightly different path when trying this process. I did follow the steps as outlined, but instead of simply recording and sharing the bleeps and bloops as the turned out, I took the opportunity to turn those sounds into sample sources and make a “clicks & cuts” style drum rack with them.
I used the technique outlined by trytyke to build up 5 different feedback loops to sample from, and make 5 potential pieces of a drum kit. While I don’t think that my sounds are anywhere near as interesting as what other people achieved, I do think it’s fascinating to hear just how different they all sound even though they were all made by following the same flow: sampling just the internal track itself and the master output together.
Listen to the sounds in isolation here, but also look at the waveforms! I love the variety of shapes and outcomes. Again, all of these noises came from within the machine alone, with NO other sources plugged in - just the OT itself being modified and overdubbed over time. I just picked my favorite parts to save here.
With these samples recorded and ready to go, I took each of these sounds and selected extremely short slices of them to sequence as a cohesive, noisy drum kit. You’ll notice that on my Octatrack, I have a few tracks set up as Flex machines and a few as Static machines. The only reason for this is to conserve RAM memory and free up space for re-sampling live playback later. I do this a lot with on-the-fly recordings: I’ll take my favorite flex samples I record while jamming and save them to Static slots instead, so that they’re read from the Compact Flash and not taking up precious Flex memory.
Note that the Celeste sample I used is the only bit of tonal sound happening - I made 2 copies of this sound (which is just a single note), each with slightly different effects and LFOs applied to things like pan, filter, and volume. The crossfader is controlling the pitch of the second layer. When I slide it to the right to move to scene 9, that layer is pitched up an octave.
Anyway, after a somewhat tedious and selective mapping of extensive re-trigging, slicing, and sequencing, here’s what things sounded like together. Here you can get the gist of how these sounds work in the context of the final video. Sorry for my arm getting in the way a bit sometimes…I had to cobble together a weird setup to make things visible for the camera.
Of course, things wouldn’t be complete if I didn’t bring the iPad into the mix somehow. In this case, I recorded the piece directly into AUM via a Zoom U-24 interface. Fabfilter Pro-Q EQ, C-2 compressor, and L-2 limiter round out the master chain for finishing touches.
Conclusions
I have to say that this was one of the more invigorating Octatrack jam sessions I’ve had in some time. It continues to amaze me how deep this machine is, and this little exercise further solidified for me why it’s one of my favorite instruments of all time. It seems that there’s no final destination within this box, and it really feels like a timeless device in an era when so many consumer electronics are destined to “age out” after a year or two. Here’s to hoping that Elektron continues to provide updates and service to this magical machine for years to come.
I’d also like to again thank the Lines community for inspiring me on a near-daily basis, and especially to trytyke for taking the time to kindly share his process and audio samples over there in the fantastic Octatrack help thread. I’ve learned so much from that group, and I’m excited to put this new technique into practice on future recordings.
On the topic of future recordings - I should note that I’ve received masters for the album I’ve previewed before. It came out even better than expected, and I’m looking forward to sharing that at some point.
I’m also working on setting up some interviews and other collaborations to feature on this page. As mentioned previously, I’d really like this to be a place that is inspiring to visit and sparks ideas about methods of music creation. Thanks again for being here, and reach out any time with questions, feedback, and comments.
-Andrew
is simply spectacular! wow