Audio Mastering in Logic Pro X

1. Introduction and What you are Going to Learn: Hi, this is Christopher Caraglio, and in this class I'm going to teach you how to confidently create great sounding masters at home using logic. Pretend I'll be going in depth into some mastering concepts, such as creating a fixed monitoring level. Recommended practices for monitoring. Finding a lap of sweet spot using reference tracks using multi bands. Compression limiting, Andi Dynamic equalizing. I look forward to seeing you in the class. 2. Peak Headroom vs Dynamic Headroom : Hello and welcome to this lecture where we're gonna talk about headroom. Headroom is someone has talked about a lot in the audio community, especially when it comes to mastering on bears a lot of confusion as to what headroom is, how much is needed. So what is head from? Well, in my opinion, there are two different types of headroom. There is peak headroom on there's dynamic headroom. Paige Headroom is basically how much space there is between the maximum peak level of the signal and zero or digital full scale. The very, very top off the scale before it clips. Dynamic head from is more the difference between the peak level on the average level off the signal. So how dynamic it is, or how much difference there is between the ferry loudest part or the very quietest part, the more dynamic head from there is the more information that the mastering engineer has when he or she is trying to optimize your master for loudness equalization on dynamics, the less dynamic head from there is, the less options they have to optimize. Your master pick head from, however, doesn't really matter that much in digital as long as it's not clipping. So if you submit your master to a mastering engineer and it's peaking up minus one, peaking at minus two, minus six minus 12. It doesn't really matter, however, if it's peaking at zero. But some of that information went above zero and thus got clipped. Then that's an issue. Anything under zero is not a problem. Doesn't matter if it's minor six or minus. Any other number for that matter. Many mastering engineers advised to leave six decibels off Peak had rooming nor mixes before you send it to them for mastering. That is a good rule of thumb, but you don't have to aim for that specific number. They can turn it up, or they can turn it down via simple gain on their end. The reason that they advise this is just toe. Make sure that you think about your levels before you send it off for master, so it's a recap on that point. If you're mixing Andi, your final master fader is above zero and you're getting the red light. It means that some of that information is going above the sealing off the digital system on buses getting chopped off which causes often quite nasty distortion. So I've got a mix here in this project on I just like to demonstrate to you quickly how I'm measuring peak headroom and how are measuring dynamic headroom as the mastering engineer in this situation. So I'm just gonna show you now how I tend to measure dynamic headroom and peak head from when I'm first reference in the mix is that I get from a client. What I do is I go over to the stereo out bus and I load up as an audio effect. The logic pro level meter under the metering supping you, Andi I simply said it to peak and RMS Andi I simply play back the audio. I was young and old, the corn and pushed Smith four moving cannot. Okay, so comparing the peak and RMS levels here. So the peak level is the ferry top off the way form on the RMS level is basically on average off the whole way form level. So it's more in line with how we perceive the way form. But the peak is the very top off that shape now here is telling me that I've got a goods 12 ish decibels off dynamic headroom between the peak level here on the armistice average level here. This is also known as the Crest Factor. The difference between the peak and are a mess level around the 12 disa Belmar. I'm pretty happy that that is a mixed down with plenty off dynamic headroom for me to master with. However, if it was three or four decibels, which does happen that times, I wouldn't solely based my decision on going back to the mix engineer and saying, You need to adjust that just based on those numbers. But it might tell me it might give me something to look at and go, OK, is that a bit too crushed? Can I work with that? Or am I gonna have troubles when I'm trying to apply some e que compression or limiting in my mastering stage? However, for mixed at this, I've got plenty of peak headroom on. I've got plenty off dynamic headroom toe work with. So just to summarize, what we looked at here in this video is the difference between peak headroom and dynamic headroom on how to measure that. Using logic, pros level meter plug in 3. Compression - Part 1: Hi. In this video, we're going to talk about compression in this video. I'm going to use a lot. It's compressor, which is a fairy capable, very versatile compressor on will fulfill ordered the compression needs that you need in mastering. First of all, I'd like to make clear wearing the chain. The compressor should generally go. So as you can see here on my track, Inspector, I've got gain utility I e e que. And then a compressor that is generally speaking through the order in which I have those plug ins. If I used them in mastering, the reason for the initial gain is to increase or decrease the headroom that I might deem necessary when using compression. The e coup is because generally I want to balance the frequencies in the sound first before compress it. And sometimes I might even use an e que after compression to slightly adjust the way the compressor colored the sound. But generally I would definitely have one before, because if the ikar is unbalanced, I can't really expect the compressor to behave in the way that I expected to. Now I've looped a reasonably loud part off Thomas's mix to demonstrate it compressor. So I'm just gonna hit play and we're gonna have a look at what it's doing. And then I'll talk you through the different parameters that are on the compressor such as threshold ratio, makeup gain, etcetera. Okay, so the first thing that you see here on the compressor is this needle. Now this needle, what this represents is the amount off gain reduction that's happening. A compressor is practically just turning down or turning up the signal as it comes in. So the the amount of gain reduction is just how much it's turning the signal down by which brings me onto threshold. So the threshold is the level at which, when the audio exceeds it, gain reduction occurs. So, for example, for every Desa bell above the fresh old. So whenever the sound exceeds minus 20 decibels, game reduction applies. The amount of game reduction is determined by the ratio. So, for example, if the audio waas 10 decibels over the threshold, a ratio of 2 to 1, it would reduce it by five. So for every two decibels in over the first sold, only one disa bell over the threshold comes out. So it 2 to 1 is basically having the amount of decibels that exceed the first hold To give you an example of why it's so important to manage the amount of gain going into the compressor under demonstrate how the compressive behaves different game levels. Okay, So as I play back the audio now, I'm going to adjust to gain, which is before the compressor in the chain. Andi, we're going to see how it affects the behaviour of the compressor. See, as I turned down again, going in the compressor basically stops working because the inputs, the amount off gain going in is now underneath the first cold. If it's always underneath the fresh hold, the compressor does nothing. When the volume exceeds the first cold, then the compressor turns to signal down. That's what you're seeing here on the meter. The amount of game reduction before it was pumping around down to around minus five, which meant it was turning down the signal by five D bees. Now what I'm gonna do now is because I don't want so much gain reduction happening on my master. I went around one db off gain reduction happening. I have two options So I either increase the threshold so that the fresh oldest higher so that the input has to be louder for compression toe happen. Oh, I just simply turned down input gain like I just did. However, I'm gonna turn it back up again so that compression is happening. And then I'm gonna make some final adjustments. So I've just turned up the input gain until the needle waas pumping down to around one decibel reduction as a general guideline. That's what I recommend when in doubt, when in using the compressor in mastering it gets a lot more noticeable after one decibel of gain reduction. However, what sounds good is good. My main motivation for using compression in mastering when I do use compression in mastering is sometimes just to glue the sound together a little bit just to bring the peaks down ever so slightly, especially if I want that extra one db of gain going into my limiter to increase the loudness. Okay, so let's talk about first hold again. So what I do is increase the game going in months more again, increasing the gain means I've exceeded the first hold more, which means more gain reduction, but maybe I want that amount of gain. I just want to reduce the amount of game reduction happening. So what I do is increase the threshold by increasing the threshold. I've put the ceiling at which compression happens a bit higher in the scale, so this compresses quite good. I'm doing a good three dp off game reduction on this master right now, and it's sounding really, really punchy now. The reason that it's sounding punchy is not necessarily because it's during three D B's of gain reduction, but also because it has a nice open attack, which I'll come back to later in the video. From here, I might want to find tune the way that the compressor behaves by adjusting the ratio. So, like I said before at 2 to 1 ratio, which is generally where I like to start for every two db over the threshold, it reduces it toe one db over the threshold. But I'm gonna make some fine adjustments now, just using my ears just to get to the sound that one. So I just lightly turned down the ratio to about 1.6 to 1. It's just compressing less so now only 1.6 debates need to go over the first hold for one db to just come out and the closer I get to a 1 to 1, the closer I get to zero reduction or zero compression, as you could see when I bring the ratio or the way down to 1.0 to 1, that is no compression at that ratio. Nothing is happening because it says for everyone db over the threshold one comes out, so that's no difference. So as you can see, there are already three things that determine the amount off gain reduction or the amount of compression that happens. That is input gain, which you can also adjust with logics built in gain stage here. That's no different to use in again stage before it. The only difference in this example is that there's an e que between them, because that's how I had it set up. But you can also decrease or increase gain at the beginning of the compressor. There, the threshold is the level at which game receptions starts to happen. The ratio is effectively, how much relative game reduction happens, and now we're gonna move on some makeup gain. The purpose off makeup gain is for us to matched the loudness off, processed on un processed. The purpose off. That is so that when we comparing the signal with the compressive bypassed on the signal with the compressor un bypassed, our is unfold into believing that the one that is just louder is better. We used the makeup gain to match the loudness off the to when we're making that a B comparison so we can be objective about the way that the compressor is handling this out. Generally, I like to turn off the auto gain, so it's not doing it for me so that I can adjust the makeup gain myself, as I suspected with the compressor on the signal was a little bit quieter, obviously, because it's actually turning it down a bit. So what gonna do is I'm gonna compensate for that with one db, maybe 1.5 db off makeup gain so that when I'm toggling the bypass when I'm making that a big comparison, my ears feel like it's exactly the same loudness. And the only remaining difference is the sound that the compressor is making. Okay, so now I feel they are both the same loudness. The only difference I'm hearing now is that the compressed sound that's just slightly more together and simple is that that is best practice for using makeup gain when using compression in mastering in the next video, we're gonna talk more about compression. More specifically, about knee attack on release, see in the next video. 4. Compression - Part 2: welcome back to part two off compression in Mastering in Logic Pro. In the last video, we talked about input gain threshold ratio on makeup game, and now we're gonna talk about knee attack and release, as I mentioned in the last video in this example, even with quite heavy game reduction, like 3 to 4 decibels, the sound was very, very punchy and quite transparent. And that's mostly due to the attack and release times that's set on this compressor setting . The attack is how quickly full gain reduction occurs, so when the fish hold is exceeded by the signal, the attack is how quickly four gain reduction occurs. So, for example, if the input crossed the threshold by three decibels on the attack, time was set to 15 milliseconds. The attack time is how quickly it goes from zero game reduction to full gain reduction. More open attack times are generally quite good for preserving the transients, especially when you're mixing, and you were trying to use compression to glue the drums together. For example, the attack time allows those transients to pass through whilst the compressor compresses the body off the sound. This also is very useful in mastering for trying to glue the body of the master together. Waas. Allowing the transients to pass through, mostly unaffected the release time is how quickly or how slowly the gain reduction stops or goes to zero game reduction. As Theo Import produces under the threshold, the release is practically how long it takes for compression to stop occurring after the import has fallen back under the threshold. Logics compressor also has an auto release function, which attempts to make those release decisions for you. However I quite like often too, especially with E. D M music to assign my own release time based on the general beats or the general rhythm off the sand that I'm mastering, they show you an example. City also release right now is doing a very, very good job off, bringing the reduction back up to zero just in time for the next beat, just in time for the next kick drum to occur. If I were to do it without the auto release order as follows, see how game reduction returns to zero a bit faster now when I bring it down to five many seconds. However, now you see that as I really slow down the release to five full seconds. The needle is basically not returning to zero, because by the time that the compressor has triggered again, the release is only happens like a really, really tiny bit. So I want to bring it down to just about when the needle returns to zero. When I'm compressing a song, not descend, mastering now I'm going to finally adjust the attack time just so that it gives me the just the right amount of punch whilst compressing just the right amount of body that I wanted to have a slightly reduced attack. That transient is coming through just a bit harder, bit snappier, just like I wanted it to. The last thing I like to talk about in this part of the compression in logic Pro 10 for mastering Siri's is knee. So when I was talking about threshold before the first Cold is basically at what level does game reduction start to occur, or at least at what level and above? Does game reduction start to occur? Nay basically widens the value at which game reduction starts to occur. So instead of being like oh, at minus 15 and above. Game reduction occurs on, and when it's below minus 15 Nover game reduction occurs. The knee basically softens that curve, so we have some knee. It slowly starts the happen below the threshold and fastens over the threshold. So with a knee at like one right now, the first cold is no longer at minus 17 even though it says is at minus 17. It's actually a bit underneath minus 17 but it's No. Four game reduction into it passes minus 17. So the knee kind of softens the point on the scale at which game reduction starts to occur . Sometimes it's very difficult to here the audible difference as a result to this. But I suggest that after you do all of the moves I've already done, like adjusting the fresh hold, adjusting the ratio, attack and release, you just adjust the knee until you find a sweet spot in the sound. Somewhere. By bringing the knee back up to 0.5, I found that the sound just a bit punchier than it was at one, probably because compression started happening a lot further below the threshold at one than it does at 0.5 thanks for watching and in the next video and I talk about the different compressor models that are available to you in logic pros compressor. 5. Compression - Part 3: Hi. Welcome back in this video we're gonna talk about the different types off compressor models that logic pros compressor emulates. Right now, we're using the default platinum Digital model of compressor in logic Pro. I'll be using the same example that I was compressing in the previous videos. But what to do now is just flick through the different compressor models on. We're gonna notes how they change the behavior of the compressor and generally, how they slightly color the sound a bit differently. So that's platinum digital. Generally, this is the most transparent type of compression that we have available to us in logic. Pros compressor. When I say transparent, it means with minimal color and minimal distortion. It does the job without having too much of an effect on the general tone. The general sound off the audio. The second option along the row at the top here of different compressor models is studio V C, A V C A stands for voltage controlled amplifier on knees. Models can respond slowly or quickly to incoming transients, and they tend to have a ferry clean tone. So let's note the difference between the two in terms of General work flow. I like to do all of my settings using the platinum digital and just get into the right amount of compression. And then I like to flick through the different models that are available to me just to see what over colors and what other type of compressions hands are available to me after I've done all my basic settings. So this is studio V C. A. And I'm gonna flick between that platinum digital just to see how it changes the sound it. Okay, well, I'm definitely getting a fuller sound from the fault is controlled amplifier. I can't explain exactly why that is, because thes model to sort analog behaviors off those compresses, but I generally suppose that is coloring the sound on some way based on the Wadleigh emulates the circuitry off that compressor. Let's move on to studio F E T. Which stands for filled effect transistor. And once again, I'm gonna compared that with platinum Digital because platinum digital is kind of the default in the most transparent. So his pardon and digital again okay, so not a huge difference in sound there. But I do note the way that it's slightly differs in a way that it behaves with transients. I'm feeling a little bit off crunch when it comes to using the studio Every team model. These are no followed fast, transient response, clean or colored tone, particularly in the mid range, which I think is what I'm hearing right now and could be pushed to a crunchy tone, which is definitely what I'm hearing. So again, I'm just gonna a be between platinum digital and studio fbt Definitely bit cruncher has a bit more of a knock to the sound, which I quite like. There are other variants off the V C A models and the F E T models available to us in logic press compressor conflict frutos Now that we've explained by fee C A and F e t in its most basic sense. Okay, so now I'm probably gonna leave it on vintage V c a r quite like the way it hardens the sound off this master, so I'm gonna leave it at that. But before I move on the last off the models that I'd like to introduce you to is vintage Opto. So vintage Opto models that often optical compressor, which are also known for their fast, transient response fairy clean sound on offering, used as limiting amplifiers on buses on outputs not much happening here, basically just because of the way that it behaves differently. Like I said, it's used generally as a limiting amplifier or across buses and that sort of thing. If you like to learn more about the difference off analog models air compressor that are available, I suggest you look into old Andi current analog compressor models such as 11 76 and L. A to A and see how they differ in their circuitry, such as voltage controlled, amplifier filled effect transistor and how they are commonly used in various contexts. In mixing and mastering. However, Logic Pro has most of that sort of sound available to us on. Like I said, our to get my basic compression. What done? So it's doing the job. And then I like to cycle food that different models are available to us just to see what different sounds I can attain. Thanks for watching in the next part of the compression Siri's. We're going to talk about side chain distortion on parallel processing 6. Compression - Part 4: welcome back in this part of the mastering with compression Siri's. We're going to talk about the in built limiter, distortion, parallel processing and site training. So let's start with side chaining what is side training? So in the compression examples that I've given you already, the compression that happens is relative to the amount off signal going in on the signal going out. Is the affected version off the signal going in with Cy Chain? However, the signal that triggers the compression can be different to the signal that the compression effects in a mastering context that can be quite useful. So instead of using the same signal as the trigger, we're going to use a filtered version off the signal as the trigger. Sometimes I do this because I want to compress the sound. But when the kick drum is so loud on its triggering the compressor, it could be to audible and not very transparent because the kick is happening and everything else is getting pushed down when the kick is happening. So I'm gonna use the side chain to use a version off the signal. It's a trick of the compressor that has filtered out the kick drum. So let's get started. So I've just flipped over to side chain here on. What I'm gonna do is I'm gonna listen to the filter, so it allows me to listen to the side chains signal select the and I'm going to assign the filter to a high pass filter. And what I've done is I've landed the cut off frequency just above where most off the energy off the kick is coming from now that the signal that triggered the compressor is not so kick heavy. The way that it's gonna compress the sound is gonna be a bit more balanced, a bit more universal. There's gonna be less pump coming from the kick time. That's trigger in the compressor. So now I've done that. I no longer have to listen to the field so I could just turn it upon because and now, with the filter on, the compression is a little less sensitive to the kick. Even though in this example is always gonna be very sensitive to the kick. As you can see here as I flick from filter on to filter off, you can hear it, but you can also see it with the filter off. There's more game reduction happening when the kick is happening because there's no filter . Now we're just getting a punch. Your sound, because the compression is getting less pumped from the kick. That's one very practical and quite commonplace example of using side chain compression when master. One thing that you should make yourself aware off when using side chain in logic press compressor is the detector or the detection. We have four different options here. Max Some Peak and RMS, and you choose one from each road so you can have met speak Mets ROMs, some peak, and some are mists. Let's start with Max and some. So in this context, we using stereo compression, which means you've got the Left Channel on the right channel. When it compresses, detection is set some max. It triggers according to the maximum level from either left or right channels. So if the left channel happens at minus two on the right, channel happens at minus eight. Minus two is the maximum in this example, so it pushes both channels down the same amount. However, when you through some, the detection is based on the some level off left and right. So instead of going okay, what's the maximum level from either left or right? It's what is the level off left and right combined on applies compression based from that level. When it comes to Peak and RMS, however, the peak is the level off the very top off the way form. Whereas the RMS is the average level off the way form over a short period of time are a mess is a bit more realistic in terms of how we perceive sounds we listened to sounds quite slowly. We hear entire cycles, as opposed to the very top or bottom parts of the cycle are generally stick to RMS. But every so often you might want to go over to peek just to see how it behaves. But you're generally find that at peak because those levels are higher, the compression is going to be a bit more intense. So moving on from side chain right now, I'm gonna flick over toe output. How much you're gonna turn off the filter, click over but output. Now we're gonna explore the integrated limiter, So logic pros compressor has an integrated limiter. Personally, I don't really really use it all. I use the adaptive limiter in logic to create my limited signal personally in mastering I don't really use. The integrated limiter, however, is there if you need to use it, but it's very simple. After the compression, it limits the peak level. So if I set the first hold 20 and I turned the limiter on, it won't allow any peaks above Syria. A limiter is effectively a very fast compression with a ratio about infinity toe one or something really, really high tow one, which basically means that nothing will exceed the first cold. So if I switched a limiter on and set the threshold to zero, that means no peaks are going to go over zero. Now. If I turned official down to something like minus six, no peak values will exceed minus six, meaning the output off this compressor will not exceed minus six peak level. Let's see how that affects the sound. You see the red led flashing it's flashing whenever the fresh hold off. The limiters exceeded, however, because the limiters on at the threshold it caps the peak level at that fresh old, so no value now is exceeding minus 10 coming out off this compressor, as you can see in this meter here on in the meter off the channel in logic, start having a drastic effect on the sound right now because the amount of game reduction that the limiters during is very infrequent and minimal. However, if the level coming out of the compressor is much higher on the threshold around the same level, Kenmore game reductions have to occur that you would have a larger effect on the sound. Again, I don't really use the limiter in logics compress open mastering because I normally have my limiter last or second to last in my chain. If it has an effect on the sound that you like them, by all means, leave it on. So now let's move on to distortion. So Logics compressor has a built in saturation unit or built in distortion unit. It gives us three different types off distortion, so distortion generally refers to the distorting or the changing off the way form in a way that colors to sound in the way that creates new harmonic content. So I think oven e que as a way of balancing the harmonic content off the signal distortion is a means of creating new harmonic content in the signal. Three different types are available to us in logic. Pros compressor soft, hard and clip. They generally do what they say on the tin. So soft is a soft sort. Saturation Hard is a more audible, aggressive distortion, and clipping emulates that when you are, for example, clipping a converter, which basically means that you are soaring off the top of the way form. So if the way from is a sign waving your sword it off, you're creating a square wave, which creates new harmonic content as generally quite harsh but can be the sort of effect that you're going for. Let's have a listen to the different types of distortion that available to us in largest compressor. Now you shouldn't be surprised every so often to find that distortion can make things a bit louder. That is true, because by changing the way form but without increasing or decreasing gain, it creates new harmonic content without affecting the peak level, so it's generally quite useful for creating extra loudness. However, you are distorting the signal, so there's generally a trade off. But if you find a sort of sweet spot. You find a middle ground between distortion and loudness. Then you can get to a place that you're quite happy with a soft setting on a distortion. Here on large, it's compressor. Generally first, who's some sort of saturation is color in the sound in some way that is increasing the perceived loudness without actually changing the level, so you might find that quite useful when mastering. But they all colored the sounded different ways, so they're gonna be each better for different types of signal. The last thing I want to show you in this part of the compression Siri's is the mixed up. So basically, this is the type of parallel processing. Parallel processing is where you take the signal unaffected, and you take the signal affected and your Sundin together, or you mix them together or you blend them together. This allows us to do it without the legwork off, creating another track and mixing compressed and uncompressed together. This kind of takes care of it for us. For example, if I turn the dial the way to the left that is now purely input, so we're only hearing what it would be without compression, return it all the way to the right. It's with compression. However, you might find it useful after finding all the right settings to just dial back a little bit sometimes, so that your blending the affected against the unaffected just to dial it back a little bit or just find a sweet middle. The mixed Nabhan logic compresses just allows us to dial it back a little bit, explore some really extreme settings and then just kind of mix it in against the unaffected input that was part for off the compression Siris in mastering for logic, pretend on, I'll see you in the next video. 7. Limiting: Hi. In this video, we're gonna talk about limiting. Limiting is the process of limiting the peak level. The importance off limiting the peak level in mastering is to stop clipping. Clipping is when the signal exceeds headroom off the system that it's played back on, and it thus soars off the top of the way form. And that creates distortion to demonstrate this and to demonstrate the ramifications off using, limiting on the ramifications off, not using limiting. I'm going to loop back off the louder parts off this song, and I'm going to increase the game until the song is at a respectable loudness and see what happens when I do that without the aid of a limiter. So let's say, for example, that the right loudness for this song is around minus 10 or minus nine. L u f s momentary. What I'm gonna do is I'm gonna load the game, and I don't increase it until it's that that loudness and see what happens. So on the stereo out bust, I'm now gonna open logics loudness meter. So I go to the stereo out bus. Here I load up the audio effects menu. It's a metering God's allowed this meter. If you haven't done so already, feel free to check out the video. Well, talk about L U F s L U F s is the unit on the scale which the loudness meter in logic uses to measure loudness. However, don't worry too much if you haven't watched the Professor video already. We're just going to use this as an example. So let's say I'm aiming for the momentary column to reach around minus nine. L u F s. So I'm just gonna play it back. And as I play it back, I'm gonna increase the gain until I achieve that number. Okay, some around there already on that is a very typical momentary Luo fest level. At this point, in a piece of music like this, however, we have a problem because I didn't have a limiter in place. Clipping has happened. How do I know clipping has happened because of this red light over my channel meter? Right now it's his 3.1, which means the level off the signal has exceeded zero db, otherwise known as four scale the ferry top off the digital ceiling by 3.1 decibels. You might be able to hear this already. But if I export this out now as a wife file, those 3.1 DP's over zero will be lost. They'll be chopped off the way, form and distortion worker as a result of during that limiters. I like really, really fast compresses. But instead of having a ratio where the signal above the threshold is attenuated or turned down, the limiter basically stops the signal exceeding the first gold at all to demonstrate that I'm going to use logic, pros, adaptive limiter, which is a limited, the are use in mastering. And of course, I'm gonna have that After my gain utility, someone's going to click on the space underneath gain going to go to the dynamics menu, Adaptive limiter, stereo as this is a stereo master, I'm gonna turn down the gang so I don't want any extra gain because I'm happy where the game is, what now? But I just want to limit the ceiling so it doesn't clip. So I've got this set zero now, not just click. Apply on optimal look ahead, which I always do. A nectar simply optimizes the look ahead. The look ahead being how far into the future It looks at the signal so that it can limit the peak level as transparently as possible. So I'm just gonna play that back now. First of all, I'm going to reset the clipping led there. Onda. Um, now I'm gonna play it back and see what the new peak level is. Great. So as you can see, it's done. It's job. I'm not the same loudness. But before I was clipping and now I'm no. And as you see as I talk with the bypass there, the loudness is basically the same. But the peak level is drastically different in the middle column. Here, you see reduction. The reduction column is how much it's reducing the peak level to make sure it doesn't exceed the ceiling, which is zero because it's a limiter, is working so quickly to reduce that peak level that you can't hear its effect on the loudness of the signal because the way we perceive loudness is very much an averaging off levels over time, as opposed to peak level. The peak level between two signals come very drastically, but the loudness could remain the same. That's why generally speaking, peak level is not a good measure of relative loudness, which is why units such as El your office and RMS serve to better predict the loudness off the signal. As mentioned in the true peak video, I generally use true peak detection on on the adaptive limiter on my cap it at minus one. Again, I explain that in the video where talk about true peak levels. So in this example, where I've got plus nine db of gain here, I would bring that down to zero on, instead increased again nine db here the effect is the same. However, if I had something between my limiter on my gain, I wouldn't want the import off that plug in tow have over nine db off signal going in, especially if it's a compressor. I wouldn't want the nine db to be over the threshold off that compressor, so it is better to use the gain within the limiter as opposed to driving again going into it. That's basically how and why you'll use a limiter in the master process. Thanks for watching. I'll see you in the next video 8. Fixed Monitoring Level: Hello and welcome to this lecture. In this lecture, we're gonna talk about how to calibrate your monitoring at home so you could make accurate decisions when you're mastering your own music. The method I'm gonna take you through today is using Spotify. The reason I'm using Spotify is because everything that you play back in the Spotify library has been loudness managed, which means that Spotify has looked at the loudness of all of the songs and balanced amounts and turn some up and turn some down so roughly they should all be playing back at the same level. We're just gonna double check that that setting is there and on. So it's a calibrate. Your monitoring means to just set it up in a way so that it's always playing back a consistent volume whether you're using your speakers of your headphones, which allows you to accurately measure the relative loudness off what you're working on against other things. The reason we used in Spotify is because Spotify is one of the top music platforms that consumers used to listen to music daily, which is why they've level matched all of the material that's on there on and If you calibrates a Spotify, your masters will quite naturally end up in a place that is optimal for playback there. So before we do that, we need to make sure that your Spotify app is actually managing the loudness off the music or what they call audio normalization. In order to do that, we simply go to the settings and we scroll down to show advanced settings. Aunt here, where it says set the same volume level for all songs. Make sure that's on. So what Spotify have done is everything that goes to Spotify. They analyzed the audio Andi. They decide whether that song is louder than their target level, or if it's quieter than their target level, and they may turn your song up or down to get it to that target level. Like I said before, the reason we uses Spotify to do this is because it's one of the major platforms for consuming music today. So what advise you do is you create a playlist full of your favorite songs that are available on Spotify and to spend some time, listen to those songs, shuffling through the playlist on adjusting the output volume on your audio interface, which is connected to your speakers or your built in output like this or the headphone AMP that you might be using on your headphones. Spend some time, listen to those songs and spend some time optimizing that output level and get it to a playback loudness. That is loud enough that you can hear everything but not too loud that you can't work at that level for extended periods of times for like 3 to 4 hours. Maybe if it's too loud, your ears will fatigue over time. And that won't help you make good decisions in mastering. But if it's not loud enough, you're not necessarily going to hear all of the detail on. Also, if it isn't loud enough, you might be incentivized to over compress or overcook your master just to get it to a loud enough level. So, as I said before, I'm simply gonna open a playlist here. I'm gonna play some music, and I'm just gonna play different music that's in my play. This and I'm also making sure that the music that see my players has a large variety of styles, but more importantly, it's music that I know it's music that I love. Andi. It's music that I know what loudness should be. As you can see here in this example, I'm using Thebe built in Mac desktop control to adjust the level. Normally, you would do this using the output gain up or the master knob on your audio interface. And then once you're done and once you found that comfortable level, you just stick to it and you don't change it. That issue of mastering level. So once you've got there, just remember what those settings are. Remember where on the clock that dialogue was if used a dial on your audio interface? Andi, just leave it there and that is your master in level on over time. Actually, listen to music at your master and level as your master music at your master level, your ears will be in a much better place to measure dynamics. Measure E que on measure loudness and drove rely less on metering to tell you what those things are. Thank you for watching this lecture. Hope that's help you understand the benefits off work into a fixed monitoring level 9. Finding the Loudness Sweet Spot: Hello and welcome to this lecture. This lecture is on finding the loudness sweet spot. So what I mean by loudness sweet spot is, and this is unique for every song that you work on is about finding the level at which it's as loud as the material allows it to be before any audio degradation occurs. Obviously, what happens when you go beyond the loudness sweet spot or beyond the loudness potential off the music, you start to audibly mess with the dynamics things that the transients won't be as impact for. And it would just sound sort of dense and fatiguing to listen to when it's pushed that little bit too far. This is also one of the benefits of using a fixed monitor and level because over time, as you listen to music at fits monitor level, you've internalized what loudness sounds like. This means that you can just bring in a mix into your project session and then start turning it up until your ears tell you it's the right loudness. Allow me to demonstrate. So right here we have a mix with plenty off peak headroom. As you can see the peak off the way form is nowhere near the top. So the very first thing that we're actually going to do to turn it up in the most basic way of turning up any audio signal is using gain. So I opened the plaguing menu, go to utility and open gain gain is simply turning the signal up or turning a signal down its linear meaning that it doesn't change the dynamics off the signal. If you turn it down and then turn it up again, it's exactly the same audio. So all I'm going to do now is play back the song and just turn it up until I feel it. Is that the right loudness? The reason I'm using a game utility at the beginning of my chain here is because I might need to create extra headroom for any processing that happens after that gain. And also the channel Fader will only allow me to increase up to six decibels off gain. So that's why I choose to use the gain plug in as the ferry first instance on my channel. So I'm simply gonna go to the chorus here, and I'm gonna keep turning it up until it's at my normal mastering level. And I'm doing this entirely by ear. I don't need the support of meters to do this. Whatever way. While dreams I was a young, you know, Hold the corn and push Smith four and you can move and cannot like a day without most so whatever. Okay, so I've turned it up Some, However, I do feel like I want to push it a little bit louder. However, I have a slight issue where I have only 0.2 desa bells left before zero before clipping. This is why we use limiters. Limiters allow us to increase gain or further increase gain whilst keeping the peak level at zero or at any other designated number. This is the primary purpose for using a limiter now, because the gain isn't Allow me to go any further without clipping. I'm really use a limiter. So I just open the plucking menu and I go to dynamics are true adaptive limiter. And first of all, how much you could turn down the three db gain that it gives me there. Andi, I'm making sure as well that the adaptive limiter is of course, after that initial gain stage on my chain. These plug ins are, of course, opened in stereo because this is a stereo track. So now I'm going to increase gain going into my limiter until I feel that. Is that the right loudness? According to my ears, I can use the gain in the limiter itself or can use the game before it. There is no difference Limiters. They only apply game reduction when the actual input level hits the ceiling off the limiter . That's when gain reduction actually applies. We could always be a part of each other we could ever I would always being you, my treasure. We could roll miss enough. One another ever, Ever. Reeses heaped high upon the pyre. The script is sold. Just your little memories. Okay? My ears are happy with the loudness off the music right now. Andi, I don't hear any distortion artifacts. I don't hear any pumping from the limiter or anything like that. I'm very happy with the way that it's sounding. However, if there were any inconsistent artifacts like the odd distortion here in there as opposed to a consistent distortion, I would work back into the chain and look into things like compression and eq you to try and manage those sounds before they hit the limiter seeding. So again I've reached what I feel is the right loudness for the material without any audio degradation. Notes also that even though I increased the gain, my peak output level remained at zero. Because that's the That's what the limiter does. It stops the peak level exceeding the output ceiling, which is set to zero at this point. And as you can see here, there was 1.5 decibels off gain reduction, so gain reduction is basically when, for example, the input goes 1.5 db above the ceiling. That limiter is turning down the level by 1.5 so it doesn't exceed zero, and it happens very, very quickly. It happens fast enough that your ears can't hear it by the average level off the material goes up without clipping. So this is how you find the right loudness or the loudness sweet spot off the music that you're working on. Thank you for watching 10. Reference Tracks: Hi. In this video, we're going to talk about how to use reference tracks in your mastering session. A reference track is a song that you used to refer to when you are mastering your own music to give you a guideline off what your master might sound like with regard to e coup dynamics and possibly even loudness. So, first of all, I'm just going to set up a basic master in session where we have one track that we master on on another track that has our reference tracker. So starting a new project, I'm just gonna select two audio tracks and I'm gonna see men. I'm gonna call this track mall stirring, and I'm going to call this track reference. So to import my reference track, I click on my reference track header. I got to file import audiophile. Then I locate the father I want to use. This is ah, on Thomas George's songs that he mastered and released. I'm going to import that now added audio file contains temporal information. Don't import. I don't need that temporary information. I'm just gonna seem out now as this is a fully mastered song. It's going to be at the loudness that it was master too. But that's not necessarily the loudness that we should be referring it to, because before we start work or early in the master and process, we might not get that now that we might not ever get that lowers because it might not be the right loudness for what we're working on. However, if we don't do anything about that, if we don't do anything about the difference in loudness between a reference track and our mastering track, then we're always gonna feel a bit inadequate about what we're working on because it's always going to be a bit quieter. This is why is really, really important that we turn our reference tracks down to our mastering level. If you haven't watched that video or you don't see yourself using the fixed monitoring method, then what I suggest you do is you take your reference track and you turn it down until the loudest parts are around minus 12. L u f s. Let me show you how to do that. So I'm just gonna bring the play head to what is likely the loudest partner song, which is quite clearly around here, Andi, on the stereo out bus. I'm going to load on the metering loudness meter And now I'm just gonna play. But so right now it's what some might say commercially loud enough, but probably too loud for us to start our mastering session with. Anyway, it was hovering around minus nine or minus 10. Sort, momentary and short term Luo fs. I'm gonna turn it down by a few decibels just to bring it down to a more acceptable monitor level. And to do that, I simply go to the track, load a gain, plug in on the track true stereo and turn it down from here. I could also just turn it down using the fader. Okay, that's more acceptable level toe workout. Um, and now what I'm gonna do is bring in the track that I'm going to master Andi. I might, from their further compensate for any loudness difference when I'm referring back and forth between the reference track and the song that working up. So it's gonna close these for now and import the song to master file imports audiophile. Locate the audio file that once master, I click open and here we go I should concede, as still quite significant loudness difference between the arm mastered Onda mastered even though I've turned down knee mastered. So I'm gonna make some adjustments to the level off the unmasked erred just to bring it up close to trial, mastering level and possibly even some reduction. Further reduction on the reference track just said I can, compared to two fairly to do that. First of all, I meet one of them, so I'm gonna meet the reference track first. I know I'm gonna start playback on the UNM mastered track when we come over here because it looks like one of the louder parts I'm a credit cycle region to leak that. Fortunately, the loud a part of this song visually I could be wrong is roughly in line with allowed a part of that song as well. A quick trick when you're using reference tracks in logic Pro is when you've got one muted and you hold the key or the option key and you click on the UN muted truck It mutes that one on UNM utes that one. So I do that again for this one. So hold Olt option. Click on the un muted meets button and it switches over, so it makes it really easy for us to compare the tracks. Okay, so the master track is still quite a lot louder. But there's so much headroom in the unmasked attract enough headroom for me to start turning it up until they're both roughly at the same level. So I simply load again. That's the first plug in. Go to utility, Got to gain stereo, and I'm gonna turn that up until they're both roughly the same loudness to my ear. Great. So now they're actually roughly the same loudness. The good thing about that is I can just focus on balance indie e que and referring to the reference tracks E Q. By ear. When I'm working on the track that I'm mastering, it means that I won't feel insecure or rent B Ford by the extra loudness off the master track simply because I've turned it down to my mastering level. Simple. Is that so? When you used in reference tracks, make sure that you turned them down to your mastering level. If you have one or just a respectable level, like minus 16 or minus 14 L U f s and you can simply measure that. Using logic pros, Loudness meter. Thanks for watching. See you in the next video. 11. Loudness Normalization: welcome to this lecture. In this video, I'm going to talk about loudness. Normalisation. There's a lot of information out there about loudness, normalisation, and it's a very recent development in today's streaming world. So what is loudness? Normalisation? Well, loudness. Normalisation is the process off making a whole bunch of tracks the same loudness or similar loudness, so that when you're listening to them, you don't have to adjust the volume so much? One of the number one complaints from users about streaming platforms that they're forever having to adjust their output volume when it goes from one song to another one video to another, which you may have already experienced on YouTube or Spotify origins. Previously. In recent times, however, those platforms have started to implement loudness normalisation, where they measure the relative loudness off the content that's on their platform and makes adjustments on each of those pieces of content so that they all seem balanced. That's loudness normalisation. So you may have heard of something called the Loudness wars. The loudness was is something that happened over a long period of time, slowly but surely on. Basically, it was where everyone was competing to have their song at least as loud as commercial releases that were out there. The problem with that is every recorded medium, such as vinyl tape CD, has a ceiling. So everyone was compressing their music heavily, limiting their music just to push the average level up but destroying the dynamics off their music. At the same time. However, now that we have loudness, normalisation and things that are hyper loud or hyper compressed are getting turned down just to match, everything else that's on the library has practically mitigated, or at least started to mitigate the motivation for trying to get hyper loud. The good thing about this is others. Producers mixes and mastering engineers canal just once again, as we always should have done focus on just making it sound great, because now, especially with loudness normalisation, no matter how loud or quiet our song is, we don't actually have any say or any control over how loud or how quiet the music gets played back out on the other end. So obviously, no matter how loud your song is, you don't have any control over whether the person on the ever and just listening to it through an iPhone speaker offer a massive, expensive highs hi fi system with lots off amp gain and stuff like that to demonstrate loudness normalisation in practice. I'm gonna play back some songs in this playlist on Spotify on I'm going to switch normalization off and on. So let's play some music with each other has must say so little. Okay, as you can hear there, obviously there are loudness differences because no single song has exactly the same loudness for up the whole song. But even through contrasting styles of music there, they were close enough to be considered a consistent level for listening. Now, if we didn't have loudness normalisation on, that's a whole different story. So I'm gonna go and turn that off. So I'm to come up to the settings here, show advanced settings and turn off the option that says, Set the same volume level for all songs, Go back to my playlist so little moving cannot. So to summarize, in today's world of loudness normalisation, it really doesn't matter how loud you make your song. It's only gonna be normalized, and it's only gonna play back at the same level as everything else anyway. So it is no need to be motivated to make your song any louder than it needs to be, because it would just get turned down. If your song is two db louder than Spotify eyes reference level, they'll turn it down by three db. Otherwise, they may turn it up by two db if it's lower. So you might as well just work to the right loudness, sort of right dynamics and of right e que for your master and not worry about how loud it's gonna be compared to other songs. Thank you for watching this lecture see in the next one. 12. Loudness Units (LUFS): Hello and welcome back in this lecture, we're gonna talk about L. U F s, which stands for loudness units. Full scale loudness units for scale is a unit of measurement designed to measure the relative loudness off one piece of material against another, the relative loudness being the difference in loudness between one piece of material to another. The purpose of WFS is to create a unit that easily describes how much louder one piece of audio is against another, its frequency weighted, which means it takes into consideration the disproportionate way that we perceived loudness at different frequencies. There are three different types of L. U F s unit each measuring over different integration times. One of them bean momentary, which is the smallest or the shortest value. The shortest Eliud fest value the shortest amount of time that it's measured. The loudness short term is slightly longer than that on Integrated is designed to give you a number for an entire piece off program material or entire piece of audio. For example, a song. I'll show you all three of them in action with this piece of music right now. So what I'm gonna do is I'm gonna bring my play head to the beginning of the truck On on the stereo out bus I'm gonna open under the metering sub menu logics built in largeness meter. So as you can see here we have Luo fs and for momentary as for short term and I for integrators So what I'm gonna do to make sure it starts counting the integrated level on a click start, I'm gonna start playing the check now The integrated Luo office is to is designed to give you one number for the entire piece of audio for the entire song. Where's momentary on short term fixed amount of times each. Obviously as this is an unmasked track, the Eliud Fest levels are gonna be relatively low from what you could generally expect to find in mastered pieces of audio. The l U Range is just the range in which the Luo fest levels differ. You're smiling face on a car are times never seen so hard like words Never spoken Love boat A look now Okay, so let's stop it then. Let's assume that what I've played so far is the entire song The integrated Luo fest level off the entire song is minus 28.3 l U F s. So that's designed to represent the entire piece. That's why you'd no longer see any momentary values. You no longer see any short term values because there is nothing making any loudness right now on 17 is the range off loudness units. That happened for the peace that we listen to. One loudness unit is equal toe one decibel. If I wanted this to B minus 29.3 WFS integrated, I will simply turn down the audio by one db to get it down to that minus 29.3. Figure one l U is the equivalent of one DB. If you want to change by x amount off L you you change by X amount off decibels. It's a simple is that in the music world there is no single value that it should be. No. All of the platforms use WFS to measure the loudness. Andi, even if they did, they all target different levels anyway in broadcasting which is what this is primarily used for. Its very important is in fact regulation that things go out at certain Luo fest levels. But that is not the case in music. However, we like to reference the Luo office levels in music as a reference just so that we know and we could make comparisons and weaken measure How accurate L U F s is for measuring loudness . However, it is not in itself loudness, Loudness is an auditory sensation and can only truly be measured as such. It's up to your ears subjectively speaking what loudness is and that's why we use fixed monitoring levels so that we can empower our is to do that. Tests and research show that the L. U F s levels that the platforms are generally putting out are anywhere between minus 16 and minus 13 l U F s integrated, which includes YouTube, Spotify and iTunes. However, it doesn't mean you need to match them. It just means that they normalize to that number. So if you master summit to minus six l u f s integrated, they will turn it down to match their level. Also, if you master it really, really low are really, really low Elia fest, they may turn it up, but it's good to be aware of these numbers. It's good to be aware of the research so that if you find yourself making a master, that's minus 21 l U f s integrated. You might consider bumping up a few decibels so that they don't have to turn it up. Thank you for watching. And I hope this helps you understand how to reference value fs meters in the mastering process. 13. True Peak: Welcome back in this video, I'm gonna talk about the difference between peak level and true peak level when you're mastering music, the true peak level estimates what the highest peak level of the audio signal will be when the digital audio is converted to analog via your audio interface. The reason that we measure true peak levels is so that weaken minimize the risk off clipping further down the line. The reason that this can happen is because of something called inter sample peaks, which I employ. You too. Do some research about if you feel necessary. Otherwise, my recommendation is to master to a maximum true peak level off minus one decibels. The reason I recommend a minus one disa bel trew peak level in your masters is to not only mitigate the risk off clipping when converted to analog, but also mitigate the risk of clipping. When your Web master is converted to lossy formats such as MP three and A C. It's very easy in logic pro to measure the true pick level off your signal and it's also very easy to limit the true peak. Level off your signal logic pros. Adaptive limiter, which are load onto distract. Here under the Dynamics menu has a true peak detection option which I've just turned on. And from now I could simply sundown the ceiling tu minus one and then I can increase the gain to as loud as I want it to go. The seething is the highest peak value that will come out off the limiter in order to measure the true peak on the peak levels. Off your audio in logic pro, simply open up the audio effects menu in logic, go to the metering sub menu and choose either level meter and then select three peak Or you can also select logics Multi Mita, which gives you a lot more information including Hey true peak are a mess in this Bring you here We can choose between Peak and RMS or tree peak enormous. You cannot like a day without most So what? But we could always be a as I'm about to demonstrate, I'm gonna increase the gain Governance of the limiter on actual See, It won't exceed a true peak value off minus one number. How much gain up in I would always being you my treasure. We control miss enough one another. Obviously, I've run out of gain here on this dial so I can go just before the adaptive limiter here create a gain stage from the utility menu and increased, gained going in. I'm just going to reopen the adaptive limiter there I was gonna add a bit more gain just to show you a drastic example. Off me, increasing gained Gonen limiter. But it's still capping at that minus one true peak level. Reeses heaped high upon the pie. The script is sold just as you can Quite clearly here we have gone way beyond allowed the sweet spot off this song. However, it's still only putting out at minus one true peak level, despite there being lots of distortion which is actually happening inside the limiter. That's what it sounds like to go beyond the loudness potential off your song. So that is the difference between peak and true peak. Again, I highly recommend that you keep your true peak levels at a maximum off minus one decibels . However, for reasons just heard, that doesn't necessarily mean you can just drive infinite gain going into the limiter. The limiter is still distorting the audio. So you still need to find that sweet spot just to wrap up. That's the difference between Peak and True Peak and keep those tree peak levels at minus one, see in the next video. 14. Export Settings from Logic Pro X: Hi. In this video, we're gonna talk about authoring your master on what settings you should Jews when exporting. Now, the first thing that I like to do before I go ahead and bounce out, my master is used the fade toll. So if I go to my pointed towards up here on select as my secondary point at all, the fatal are, then zoom in to the ferry, beginning off the old air file, do a first like fade just to make sure that it begins at zero. So there's no popping or clicking, then go back to the end. I see MT. Then I come to the end of the file the same in again and imply a very gentle fade ride. The end there as well isn't a fatal. Okay, now let's export, Let's bounce. So I click on the file and I used the command be commands on my keyboard. Which brings me up the bounce menu. Now when you are bouncing out your master for online distribution platforms like CD baby tune core over new uploads, a sound card or any other online audio platform is generally advisable. Toe upload un compressed so un compressed When I say compressed, I don't mean like dynamic compression that the compressor that we used in our mastering chain, but I mean un compressed, as in not an MP three. For example, MP threes are type of lossy compression, which means it works, too. Delete the information that you're less likely to hear and leave the information that you're more likely to hear to reduce the file size, which is a very simplified way of explaining it. But that's pretty much what is designed to do. However, UN compressed file formats such as wealth or a I F. F. They don't reduce the quality off the audio. That's why they tend to have much larger file sizes. Then formats such as MP three now is very likely as well that when you upload toe online platform that it's going to get reduced to something like MP three on the other end. So if you submit to them an MP three and they render it to MP three again, you might get twice a distortion on the other end. So it's advisable to send them un compressed versions off your audio so that the very end result is the highest possible quality. Most online distribution platforms such a CD baby and Tune Call will only accept 16 bit wear files at some poor eight off 44.1 kilohertz. So in order to meet their specifications, we need to make sure that we applied the right settings in the bounce menu. That's right in front of us. So which is PCM? Which gives us the options for the UN compressed formats? PCM stands for pulse code modulation, and over here, which is wife, we bring that down to 16 bit of resolution, which is very, very high quality, very high resolution. And it gives us a dynamic range off around 96 decibels, which is more than enough for most music, which is a sample rate of 44,100. The some part of a given piece of audio determines what the highest frequency that can be captured is without alias ing occurring. To learn more, I suggest you simply look up the Nyquist theorem. At 44.1000 hertz, we could capture frequencies at 20 kilohertz, which is generally understood to be the very top off the human range of hearing, and at 44.1 kilohertz. We can capture that frequency without any alias ing. Artefacts on the file type were generally gonna choose into leaved, which means it is this one stereo file with both the left and right channels embedded into the one file. Otherwise, if we choose split, we'll get to mono files one for the Left channel on one for the right channel. Generally, we're gonna choose into leave now for dithering. Dithering basically reduces the risk off something called quantum ization distortion where the least significant bit is rounded down. Possibly creating harmonic distortion in your dear file, which is very, very low level, very, very unlikely to here in the vast majority of cases. But generally is better to do it then should not do it. It creates a very, very low level noise for much like tape pistes. But even at 16 bit is at least 96 decibels below zero. So generally advise that you do together, which did the use really, really doesn't matter. In my opinion, however, I would experiment with the different diverse settings and see if there's one that you find yourself more comfortable with in terms of listening. But generally, you probably won't notice any difference. So let's just go ahead and just choose normal. Dithering. Now normalize. Just have this off. If you limited correctly like I demonstrated in the limiting video, that's a true pick, sealing off minus one. As I talked about in the true pick video, there should be no need to normalize. Normalising basically takes the peak level and brings it up or down to zero and brings the rest off the level off the signal up or down proportionately to that, there is no circumstance in which I devise normalization, so I would just turn it off and then we click, OK, and then we simply choose where we want to print or bounce the file down, too. So just gonna pull it on desktop as Chris Master test, and now it's simply bounces down the audio file with my processing. So if I want to go and open that file now, I've got to find a go to desktop, and there's my file and just a double check that is all of the right specifications with the file selected, or use the command I command on my keyboard and have a look at the speck of the file, as you could see his wife from audio, which is correct on Under More in fire. The duration of the file. Two audio channels for left and right, 44.1000 sample rate and 16 bits per sample or 16 bit. Bit death. Great. So that file is now fully mastered. That is my production master. And that is what I'm going to send out to online distribution when this song gets pushed out. So that's how you print out your final production. Master, thank you for watching. 15. Album Mastering/EP Mastering: Hi. In this video, we're gonna talk about mastering a collection of songs like an E P or an album. In this example, we are mastering some songs for Thomas. I've started with one song, and I've already mastered it, but I've mastered it on its own track, with its own settings, so that by itself it sounds as awesome as optimal is possible in this day and age. Even if you do put out an album or put out an E P, it's very realistic to consider the likelihood off the songs in the album being listened to or consumed individually, either just by themselves or in other places that the end listener curates themselves. So my advice for doing this is to master the songs separately, each realizing their own unique potentials and then work on balance in the album as a whole . Great. So now I've actually imported all of the songs in this e. P. On I've mastered them individually based on their own optimal requirements. So, as you can see, even though they will have pretty much the same plug ins, I didn't copy and paste the channel settings. I just mastered them individually from scratch. So as singles they are each as optimal is possible. But now I'm gonna work on making sure that they work as a cohesive collective as an e p so that when one flows from another to another, there are no surprises. One isn't a lot louder than another, and that they all just generally feel balanced and work together as a single piece. So first of all, I need to structure them like an album so that I can audition what the actual album will sound like. So I'm gonna move the second song along. It's about there in the third song on the fourth, so and I'm going to use the loudness meter on my stereo out bust, which already had loaded. But if you don't already have that loaded, you simply open your dfx menu. You got to metering and he gets a loudness meter on. What I'm doing now is the first thing I look for is balancing the loudest parts off each song so that when the louder part off one song comes in, it's no way louder than the loudest part of song before it, so that the end user feels they have to turn it down or or anything like that. Andi, for that number, I'm generally looking at the momentary Luo fest levels using the loudness meter. So, for example, if this is allowed apart, this song that's hovering around 12 or something like that now this part this okay, so that's actually quite a bit louder. But we'll have to further investigate whether that's a problem and for this song, more empowered the first song on this one. So there's clearly one that's just way louder than the others. But I'm thinking as well that it might just beat this part, this song now, as that is way louder. What I'm actually going to do is make sure that this track header is selected and go back into the plug in, chain on and see what the compressor is doing, Theo. Okay, so the compressor isn't overly behaving. So what we do is we look at the limiter and see if I can dial down some of that game. Okay, so you can see there's quite a lot. Game reduction happened and it is a little bit distorted, and it was a good few di B's above the other song or turned down again in the limiter by a few DB here. Let's see if that gets us back to a place where the album is quite balanced so she could see the Elia fest level, the momentary Luo fest like around minus 11. Right now, I could see there's quite a bit of game reduction happening there right now. I'm not really hearing that much audible distortion. I think it sounds OK, but if I wanted to reduce the amount off game reduction that the limiters having to do with minimal distortion, I would go back to the compressor before the limiter probably reduced the threshold into. It. Just compresses a little bit more, reducing the peek underneath the ceiling of the way limiters during a little less work now . And it's about the same way. Okay, great. So I think the louder parts off each song are generally kind of on power of each other. So now what I do for a few times over is I just listen to the entire album as a whole. I make sure that things end when they should, and the next thing begins when it should, and vice versa that the gaps aren't too long or short. So I'm just going to go over and do that a few times now. So right now I'm just gonna audition the way that song one transitions into song to over that feels right with the timing feels right, and I just it from there. So let's have a listen thing, Okay? So already this is gonna be way, way too late as far as the album's concerned. So I need to make some adjustments from here. So I stopped the play head at roughly where I think the next song should be starting to come in Andi with the play had the in place, I'm going to grab the region by the very bottom right corner here and drag it into its roughly at the play head position. And then I'm gonna bring song, too. But in fact, I'm going to select all of the songs said I don't affect the gaps in between them. Zoom back into where the play head is on. Bring those along as well. And then I'm gonna skip back and see how that feels. Okay, so I still think is a bit late in my head. I counted two bars after that song had finished. I generally do that as a sort of guideline to roughly when the next song should start. That's by no, By all means not on absolute rule, You can just do whatever you feel was comfortable. But when I'm not entirely sure, that's generally where I start. So I'm gonna bring the next song a bit further in and then try again. Okay? I actually feel that I was a little bit early, so I'm gonna make some fine adjustments and keep auditioning it into I feel it's right. Okay. It's still bitterly Still a tiny bit early. Okay, I'm happy with that now. What I'm gonna do, however, is because when I bounced out, each of these audio files are gonna be bouncing them out at the current left of those files . So if they were uploaded to CD baby and push outs an album, this would end at the end of this song on. This will begin at the beginning of this song. The problem of that is that difference will be back to where it waas and thus very sort of long drag into the second song. So my best option really is to take this region. It's at the bottom right of it and drag it until it's basically at the beginning point off this one so that when they played back as an album in Spotify or iTunes, that timing that I've created gets reproduced on the other end. And, of course, I use my trusty fates. Also again, secondary point. It all fatal is the command key and create a fade just to make sure that the beginning of the file starts at zero and there's no pops or clips or anything like that. So that's generally the practice that I deploy when I'm making sure that the transitions between one song toe, another of recreated, as are intended on the other end and I just go ahead and do the same process for the other songs. Then, just like I did it in the exporting video, I print each file out individually, so I select the file, and then I used the Bounce Command and printed L prints out those individual files, and then those individual master files get uploaded to the distributor. You can then bring those files into iTunes toe, audition them yourself as an album. I highly recommend that practice just so that you can get a feel for what it might sound like on the other end and then take any notes, come back to this session on make amendments. That's generally best practices when you're mastering an album or an E p in logic pro fence for watching and see you in the next video. 16. Dynamic EQ: Welcome back in this video, I'm gonna show you a good case study for using dynamic e que in the monster and process. Logic Pro doesn't come with a dynamic equalizer. However, TD R Nova is freely available from their website for free. I'm in no way affiliated with Tokyo Dawn lapse. However they do give one of their better plug ins away for free. It's called Nova, so I definitely go on and go and get that. So let's have a listen to this really basic really entry level master that I've started during so far. You're smiling face on a car times. Okay, so it's loud. But there are some problems. There are some artifacts. The bottom ends mostly that kick drum is creating some distortion against the limiter. However, I don't want to lose any of that loudness. Andi, the problem is inconsistent enough for me to just try and solve it without losing any of the loudness that I've gained by using the limiter. So I'm gonna come over to my channel strip on in between the gain stage and the Adaptive Limited. I've got loaded. I'm going to load from the audio units menu Tokyo Dawn Labs T d arnaud over stereo. So on the surface, it looks like a normal week. You however, unlike a normal e que it allows me to equalize the frequencies dynamically, meaning that the amount off gain that it adjusts in that frequency band is relative to How much signal is going in? Much like a compressor. I'll show you a working example of what I'm talking about. So I'm just going to rewind here, and I'm gonna click on this band here click band solo because I'm roughly estimating that's where the kick is on. I'm just gonna have another listen, you're smiling face on a car. Okay? The kick is quite heavy, so I want to tame the kick, but I don't want to necessarily e Q the track. I don't really want to change the overall frequency balance, but I do want that kick to just come down a little bit sometimes when it's loud enough. So I select my band, Andi I click threshold just like I would use and the compressor and I turn it down until I start seeing it moving. Never seen so hard. See how it's turning down that band dynamically as the kick happens of Snow Boat. A look never. I'm just going to adjust the quality factor here just to narrow that band of it. Name has told with each other. So it's taming that kick drum dynamically now, so that when I unsold load that band and I hear the entire makes, the entire master is going to help that kick. Just sit underneath the ceiling of the limiter so I don't lose any loudness. But I solved the problem off. Some of the distortion at least must say so. Little way, Find way, whatever way I want. No dreams way. Have a bypass button here so that we can hear the before and after off the effect the corn and pushed the board. And you can move and cannot very slight adjustment. Very slight cleanup of that bottom end bands there. So that is a really good use case scenario for using a dynamic IK use a very simple one. I simply selected the band a solo Did it, found roughly where the problem was in the frequency spectrum but quits of fresh old and brought it down until it started pulling down on the element that I wanted it to pull down on Simple as that again, TD on over is freely available from the Tokyo Dawn Labs website. I suggest you go and get it for me. It's the Swiss army knife of mastering. It has so many applications, is quite often my go to general e que as well. But when I need to solve a problem dynamically like I have just now it's the only plug in for me. But that was useful, Aunt Hope that provided you of a good idea off the sort of problems that dynamic Ikuko unsolved in mastering see in the next video. 17. Mid-Side Processing: Hi. In this video, we're gonna talk about mid side processing. MedStar processing allows you to treat stereo image a bit differently to how you can normally just treat it using the left and right channels when you're treating the mid channel, what you were treating are the some off the left and right channel. When you are treating the side center, you are treating the difference between left and right. Put it simply, if you increase side otherwise known as difference, you widen the image. If you increase the mid channel, you some left their mark together, meaning they become one, and they both for into the center of the image to demonstrate that I'm going to use logics e que. So I have one of Thomas's mixes here. Unmask turd on. I'm just going to demonstrate to you what it sounds like when you're treating the mid and side channels separately. So I'm just gonna highlight a section here. Okay, This is a good example cause there's some stereo information on this. I can hear some stuff to my left. I'd like to hear some stuff to my right on. The first thing that I'm gonna demonstrate is how you can use logics e que to narrow the image. And then I'm gonna show you how to use logic CQ to widen the image. So I'm gonna bring up one of the issues here. So on the track, inspector, I go to the audio effects menu. I go to the e que menu, and I choose either channel eq you or linear Ikea. Now logics e que has an option where we could choose the style off stereo processing that we used in mystique you. So I just choose from I never eat curing the stereo image or the left image, the right image or just made or side channels song transmit on. What you're gonna do is a basic gain on the mid image to demonstrate to you how it narrows it. Okay, So by increasing again on the mid only channel, it's increased the sum off the two. So it's louder, but it's also narrower. If I did the reverse turned the game down, it would be the some image would be quieter, leaving only acquired toe but wider image. So now I'm going to use a similar process to demonstrate the side channel. So I'm just going to recall the default preset, ongoing toe open side only, and I'm gonna play it back. Okay, So by during that, by increasing the side channel or the Difference Channel, it has increased the difference. So the more difference between left and right, the wider the image. This can have very practical uses, especially in popular music, where you might want to narrow one part off the frequency spectrum, but wide in another part, this is quite common in E. D. M and hip hop, where you might want to force the kick in the base, all of the base frequencies into mono or into some. But you want to leave the highs quite wide or even wider to do that, use an example that I've just used. What I can do is simply, with the gain increased, I can create a high pass and bear in mind. Again. I'm only high passing the side channel, and by doing that I narrow the image below the cut off frequency. But I widen the image above it, so that's a really common place and practical application of mid side processing in logic pro. So you simply load up any Q. You choose which channel you want to effect, mids or side only. You can use both as well, so I can leave this onside only and load up another e que and use this one to effect mid only I might, for example, want to make this bands around 750 hurts more narrow on a bit louder. And I can do that by selecting mids only and then just boost in that channel. So that is an example of how to use mid side processing with logic, stock plug ins and how it might be practical to do so when mastering electronic music. Thanks for watching see in the next video.