 Hey, it's Anfa, and you're watching not Anfa Vlog this time. This time it's a video, actually it's gonna be a few videos about how I made the track Bodacious, which was my entry to the open-source music FM synthesizer challenge, that's a hard name, and it won, actually I was voted first place. So I guess that making off could be interesting, but it's very long and uncut. So you can see the whole process from start to finish, exactly as I just did it. And you can see all my struggle and going back and forth and trying things and failing and trying different things, and trying to fix what didn't really work out, and I think it's gonna be interesting. Yeah, so bear in mind, I recorded this quite a few days ago, and I didn't know things I know now. I wanted to make this actually an exercise and learn Oxy FM synth, because I think it's pretty nice, but I also resorted to using two instances of text as well. So yeah, enjoy. Hey, it's Anfa, and you're watching Anfa Vlog. Today's video is gonna be different, because lately I've been trying to make my videos shorter and more to the score, to be more robust, less full of digressions and stuff like this, but this video is gonna be absolutely the opposite. This video is gonna be long and weird. So I was notified about an ongoing open source FM synthesis challenge, a music challenge hosted by a community of Linux nerds in Cologne, which is also the place that hosts Sonoy's convention, which I went to and I'm going to this year. Anyway, the thing is, there's a list of available open source synthesizers that can be used. You can use any open source FX processors. You can't use any proprietary ones. You can use proprietary DAWs and a limited number of built-in effects in these proprietary DAWs, but not much else. And well, I decided to give it a go. I had some experiences with DAXID and OXA FM synth. These are two strictly FM synthesizers I've been using a bit, apart from Opula NZ, which I didn't too much use in LMS. So DAXID was kind of a... it was very difficult to understand. I couldn't understand how the envelopes are supposed to work. It's super weird. Totally not what I'm used to. Not ADSR. Very strange. So I gave it a go once, not really got on with it twice. Didn't help either. So I tried OXA, OXA FM synth, and it sounds pretty nice. It sounds capable. It has a nice routing grid. It has various waveforms, so not only sine waves, which can give some cool options. So I thought, let's give it a go. And actually, I had some experience with it, but I'm not super experienced with it. So I'm going to be learning this synth as I go. So I just wanted to document my process of creating the track, my submission, my entry for this open source FM music synthesis challenge. So here's my Ardor session. I'm creating 16 tracks with OXA FM synth. I titled the track Bodacious very boldly. Hopefully it's going to live up to the name. Alrighty, let's assign a MIDI keyboard input. That's the default sound. Alright, I'm going to first try to figure it out and make a kick drum. So let's just make one bar. Oh dear, nope. Oh, the bar doesn't start at the start of the bar. Too bad. Let's just put four on the floor. Oh sweet, sounds nothing like a kick drum. Alright, let's figure this out. I need just one operator. That's it. Let's make it loud. And every one of them has pitch curve. What is this? Oh, it's portamento. Okay, operator F. I'm using operator F right now. Okay, keyboard tracking. Wave, sign. Can I have a frequency envelope? That's the question. There's an LFO. Sign LFO. Interesting. Things are moving when I am moving things. I'm going to disable all the operators. Oh, panning. Panning in the middle. Okay, double click resets it to default. Great. So it's F. Okay, so this is E influence on F. This is F to output. Okay. Also, this is pretty quiet. I'm going to put a limiter on the master. Limiter. Super quiet. Okay. How do I... Why does the pitch change when I move the knobs around? That's weird. Actually, when I attach something else than an operator. Well, that's nice. Okay, maybe I can use a different operator to make a pitch envelope for this one. For example, if I use operator E to influence operator F. Operator E is too high frequency. Oh, sweet. Yeah, let's make operator E decay to zero. Okay, this is kind of... Kind of doing it. I wonder what is the LFO used for and how can I actually source it? Oh, this is self modulation. All right. This screen is nice. It shows lots of things. Maybe I can understand stuff with LFO waveform rate, depth, delay. Destination. OPX. I see. So I can send a wake, but I think OPX is noise. Yeah. So we basically have a little bit of the noise would be... Oh, there's some settings for that. Cutoff. Okay, so we have a low pass. Oh, it's actually bundt pass. Hey, but isn't this the same thing as here? I think not. Okay, let's give it some very short. Yeah. OPX bypass. What is that? No idea. But hey, we have something like a kick drum. Well, let's see what we can do with external processing, I mean external. Other effects to make it cooler. It's distorted. I think it's a little bit too high pitched. Oh, this one is way too loud, I think. Hey, better. Maybe. Okay, operator F. Let's turn out the chorus. Maybe try this. I'm gonna add an EQ. EQ 10Q, yep. Everything I have here is open source, so it doesn't... Apart from the initial sound, it really changes for me. Which is really nice. Okay, I think I want to distort it after this happened. Maybe we'll tap tube warmth. Ah, yes. Ah, it's too loud, it's too loud. Compress it. Maybe we should just turn down our monitoring. Is it too loud for you? I think it's not. Oh, sweet. Okay, we have kind of the middle ground. Let's distort it again because distortion is great. I see. I think we're discovering some minor problems with maybe the quality of the processing in Oxy FM synth. Or maybe that's just the horrendous amount of distortion I had added. Three different processors. I think we'll need to add another equalizer. Or maybe just a little bit of reverb. I think I'm gonna go with guitar X, Zeta, reverb. Then let's compress it again. Finally, some again. Some 808. Okay, we have a kick. Now, what actually what I want to do? We have tempo 120 as the default one. I wanted to do something hectic. Let's go with 174. Let's try some drum bass, maybe. I don't know. We might change this later. We hit snare here. So this is gonna be the snare. You know, drum synthesis is one of the most difficult things. I think, like, decent drum synthesis. So once I get that done, I'm really not afraid about the rest too much. Oh, that's funny. It feels kind of dubby. Alrighty, again, turning off what I don't need. Turning up what I need. And I need noise. Now we'll try to make the operator E. Influence operator F again. Whoa. Whoa. Oh, I changed the operator F. Okay, operator E, sorry. Well, this sounds pretty decent. What is operator Z? Who knows? There's some info. Oh, it's another noises later. Is it? Should be. But this one has variable type. Let's disable everything else and try this. Why can't I hear it? Oh, I see. It's a freaking filter. Awesome. Is it? Low pass. Okay, let's disable. Put everything through the low pass. Yeah, sweet. Can I have a... I have an envelope. Sweet. This is nice. This is interesting. I like it. I'm not sure what's the pitch of this snare. Come on. EQ 10Q, where are you? Oh, there it is. Okay, could be good. Yeah, I have a little bit too little high end. Now some reverb. I could actually maybe use a burb. Make it small. So it just adds some ambience and feel of a room of a space. Now, of course, what else are we going to do? We're going to distort it. Pretty sweet. Maybe a bit too much. Yeah, we're going to compress it then. Accentuate this transient. But also make it sound bigger. You know, reverb on this doesn't work. At least not with this amount. This kick doesn't sound too good, actually. It's the first kick I ever made with Oxy FM synth, so probably I'm going to improve. How about putting this all through, or at least this one, through Z. Okay. I don't like this strange afternoise, but you know what? I'm going to maybe try and mitigate with a side-chain gate. Also, well, let's see if it's introduced by something else or it's in the source. Well, the source is pretty quiet. Is there a master volume somewhere here? I know. I'm probably... That's weird. Yeah, we need side-chain compression or gating. So, I'm going to use a side-chain gate. Not because I can introduce an additional input from the side, but because I have filtering options right here that I can use to make the gate only let the sound through when there are low frequencies available. Because the dirty stuff that I don't like is only in the high frequencies. So, let's go to listen mode and make a low gate. See? There is no bad stuff here. However, we want the click. Initial click. So, I need to do it before... Oh, that's what I wanted. Sweet. Awesome. It's not perfect yet, but it's getting there. Pitch it a little bit down. This is the first one for CQ. It's interesting. It sounds very acoustic. It does. Wow. Interesting. Okay, that is starting to sound like something useful. Let's save the session because that is very important. Oh, my. This is interesting. What I need now is a hi-hat. Actually, a couple of hi-hats. Let's go with hall-hat 1. Let's do a simple backbeat kind of thing. I know. I'm going to make it quarter. Make these two quieter. Oh, that's too fast. Stretch it. Yeah, that's great. Now, I can just consolidate this range to... Oh, something broke. Not super cool. Okay, we can leave it like that. Pride. Hi, Hattie. Ask me. Oh, that's pretty nice. I have no idea what I'm doing. Hi, kid. I have some idea. Oh, that's sweet. It's pretty crazy. It's pretty nice. This is velocity sensing. Okay, so that's nothing. How does the reverb sound in here? Oh, it has some... Okay. Maybe I'm going to try to push it through a filter, too. Oh, that's pretty sweet. Oh, man, this is the awesome goodness. So ridiculous. Okay, I'm going to try to run it all through the filter. So I need to route it to Z. Use activate Z. And then this is the filter. Am I right? Type. High pass. Okay, that's pretty nice. You have a nice click. Oh, this is really loud. Well, let's go with this. I would just go with an EQ. See what we have here. I tried to high pass it, but it wasn't super successful. This is a little bit too much. There's another one there. Pretty piercing. Overall, I could maybe kind of boost the high high. Ah, there's... I know, there's too little noise in this. I also use some more variety. Oh, control. The velocity. Sweet. We have a drum and bass-ish beat. Here's my coffee drinkable already. That's interesting. Okay, let's have fun. Let's have some fun with the beat. Oh, that's really fun. Can I make some ghost notes? Maybe I can make a ghost snare. Save it and duplicate my snare instrument. Let's make a new playlist. It's on the bottom. You see, I made myself quite a few tracks to work with. I'm going to insert some... Wait a minute. Should I make longer notes? Like it. Okay. Yeah, that's what I was talking about. But this is... I think we need to relieve the compression a lot on this one because otherwise it's going to sound like a turn-down sound, not like a quieter sound. Something strange is happening. Maybe it's due to the compression. Oh, I think it's due to the timing. Okay, that's the model consistent. Once it's consistent, I can use something like a KALF Transient Designer to add the spike in. I usually do this with a compressor, but because we have already messed up this sound pretty bad, I'm going to try to not do it with a compressor. Also, I'm still lacking some high-end. However, I think we have a trouble with some real bad spikes, so I'm going to add an Alpes filter to kind of destack the whole thing and maybe a DC-offstream mover also. And then a... Let's use a compressor, maybe. No, let's use Wolf Shaper. It's pretty loud, but too much. I'm going to try to tame it with compression again. Maybe not turn it up again. Let's use it like a limiter. Can we tame the peaks? All right, better. How about just using a saw-turner? Okay. I think... Yeah, that's very unrealistic. The noise is just... The noise is just... It just has no decay. It just goes forever. Okay, I have the operator X. Yeah, it has sustained it. That's more realistic, right? And now this doesn't have such an impact. I must say this kick is pretty weird. I'm not sure if it's good. I could try to mess it up more. How about actually adding some proper gain before all that stuff? Because I'm trying to make it louder. It's not super quiet, which is strange. Okay. It's pretty strange, if you ask me. I think it kind of works. Let's add a bus. So we have... We can mix our sweet drum kit into just one bus. It's going to be enough. Oh, I should have changed to insert it in the beginning. I just have to manually move it. All right, now let's assign it. The kick goes to drums. The snare goes to drums. Snare goes to drums. And hi-hat 1 goes to drums. Now... Yeah, the kick is definitely too loud still. And it's lacking... It's lacking a pitch slide. But I'm not sure how to make it. Let's save it. Hmm. Let's try from scratch. That's what I need. I need to disable the keyboard tracking. So now I have an arbitrary frequency. Which can do some weird stuff. Is it better? I dare you to make a kick drum out of the rack impulse. You know? I need to pick up my headphones. Because I might not be hearing the bass. As it is. I might be hearing it skewed. And maybe... Maybe it sounds entirely different, I think. Holy crap! Holy crap! How about disabling the limiter? Okay. That's a lot of bass. And a deep bass. This is heavy. Wow! Unfortunately, I think because of the gate we are still losing quite a lot of the transient. I actually can remove this stuff with this transient designer. I'm going to try and use something called Hard Clipper. Limiter. By Steve Harris. It's super nice because you can listen to the residue. You can listen to what you're actually clipping. Okay, I'm going to try and use compression one more time. Can you see the amount of plugins I'm using? To shape this kick into submission. To kick this kick. Let's hear it in the mix. Well, well, well. Let's save a new version. I'm not sure if we're making progress. Maybe I should leave it for now and focus on other sounds. Or maybe just call it a day. Leave it for now. We are just, you know, 15 minutes in. And do some other stuff. No, I just can't let go. This kick doesn't sound well. Let's maybe disable all the plugins. I don't know why doing AB doesn't disable everything at once. Let's listen from the source. Oh, that's a problem. I mean, it's not a problem, but really big spikes here. Scooped mids. I'm not sure if I like this curve anymore. Let's reset it. Or have another curve. Just compare them. It's a nice feature of EQ 10Q. You can have multiple curves. You can have two. Actually, you can switch between them. I actually want to make this shorter. Oh, I should do. I shouldn't use the sidechain gate. I should use the multi-band gate. I have used it to deal with noisy recordings of voice before, because we can basically do stuff like this. So we have the click, but we don't have the nasty stuff that goes after it. That's what we needed. Now compression. Lots of compression. We still have some of that. Let's see if we can... what's happening here. Oh, there it is. It's interesting because it changes when I switch between peak and run mass. Sweet. Multi-band gate is an amazing thing. See if we can clean up the sound in different ways. See if we can shorten. Sure we can. Why not? Change the ratio. Now it's a bit too short. I can bring some of that back just to leave it some character. Oh, that probably wasn't helping. Yeah, we don't need this reverb. Let's delete it. I think we don't need more compression, actually. That's the sailage for now. Oh, that's an overkill now. No, I don't like this at all. It sounds terrible. This sounds terrible too. Let's make a new curve and see. We have huge amounts of bass. Let's see if we can... There's something we're peaking. Oh, this is helping. The Alpes filter makes it sound entirely different. We are rearranging, like, you know, making some frequencies lag behind the other ones and we are, like, re-stacking our transient. I actually like this. I think the DC offset is no longer needed. It doesn't change anything, so let's delete it. That too. And this too. Oh, this is pretty sweet. Actually it doesn't do much. I think we need these two, either. Okay. I'm going to actually snapshot another version. Don't save now, just the snapshot. This isn't with the whole thing. Alright, it's a bit too clicky. I think we have to mitigate it with DQ. Yeah, that sounds better. Okay, we're almost hour in. I think we're going to leave it now. This is end of day one. Yeah. If you want to see more of that stuff, join me in the next episode as I'll be finishing this track learning oxyephemsins as I go, which is very unusual. Usually I show you stuff that I have already figured out. So if you want to learn stuff that I already figured out check out my other videos. Thanks for watching. I hope you learned something. And as always leave your comments in the comment section. Have your suggestions heard and if you want to see more stuff like that consider joining my patrons at patreon.com because these are the guys that allow me to take off time from my full-time job to do this. Share the knowledge. So yeah, thanks again. And I'll see you in the next video. Bye.