 Hello, Oslidesync here. If you've been following the channel for a while, it'll be no surprise to hear that I'm currently into what I'll generically call slow music. And while I've not shied away from it in the past, it's definitely the case in the last few weird months. I've been going there a lot more often because it offers me a brain break. The process of preparing and arranging ambient or drone music is kind of meditative for me, just as much as the listening is. One of my favourite instruments to create drones with is the electron digitone, and I've been asked about my process a few times, so that's what this video is about. And it's going to be a long one. I know because I'm recording this introduction last, because we're going to go from nothing to a finished piece. But hopefully along the way there will be some ideas and tips that you'll be able to apply to your own music. And while the digitone is the instrument of choice here, a lot of the concepts are going to apply no matter what instruments you use. And either way, there's going to be some very chill music happening most of the time. I'm not trying to imply that I'm an expert in creating this kind of music, and I'm definitely not going to suggest that the process in this video is the definitive approach, but I wanted to share a workflow and some sound design techniques that I've enjoyed using. Before we get to the digitone though, I'd like to introduce my one rule for drones, which is move slowly, but never stop moving. Whether that movement is harmonic, timbral, rhythmic or melodic, move slowly but never stop moving. One of the elements of drone that especially interests me is how the line between all of those things is blurred. Often the notes that you choose to make up a drone are better considered to be harmonic or inharmonic, overtones of the drone. Conventional rhythm might be hard to find, but the beating between different frequencies or the interleaving modulation of sounds can imply a rhythm in the moment. And the compositional choices you make amongst these blurred lines can create tension and release. Musical questions and answers, themes and developments, just as you would in other arrangements, just, you know, much, much slower. So I'm here on an initialised pattern here, and let's start by thinking sort of in broad terms about the arrangement, if you like, what we're going to be using our tracks for. So my plan is firstly tracks to have a kind of a paddy kind of drone. Track three can be our sort of bassy drone, and track four will add some, probably some more sort of ambient melodic motifs, maybe just to create a little bit of interest here and there. So let's start with track one, which will be one of our paddy drones. What I plan to do is have one that is more sort of stable, that's a bit more grounding what's going on, and one that's a little bit more unstable. But we'll start with the more stable one first of all. So going to our amp page. Now I'm not necessarily an advocate for making all drone sounds have long attacks and long releases. And in fact when we get to the bass sound, we'll probably not work that way. But it does kind of sound good. Right, before we get into any of the other sound design on this one, I'm actually going to come into the second amp page here. And I'm just going to check on loads of reverb. Now again, adding loads of reverb on everything that you're doing within drone, not necessarily the best thing of all time is it gives you no back-to-front contrast if everything's drenched in reverb. And again when we get some of the other parts, we'll possibly try to make them a little bit drier so that they're not sort of just sort of soaked in reverb. But having a big long reverb does allow you some sound design tricks and by having it here at the start, you're kind of working it into the actual sound design process. So let's just make it nice and long. And also a bit darker. I'm just going to go down to one of these other tracks here. I'm just here, the tone of it. A little bit of rattle at the top, but otherwise I think that's just about right. And also we'll check on a load of delay as well. And we'll make that delay ping-pong. So we're going to delay page, done on ping-pong. Again, a little bit more feedback, but then darken the feedback. Low pass filter quite a lot. Again, we'll just come into this spare sound here. It's a much darker delay compared to the original sound, which I like. The other thing that we can do in the delay page, we'll actually get on to making the drone in a second, is we can send some delay into the reverb as well so that the delay itself reverberates. Which is kind of my preference most of the time. More feedback. I'll come back and tweak that as we go along anyway. I'm not surprised with the note. We now have a sound that's going to go on forever because it's got a load of reverb and a load of delay and then the delay is going into the reverb. Makes us able to have notes ring out for longer than they're playing, which when you're dealing with only eight voices of polyphony might well be a benefit. So let's create a bit more interest and indeed movement because as I said, everything has to be moving at all times. So, which algorithm? Let's go with algorithm two, which is essentially two lanes of two op voices. And let's start playing with the FM parameter a little bit. Let's maybe... I probably don't want to push this too high. It's kind of nice though. So one of the things I tend to do with these sorts of voices, because it helps with that clicky attack there, is I tend to come into the second page of Sin2 and I turn the phase reset off. That hasn't fixed everything yet, but then if we come into the amp page and if we turn the amp and vote reset off, got rid of that clickiness. There are other things you can do with the filter reset and stuff which can help as well, but that should do us the trick for the moment. So I'll listen to the other side as well. I think I'm particularly clever with the FM voices, but let's have them move a little bit. So maybe we can have one fade in slowly and a K down. So we can just set that on a regular way to zero. Sometimes I feel like these voices on the digga tone, especially when they're going hot into the reverb and the delay, when they're quite still siney, they can take up a lot of space in the low mids. So quite often I will come into the second filter page where you've got the bass width filter and just start to take out some of the bottom end. You can take quite a lot out, because of course remember we're going to have a bass voice in here as well. I just don't want it to take up quite so much of that bottom end, but more the low mids really. Okay, so let's think about our filter. We're going to switch over to the four pole. Bit of resonance on there. And again we can have crank the sustain all the way up and adjust the envelope to find my highest point. I think this one is still coming on a bit too hot there. At the moment everything's a bit stable, even though this is the stable voice. I'd like this to be a little less stable, so I'll just bring in a little bit of detune. So we get some quite low detune. So we get some of that beating. It's accentuated by a different ratio. Maybe if we come into the SYN-1 second page, maybe let's just try detuning at that higher ratio, a bit more of a bell tone, having that detune there and further detuning it manually on the ratios. Give it more of a bell tone. I like that. Okay, I'm going to go and adjust the LFOs next, but before I do that I'm going to set the tempo, which obviously is going to be really low. Should we go all the way down? 31. Let's show some restraint and not go all the way down. The reason I'm going to set my BPM for I do my LFOs is because in the LFOs, when you look at the multiplier here, you'll find that you have two different sections. So you've got the multipliers which are related to the BPM as they are now. One, two, four, power's two going up. And then you have the multipliers which are independent of the BPM. So if you're trying to make really slow movements, which is of course what we're looking for here, which one of these you choose, whether it's the one times BPM or the straight up one, will depend on what your BPM is actually set at because the ones without the BPM set here, I believe are set to relate to 120 BPM. So if your BPM, if your pattern is higher than 120, do you want slowest possible modulation? You want to be with the unsinked ones, but in our case, we want to be with the sinked one if we want to create very slow modulations indeed, which is what we're all about today. So back to the first page here. What are we going to do in terms of our movement? So the first thing I'm going to do is I'm going to add a small amount of very slow pitch wobble. So make this achingly slow. And something like, something like 0.1, something around there, it's probably the limit of what you can have without it becoming obviously out of tune. But now that we have LFOs and our reverb, so as the pitch drifts, the old pitch is sort of hanging out in the reverb and in the delay, just having that tiny bit of pitch movement. I'm just going to the second filter page here and just take out some of the extreme highs. Make a bit more mid-range focused. You can hear that you can get a little bit of beating in the tuning there, maybe push it a little bit further. You can listen there. Try it on the cusp of being out of tune now. Noticeably out of tune. And that's about right, I think. Okay, so we've got lots of movement in the envelopes, but the envelopes are all eventually and so we need some more timbrel movement to be introduced here. So what can we do for that? Well, we can maybe look at moving the level of A perhaps. Let's try that. So again, we'll set that really slow, which will be our BPMs. And slow, slow, but not the same as the other one. We want to have differences between them. Very pleasant. Okay, that'll do for our first voice. So let's actually sequence some notes here. So, again, let's make things as slow as possible because that's what we're doing, making slow music. So we'll go into our pattern here, funk and page. And we're going to do function yes for length per pattern because we want to have our patterns roll over each other. So if we want to have our patterns rolling over each other and different pattern lengths taking advantage of that, we want to make sure that our master length here is set to infinite so it doesn't keep resetting by default 16. So if you have different pattern lengths and this is set to 16, what will happen is that every 16 beats everything will reset anyway and we don't want that. We want it to go on infinitely. We're going to set our scale to be slow and we're going to set our length and I tend to just out of habits tend towards prime numbers that just seem to work nicely. So let's put an angle. Hang on before I do this, we want nice long notes, don't we? So in our trick page before we lay down any tricks, we're going to set these really long. Remember with our multipliers, you need to go really, really long. Now let's try 100 for the moment. So what I'm going to do here is I'm going to set down a trick on every single step, 13 steps. Obviously if we want to make something a little bit more complex, we might go on for more than 13 steps but we'll go for 13 in this case. So if we play now obviously all my lecture notes on, I'm just going to keep playing same note again and again. Obviously that's not what we're after here. So let's go in and actually lock in some different notes. So this is our more stable track, if you like. So I'm going to stick mostly these sort of diatonic notes. So we'll stick and see just because it's easy. So I don't want to speak too clearly to major or minor. So we'll try to avoid the major there too much but we'll stick it in sometimes. Have a nice high D, a little lower C because we tried that. Get the fifths. I was playing in a sharp B flat in C which I quite liked. So a higher one as well. And this is the programming of the notes is kind of going to be to taste and a little bit of music theory is not a terrible thing to have in this case. On 13, unlucky 13, let's put something that's not as diatonic. Let's try G sharp, maybe a higher one. When you're looking at discordances, let's be actually discordant. Let's try and have sharp. If you're thinking about discordances, the higher you put them from the lowest note if you like, the less clashy they'll be. So having our F sharp up in octave seven is probably not the worst thing in the world. I'll just save this. In fact, let's save the project line up or in the middle of it. What I would usually do now is go through these hold them down and assign probabilities to them. But because we now have the master probability here, I'm just going to set that to 20. Let's go with 20. And let's set that playing and see what comes out, obviously. We're not going to get a note on every single step. And of course, now we're recording a video. I still think that's too bright at its highest point. Let's make it a bit darker. And the note choices are... In many ways, it's possibly useful to think about the note choices in terms of harmonics that are happening here because we're building up an overall drone, of course. So it's not so much a chord progression as it is building up layers of harmonics within a drone. Okay, something that's moving too quickly for my liking. It's not going to be that decay there. I guess it must be one of these decays. Okay, so a couple of things we can do to make this pattern more interesting even on its own as a drone. I'll probably lower the probability when we start bringing in other things, but just so we can hear things at the moment. So the first thing we'll do is we'll come through each of these steps and let's parameter lock a different pan for all of them so that as the notes play, they spread themselves around the stereo field. And I'm just doing this more or less at random, but I do know that I want that discord note to be right out at the side. So what else can we do to make this a more interesting sound field? So one thing I like to do, if we consider in our LFOs, currently all of the notes that are playing are going to have themselves moving slowly but at the same rate. So what I tend to do is go into each of the steps and just nudge the speed for all of them differently so that when they do play over each other, they are moving at different speeds. Also maybe we'll just have a couple of them nudge their pitch a bit further outside the realms of good taste. Of course it's possible because of the probability that sometimes drop out together, but that's fine because we'll have other tracks going in there as well. And now we've got more interesting beating happening between those notes. We're creating tension, which means that we can have release to keep an eye on the voice usage a little bit as well, but we'll come back to that once we've got the rest of the stuff in there. We could also come into them and on some of them maybe push the detune a bit harder or less hard. Just create a little bit of variance in the different notes. Let's move on to our next part and create something with a little bit more instability, a little bit more tension for us to get release from. Okay, we're back with a sound here. So I'm going to create something with a bit more tension. So I'm going to go over to the most tensiony of all of the algorithms, which is algorithm six because it has all that cross modulation going on in there. Delicious. So this one again is going to be kind of a paddy sound. So we're going to set our, we'll give it an absolute reverb. We'll also set our amp envelope reset to off and in page two of Sin 2 we'll turn off the phase as well, phase reset as I should say. So let's set about creating a more tensiony kind of sound and we'll just kind of get the main sound happening and then we'll start introducing envelopes and stuff. So in this case A is going to be modulating both of these operators once I fade them in. And let's give a some manual detune here in page two. Here by detuning that we're getting that beating happening. Let's also tune C and B1 or two carriers away from each other as well. And here you can see that we can get rhythmic movement that's also basically timbral. So as we get different notes happening in this track, because of the frequencies beating against each other in different ways, we're going to be getting an implicit rhythm happening which can be quite exciting in the context of a drone. Let's try changing the waveform a little bit. It crass I think. So we'll stick with our sine waves or as close as, let's have a look at our ratios on B now. Now I can go something quite high but then detune that off wildly. You can hear as you start, because of the cross modulation, the fact that everything's detuned as you bring up that level there, we're getting quite interesting stuff happening. So perhaps we'll use Penelopho on that. Yeah, why not? Okay, let's just go ahead and say that we're going to do that. So sine A level. We want to be BPM based and slow, slow, slow. Slow. That's our midpoint and how far we're going to deviate. 40 and go 30 either way, which is quite a lot. So all of your LFO parameters are going to be relative to the way that the parameter actually works. So if I set SIN level A to 40 and I set my bipolar LFO to have a depth of 30, that's going to swing down to 10 and up to 70. Harshness has started to die off. I think we probably want this level to drop off over time as well. And I think we definitely don't want it to be this bright. And again, quite a lot of stuff is happening in the lower mids there, filled to some of that how quite vocal down at the bottom end there. Yeah, let's introduce some feedback. Not a lot of timbral movement happening now. So let's maybe have that fading over time. Maybe the same with A, bit of drive. Trying to make it a bit more dangerous sounding. Now that beating that's happening because of the detuned operators, it's happening at a very steady state. You can hear that extra menace come in there. So we can go a bit brighter. Anyway, sorry, high digress. That beating is happening at a standard rate and that's because the detuned ratio is always the same. So let's also nudge that a little bit. So if we go into our LFOs here, we can go down to... We can do this one of two ways actually. Okay, let's try this way first. So we're going to go down to ratio A offset, which is what we were adjusting on page two of SIN 1. Again, we'll set it nice and slow. Again, the LFO mode that I'm using here is the free running one, which given that I'm going to tweak the speed for all the different steps is what we want here. So just nudge this. Not by much. We can hear there that quite a lot of movement happening. Let's try B2 instead. Okay, so the other place we can do this. I'll try it and see what we like more. It's the pitch of A and B2. That was really slow. We get those side bands coming in more strongly than with the ratio offset. I like that, actually. I like that a lot. Is that still too bright? Maybe let's just take a little bit off the top here. Yeah, I think we can work with that. So again, let's set our length here. Let's go 17. And same deal. I'll lay down a bunch of steps. And I'll just go through and do the same thing. I'll set the notes. I will set the panning one more tweak the LFO times as well. So I'll do that through the power of editing. I think generally speaking, I'll go for higher notes, maybe with this one. Okay, here we go. Okay, then same deal. We've got our scale nice and low. We've got long notes. Let's have a look how these two parts now layer over each other. Okay, so I think in retrospect, the depth there is probably too high, maybe too bright still. Voice use as well, because that's going to inform. Okay, so I think some of the really discord notes on the second one. So when I'm doing a particularly spicy boy, probably can have them find a spicy boy. There's some spicy ones in here. Where are they? Where is my spicy one? Did I really not put like an F sharp in here? He's a bit spicy, so we'll turn him down having fade in over a longer time. So they're not quite as abrasive. So I think in terms of an instrument here, the really spicy boy was on this one. Again, the composition is kind of mixed up with the sound design intrinsically with drone music. Nice, okay. Right, time to ground this with some low end. So let's think about our bass drone. Okay, so let's give this some weight and ground it in a key so that things are held together. I'm going to do that of course with some bass. Digitone low bass is really easy to come by. Really low bass. You might need headphones for that one. Okay, so let's go to algorithm two again. Let's set C down to 4.5, so an octave even lower. Then mix between the two sides, so we have two sine waves and an octave apart. Slightly, a bit of touch and modulation. Drive it for some minutes. Okay, let's get some movement in there. Let's try just detuning the B carrier with a phasing. Here we don't want this long onset, but what I will do when it comes to the performance is I'll set the release time to infinite. Now, if I do that now, I start adding other notes. Obviously, we're going to get a goddamn mess. But in the most recent firmware, what we got with the Digitone is the new voicing modes, which is amazing. So what we can do now is set our play mode to mono instead. So that means that we can do things like this. So we set our amp back to infinite. We now have a voice, which is going to ring out ever, but not play over the top of each other. Scrumptious. So again, the same deal there. Set the amp envelope reset to off and set the phase reset to off. Try to help us get rid of the click filter movement each time we play the note. Standard envelope. Should try setting C even further down. Now, it comes to the question, do we want to put this through the reverb and the answer is probably yes. So that it spreads itself into space a little bit, but not as much as the other stuff. Chorus. What I want to do here is just high pass filter. The chorus a little bit. The pattern, see how that's feeling. Thinking about the sound design, a little bit of probabilities. Drop that filter even more and have a bit more resonance. So now I think that both of these need a little bit more top end taken off them. Now we have them in context with some bass. On this second voice now, modulation is too strong. Drop the range of the LFO that's moving the level of operator A. See what happened there? Our bass cut out. That's because its voice was stolen by one of the other tracks. So what we want to make sure is that we actually have one voice locked to our bass track. So that it never gets stolen. So it's a fairly simple bass line. It's not moving very much at the moment is it? So we should maybe think about how we can make a little more time roll movement here. So I think the first thing we should do is give it some very slow pitch movement so that it starts to beat against some of the other elements. Something around. Let's try just some simple filter movement. Let's just try the feedback. Closed off. You can't really hear it though. It's just more of a change in the weightiness. Let's give it some portamento. So the new thing with the portamento controls on the digitone is that we don't have to have full portamento so we don't have to have it go right the way across from where it started to where it's heading up to. I'm going to do that by lowering the amount. And I quite often set it as low as like 30 so it's only the last bit of the sweep that you hear. To make the time slightly lower without it. Sorry slightly longer. In terms of the programming of this, it tends to just to say I'll play this live. Otherwise the program will be much like we've already seen. The only difference would probably be that I would have the probability of anything that wasn't a C. Yes we will. So again we'll set that nice and low. Now we don't have to make the steps very long because of course we've got our envelope set to infinite. So we don't need to worry about that. So sorry notes. Okay so we'll just stick a bunch of these in. And again through the magic of editing there will be a baseline in a second. Okay so what we have is our sequence set up. The conditions of that first C is off so it will always play. And for most of the notes it'll be 25% and then when we get to a C I've upped it to 50%. We might tweak that as it plays but we'll see how it feels. Okay I think that's changing too often. So we'll lower these I think. Going for the boys grout itself back to that C there. So we don't have anything particularly unpleasant hanging for a long period of time. All of that time for us to get one of our other pad notes playing. That's how it can be sometimes with random tricks. Anyway we've got a track left so let's do something with it that's a little bit more melodic to just add a little bit more interest here and there. Okay so with this final track here I want to create not necessarily a melodic part but sort of a sequenced more notey less droney kind of part to punctuate what's going on with the rest of the drone. So I'm thinking maybe something that's sort of quite bell-like. So let's set about creating something bell-like. So go to the plucky way. Just go to algorithm, reduce the level of the oscillation, make it softer. So we've detuned everything by hand a little bit, chosen some slightly odd ratios anyway. Get some goddamn reverb. So I think it's a little bit too modulated, darker, darker, something like this I think. So let's try that in context. Okay so this sounds better tuned down so I'm just going to set its lift to minus one to dirt it up. You can hear that we've got a little click, a little glitch because of the voice stealing. So we do need to maybe address that a little bit. So we may want to, let's just say that we're only going to play one note at a time here. Let's give it some of that really slow pitch modulation as well. Automatically with each note going nice and fast and we want it on hold mode. Okay so in terms of the programming, our pattern length, just some bits. You can start to hear that some of the stuff is happening on sort of regular pulses and obviously we're going to lower our trick bar probability down real low. But let's make sure things aren't happening on a real obvious pulse. The way we can do that is by using our microtiming here. What we might want to do here is microtime things so that they're really close to each other as well. We can put more spicy boys in there. So if we were trying to avoid all together we could maybe lock a number of voices to each of these. Or we could just pray during a recording that we don't get any voice stealing. Drop the level of make more space for the bass. Always need space for the bass. And when you've got such a low sub bass you can be quite brutal with how much bottom end you pull out. Maybe we can go to the master section and see what it sounds like with the master overdrive on. Tape saturation vibe going on. So again from a mixing perspective I think maybe on these tracks here we could just make sure that stuff doesn't encroach so much on the center. Just make things a little bit wider just to keep it out of the way of the bass. The idea of mixing everything actually on the unit and only really having to apply mastering as far as possible don't tend to use over bridge so little tweaks like filtering moving sounds away from the center of the stereo spread especially when you've got so much low end going on can be really useful. I think that's pretty nice. I think we've been going on for quite long enough. The music you hear now and that you heard at the start of this video comes from a performance of the finished patch. I'll just add a quick note here about how I approach the final performance. Throughout the piece I'm adjusting the relative trig probabilities of three of the tracks. I'm doing this to try and influence the ebb and flow of the piece in real time knowing that if I want to increase the density of the main drone or the melodic punctuation I can bring up their probability or back them off to create space. Similarly if more tension is needed for contrast I can bring up the more unstable sound we put on track 2. I had to be a little aware of running out of voices but from putting the track together I had a fairly good feel for how far I could push the percentages. Ultimately whether a note plays is still a role of the dice but I enjoy this approach of trying to nudge the music where I wanted to go. It's something like trying to herd musical cats. Anyway if you made it this far thanks for sticking around. If you found the video interesting then please do the usual subscribe and like malarkey but seriously your support for the channel is always super appreciated. As always thank you for joining me. Take care of yourselves. Until next time.