Lessons

Lesson Info

Compressorless Compression

So this next bit sort of taking it, taking a little bit of a break from actual. So this is concepts over its related trust me, but we're gonna start talking about compression and distortion and that kind of business, and we're going to begin by addressing sort of the basics of the basics. Caligari day with synthesis in general were started talking about the idea of what headroom is and what dynamic range is these ideas that are central to the concept of compressing them again? Which is what all we kind of have in mind with up. So to demonstrate that I have a vocal that's, actually, nfl studio. Um, if you go to the miscellaneous folder tab has all the files and audio samples and stuff that are special to specific dima projects that iran and most of you and this is the vocal file from aren't you clever. I forget the name of the woman who sings this, but let's have a look at it, maybe never made it tends to phyllis k with every kind of light. Nice. Now I used the limiter here to kind of l...

ook at the levels as as the party goes by and see very clearly how dynamic it is. This is pretty much a nice little snapshot of what does demonstrates what dynamic range is, um, I think actually, technically, the definition is the measurement between the lowest in the highest peaks, not necessarily the lowest south from the highest sound, because that would just be silenced with maximum and that's how it was that but most picked the highest peak in that kind of thinking could measure indeed be decibels what the dynamic range is now compressing and why we compress and the idea of loudness are all kind of related, and this this concept of many people have probably heard. I called the loudness wars, where people have tried very, very hard to make whatever they're doing, the loudest version of whatever it is they could possibly do that without sort of sacrificing quality, quote unquote, or sometimes by sacrificing quality clinic because they want to be louder, they have to be a better idea. Um, and that's totally fine, you know certain styles work with that certain styles dealt that's actually super duper important to remember when we can't, we can conceptualize the idea of when to compress something and how and for what reason. But before we get kind of to that, I want to talk about the idea of positions based basic mixing concepts and adding to think two signals together and the idea of head room so, for example, I was decision the vocal demonstrate that I grant you we'll talk about that when they get the actual compression, but for now let's go back to our sort of our sol wave sound idea that we had earlier and now I have it so they have a snare that I have pulled in still there was no point syria is there here you haven't e q with the snare because you see the spectral behavior of the snare and where and what frequency? What sound is happening about point? That kind of language is super duper useful information and again, look, look at our saw wave way forward base we can kind of see that this all wave piques the hell out of everything because it's very loud and solve that problem I'm going to put a limit on the master and I'm not gonna worry about that that's just there for utility that's better so we could see that takes up your pretty huge swath of the spectrum when you play a lot of notes takes up more that's all and look it's there we've snare wave again it's always there with we can see that if they said a lot but it has specific kind of emphasis on particular parts and in fact a difference between most sounds just sort of in general off other than like note content is pretty much that's harmonic makeup and I say most sound because a snares does every totals now sound first of solid, which is a perfect total sound like a saw waivers like a piano, for example, like it's got most of the same figures he's involved just have different ways in different moment change what that character, the sound, is very important understanding for sort of sound design. I understand that, but for what we're talking about right now is the idea that essentially way have conflict. We have frequencies that they're going to conflict and for mixing listen, super important because what's going to happen is that if we had we place since the song, right, we larry this to sounds of governments they're playing, and in the solid play the resulting level of all that together for this particular frequencies are going to be more because this favorite is happening doubling in the same spot than they were before basic mixing concept is just to go toe like this are wave and take out the area of the set, list their heads or would head if I pointed, you know that high and that right there is actually an example of compressing actually for real, because what we're doing now is we're reducing the amount of conflict between these two sounds and what that means is that we've reduced, um how loud everything is together, it doesn't sound that we've produced it, because that still sounds like it's a side of this, but that's the difference between sort of loudness and then the measurement of levels, they're not exactly the same, because, I mean, in the other day, when we're pushing everything I've already be, everything is that zero? T me, but there's, a difference between everything being insecurity be everything is going to be even something sounding loud, because injury of experiences yourself, and you've attempted this sort of massive around your tracks and make it yourself. This is a common problem, but right here is sort of one of the ways that we can get around that. And so I guess, that this is a result of essentially compressing something by reducing sort of the dynamic range, because with this conflict removed, if fits together, well, so it means the peaks are picking up his high anymore, because we reduce that conflict sets, but that haven't rate is produced so that you have to keep that in mind that, like compression and reduction of dynamic range, and controlling your head room all the same thing, and that an actual compressor is not the only thing that actually allows you to compress very important sort of thought experiment, because a lot of you are going to come in with the idea that ok I have to compress so have getting oppressor or that I have had this hundred problem that can only be solved with compression that so I won't try and only use compression to kind of deal with that as a problem and they might forget you know kind of mixing shenanigans but then um the question that comes up a lot when I do I take on students for when I want lessons and the question comes up about like when do I compress things and for two what ad like you know other than reducing dynamic range so that you could turn it up louder in front of the latter that's fine but like why is it that an individual sound or a voice even why is it better that it gets compressed? Why does that help let's go back to the vocal look at that for a second to answer this question maybe never put its own channel so that we could experiment on it cos maximus for this because it's got a similar interface tio devotes about interface to limiter you can see maybe never made it tends to fill the sky cool now that conflict and we took care of with the sol wave on the snare right uh a subtle way was kind of interesting sound when it comes to india diagrams and since in general very basic since like that because they will have dynamic range they don't they're already compressed there are one hundred percent breast they're flat you play the notes alive that's you know really needed the president further, but the trouble comes when we tried to mix it with something dynamic like a voice like this woman voice here because this woman's voices that is really everywhere but it is more places it's kind of around and this means that that conflict that exists doesn't exist as bad all the time or even in the same place and so it was not consistent level not consistently the same problem then a very simple solution we the one notch in this always was not going to cut it and so compressing sense signal maybe never let a terrorist phyllis k with every kind of right kind of helps to solve this problem by forcing is taken to be bored, consistent and louder. You might have noticed that even though it doesn't peak any higher than it did when it was this loudest, it sounds louder now and this is this is again what we were talking about earlier about the difference between things just kind of being absurd to be a little time with the idea something actually being loud and well it does is related to the ideas already be in a sense they were doing my sixteen year old right now, but the the reason why this now sounds louder is because it spends off whatever it's doing it spends more time closer to the limit your threshold whatever is that we said that it did um you know and that's just really hard to that and so instead of instead of having big peaks and valleys and house dynamic range whatever it's it stays mostly at the limit and it's consistent and this makes it super duper easy to mix whatever that is and also makes super duper easy to keep it consistently heard because at whatever level it is, we can listen to it with key into it and if it doesn't change, it was we were listening to a song then it doesn't need to like be pushed any harder for us to hear with clarity, which is important first basically making a general now I mentioned some things in there I said stuff like threshold, I don't think I said ratio, but if I did, you probably thought about it. Um maximus doesn't really have any of that and it's a big mystery for that right now, but we'll talk about what threshold ratio actually mean because if you've ever seen the compressor, you have probably seen these two parameters threshold and raise you know that when I'm using a limiter on the compressor side to see that when I moved threshold richer around increases photograph right there for a second that looks kind of like maximus is graff they re arresting one a minute ago because they're related but it's little bit borders thing that I heard today today, right? So what a compressor does to control dynamic range is that it essentially really, really fast volume automation and it's it's controlling just the peaks of things they could bring effect pickles above to hide by how you become the pressure on radio in the brain that will bring the volume down on the entire signal? So that peak is down there we're supposed to be and for you could make it go up and up and down or whatever and I kind of thing and you might be thinking that might be something called expansion, but what expansion at least I am pretty sure it means is that you are increasing the demand dynamic range in a sound versus not doing that. Um maximus makes it a little bit easier, but what was generally what you do is that you can press it down pretty hard and then you use makeup game to turn up again because you've turned up all the peaks down the average levels a bit lower, but then when you bring it all up, you kind of inflate everything and things that were once quiet are now a lot louder and things that were loud or quiet and as a result we have decreased dynamic range and increased loudness now first I was pretty straightforward. You just move it to put to where you want to be in any peaks of the air above it get pushed down according to the ratio, the ratio is a bit harder ever had around but simply what it means like when you see this as two to one ratio, what that means is that the signal would have to be twice as that is it originally was for its reach its official height I'm not in territory that means it is just bringing it down twice as much on actual thing that's true but there's some math there of that relates to that and if you could bring it down pretty hard by the maximum for this compressor is twenty to one so that means that this thing would have to be twenty times lives is now for it to restoring the original height which brings it pretty hard down but not fully that completely hard and for completely hard that's what limiters before limiters, he said threshold and then nothing ever goes above it at all and, uh actually like right compressing with the ratio of the threat that we have here is not designed to completely flattened. Everything is designed to kind of curve it a little bit to make it look like a little bit more usable but not to the story of a single I say destroy but I don't really mean that in a negative connotation it does change the bit but doesn't necessarily make it worse or better that's up to you to determine that kind of by your instabilities and limit limit just kills that keeps it flat forever according to the time variables which we see here being attack release and ahead attack and release are kind of really actually just release is sort of the important one and most compressors that you'll find they burst in the eighties stars have things for them the movement of it is our attack two k so stay in release I kind of think envelope controls for the volume on the nation that is reacting to the pressure the faster that reacts the more tightly adheres to individual peaks and the louder everything gets the slower reacts the smoother it it moves around and that sort of quiet everything gets um we can kind of see this in action and maximize a bit maybe never turn it way up maybe never and then bring it down hard limit like this maybe never so we could see the volumes that this is traditional kind of compression and where they were bringing everything down and that making of the game making me later you could see the volume on the curves that off the boil animation kind of keeping with the peaks and were doing it with pretty modest sort of accuracy if we zoom in maybe in that you see that again here is pretty well if I were to increase the speed dramatically, maybe now we can see that it's it looks more like uh someone's drawing a map of a state somewhere kind of wiggly being very detailed, sort of and we can reduce it even farther maybe never that was getting really detailed it's a hearing to almost the individual oscillation of the actual way for itself. And in fact, if I go even farther maybe never that's exactly what it's doing and just based on terminology that I just use you might be thinking wave shaping right now and that's actually what's happening if you have time variables in your compressor that are reacting faster than the oscillation of the salvage of using it's going to change the shape with away form, which is distortion what distortion does at least specific specifically in the way she having to starting that we use and nfl but then again, most distortion you don't just do that in the background and have a set in a particular way for the sound of particular way in that particular looking pretty cool stuff so that's why a time variables are important now I guess I kind of need to go over sort of specifically how what these knobs mean in maximus and the limiter because they do they say things, and they kind of don't mean what they say before we do that. Does anyone have any questions about anything? I said? We have a question from rift er wants to know about conflicts and frequencies looking at the spectrum analyzer at which level in decibels should we stop bothering about overlapping, or if it isn't a fixed level, which ratio between two sounds is the same range? So for that kind of stuff, that's not does not much of a there's, not much of lycan established amount of like dp you need to care about about four months out, another um when you mix stuff, the reason where we have to next thing is the reason why we have to bring down conflicts is not just because of the head room. The headroom helps that's the problem that we have to deal with sometimes, but you might have noticed that we, even while I was playing it wasn't really easy to pick out there was a conflict and that's, because honestly, there wasn't like that we weren't, we weren't limited anything we weren't clipping different, really ahead of problem, those two sounds were interacting with it to each other, and they were not damaging each other probably should have used a more damaging example, but usually with the end up happening is that you work on a sound, you have a local issue, have a kicker staring something, and then when you play it all together in a song, none of it sounds like a date, a man to go when I was all by itself, and the goal of mixing is that you want it, we want all the things to sound like they did individually when they're all together, and so does mean stuff like, you know, if you have a job kid that has a kicking us there and a high hat that the high hats typically will interact with stuff in the kind of four away, and it won't really sound like, oh, the hires or too much just in general, the sound like whatever one thing that there is suddenly has to blakemore higher frequencies in that, then it didn't make me go like that's always my my, my selling, those eyes too sharp, all the sun. So you rolled off the highs in the hallway, and then if you listen to stuff by itself, is that you kind of interesting if you listen to like a fully mixed finish track of your own, that if you go isolate with drums, you might realize the weight of them's on anything like the drums. Did, or even do in the mix, they might be thinking about that. That sounds bad, but how could anyone have arrived individually? A decision to make something sound like that answer is you don't. Those are decisions that you make when you listen to stuff when they're playing, and that decision, it is entirely up to your own listening sensibilities. There are no measurements, it's going to defer for everybody, and you just kind of got to do it until it sounds good.