 Hey, I'm Anfa. I'm an electronic music producer and sound designer So I only use open-source software and Linux for all my work In this video, I want to talk about loudness and dynamic range. I was thinking about discussing this topic for quite a while and I thought I'm gonna do it in this beautiful place listening to The sounds of nature. All right, so what is dynamic range and the broadest understanding it's the difference between the smallest and the largest amount of something that you can register For example in a camera a dynamic range would be the smallest amount of light versus the largest amount of light that the camera can register in In a microphone, it's The quietest sound versus the loudest sound that the microphone can capture and then Let's add without distortion because well Absolutely any microphone can capture a nuclear explosion from five meters away. I Would say but none of them can do it without distortion So for the level of input to count it he it has to be clean. It has to be reproduced with Great fidelity. Of course, there are some limits and some thresholds we could apply like how much distortion is acceptable Etc because nothing is perfect and nothing is actually linear in the analog domain and This very much happens in the analog domain, but there's also dynamic range in the digital domain and That is extremely well-defined Most of the time for example a 16-bit PCM wave file Has a dynamic range a theoretical dynamic range of 96 decibels How do we know that well There are 16 bits if we have a One-bit audio stream that means it's either on or off So the highest fidelity we can really encode is a train of pulse waves and They are all at the same loudness maximum loudness if we have two bits in a room formation We can have either complete silence a Square wave at one-third Loudness a square wave at Two-third loudness or a full loudness square wave But then if we try to jiggle the samples We can create something that will sound more or less like a triangle wave if we have like One two three two one zero Etc. It may sound like a sign like a triangle wave Again, that's extremely low fidelity audio still the least I has probably been used in like Consumer applications was 8-bit. I believe at least in computer age 8-bit samples where common for mod music trackers and stuff and Even I I'm pretty sure that even the first half-life Has 8-bit resolution of its sound files of its sound effects. They sound extremely Crusty and it's a big shame because the sound design is interesting. Here is an example of an 8-bit Sound recording notice how the background noise is all distorted and crackly Now the bit depth is totally unrelated to sample rate Before you've heard 8-bit PCM sound However, it was still at 44.1 or 48 kilohertz sampling rate Now you are listening to 8-bit PCM 16 kilohertz sampling rate That's quite a difference, isn't it? If that sounds nostalgic, you're probably over 30 Okay, but How do these bits actually define the dynamic range? Well adding a single bit to our dynamic range or bit depth of an audio signal digital audio signal is Is giving us extra 6 decibels So if we have 8 bits When we take 8 and multiply by 6 that's gonna be the dynamic range the theoretical maximum Dynamic range that we can represent in this 8-bit digital audio stream Similarly with 16 bits we just multiply 16 by 6 Okay, now this is all Dynamic range of the medium Which doesn't mean the Information you put on the medium is gonna utilize all that for example your 16-bit audio CD could be Produced so quiet that it will effectively only have like 30 decibels of dynamic range because you'll have to play it so loud that you were you will hear the noise floor of the medium Noise floor, that's also a very important concept in dynamic range and Photography or film Noise floor is the grain of the film itself So if you have amounts of light that are so tiny that the grain of the film is hiding them well, that means The amount of signal or light is lower than the smallest amount of light that can be registered cleanly and if for some reason there is more noise our Practical dynamic range gets smaller because quieter sounds Darker lights all get drowned in the noise and that leads us to dynamic range of music There's a well-known concept at this point of loudness war which in summary is a trend of producing the digital releases of music recordings Which are which have less and less dynamic range and Are pushing the limits of digital audio which is pretty much pointless But the record company started doing it, I think because they believe that boost sales It didn't it never does People don't care because every playback System on the planet has a volume control and unless you are making your music so quiet that it's Practically very difficult to play at a comfortable level It is really Irrelevant how loud you make your music There's some very helpful technology especially developed in the past decades and the best standard for measuring loudness in a digital format is The ebu, which is European Broadcast Union are 128 which is the recommendation 128 that defines the loudness units and loudness units full-scale Measurements as well as a few methods of calculating different types of measurements Like the integrated measurement that averages the level across an entire program now European Broadcast Union mainly deals with television That's why there's the term program but we can also apply this to a musical piece a track or an entire album or a DJ mix whatever has a start and an end if you measure and average the Loudness units or loudness units full-scale across the entire length That's your integrated loudness level There have been older methods of measuring loudness the first I believe was VU meters Which were physical needles bouncing back and forth these meters were pretty cool because they average the signal So you they wouldn't react very much to peaks because we as people don't react very much to peaks as well if there's a short pop we don't register it as loud as a Continuous tone that would be of the same peak loudness. That's why short Transients don't bother us that much. Well, unless they Cross the threshold of pain, of course. This is why it's so ironic that the loudness wars or the loudness war have Driven the music to be completely devoid of any drum transience whatsoever because The recording is so compressed and limited beyond belief that When you look at the waveform, you really can't tell where a kick hits or a snare hits You can't tell where's a verse. Where is our chorus? Where's a drop and where's a build-up almost and It really hurts the musical expression and it really hurts the clarity of the sound and Yes, you can be a producer that prouds themselves on being able to push the numbers Extremely high with your music still sounding acceptable. Let's put that in air quotes because what is acceptable to? One person may not be acceptable to another person But what is really the point of being? Like chasing that goal. Does it make the music better? It does not it makes it worse You may say that oh But if the music has less dynamic range It's gonna be easier to mix For a DJ and to play back on a loud PA system That is false squashed music Destroys PA systems. It doesn't sound good. It doesn't play better The sound systems are designed to handle a short Transients so short bursts of loud sounds very well and they can do it very cleanly So if your music has a loud snare and kick transient you can easily have you know like 15 decibels of headroom and the music can be loud and The transients can be much louder and still your music will be clean on the PA system of Course 15 decibels is quite excessive I would not probably want to have as much dynamic range between like the peak of my bass and the peak of my drums That would probably not sound very balanced However, I do like to have natural sounding Transients or natural that doesn't necessarily have to be natural. It's not about naturalism It's about expression and making the sound move you Because when you remove all the transients the power of music is diminished in my opinion Music requires dynamics that is Changes in intensity if our music would be Exactly the same loudness throughout the whole piece. It would be boring as heck the most exciting music is the one that can vary the levels and Inspire motion thanks to that Human experience human life is not steady. It's anything but and the music reflects that and Music is always swelling up and down. We have quieter sections and louder sections and the louder sections only hit hard If the quieter sections are there to lay a baseline if we have no baseline because we are allowed 110% all the freaking time There is no music left and you know, I'm not Like some classical music junkie Who doesn't listen to bass music and doesn't understand what rock metal and drum and bass is? No, I Really like hard-hitting sounds and hard-hitting music, but it pains me that when I listen to Noisiest outer edges album on YouTube. There is like Eight decibels of dynamic range that they could have had but they didn't YouTube is Using loudness normalization, which means that the integrated level of your video Or whatever you're uploading as a video could be your music track could be your album The integrated level is measured and compared to a reference level on YouTube The reference level is around negative 14 LUFS and that is our level that I'm roughly targeting whatever I master music or Music videos music releases talking had videos necessarily should be quieter because Speech alone does not sound well when put to such high levels in my opinion and Also when I'm doing tutorials about music production There has to be more dynamic range because there's a lot of silence in between sounds and Also, you need to be able to hear precisely what's going on and Smashing the whole video with a limiter Would only make that harder. I Don't need to push the limits because the video is gonna be too quiet. There is a large margin Before it becomes too quiet to listen to comfortably There are some videos on YouTube that are made too quiet and I can't hear them But it's not because of the concerns of dynamic range. It's because they were poorly engineered and the person making them either didn't know or Forgot or Didn't pay attention to the loudness of their sound So sometimes listening to some videos on headphones While on the go is difficult and it can be difficult for music as well But I'm not talking about making your music so dynamic. It's unlistensable outside of a quiet Listening space with perfect speaker system. I'm asking for a bare minimum of air in your freaking masters if you've ever Care to measure any of my music releases You'll see that the average around negative 14 negative 12 LUFS Integrated and I think that's that's a good level it has quite a lot of dynamic range for drums to hit hard and sound punchy Without relying so solely on distortion and without having to completely drown out all other audio in your mix To achieve the punchiness which will still be puny and miserable compared to an actual freaking transient I would understand this if we were we were stuck with like, you know 10-bit digital audio Where you actually have to make sure you make the best out of every single bit of dynamic range But we freaking don't We have 16 bits 96 decibels That's a lot of dynamic range and ironically with the higher fidelity of compact disc which undisputedly beats vinyl records if it comes to dynamic range and frequency reproduction and distortion the compact disc is a Fantastic medium and it was an a huge leap in audio quality and consumer audio quality It doesn't mean a tape machine can't sound as good. It sure can but it's freaking expensive and It can't easily the tapes can deteriorate more easily. It's harder to store them, etc and with the advent of the best music storage medium and History the music quality started diminishing because people were like Doing everything they can to waste as much dynamic range as possible and to squash their music To completely unlistenable degree It hurts me so much when I listen to system of a downs toxicity that this album is so limited There is so little transience there It's squashed the drums the drums can't breathe and they are so quiet in comparison to everything else and They could maintain a similar level but still hit much harder and have more punch and air if they have had like You know just six decibels of more dynamic range but no They had to squash it because sales which Didn't affect sales. Anyway, if anything I think people who are aware of the problems with excessive limiting and multiband compressing your masters Would buy less albums which are highly processed in this manner Again, I'm not talking about sound design Smash things in sound design as much as you want I love distortion and multiband compression and removing every single bit of air from your synth sound your basses and your drums But then when you do it on your master bus You're killing your music and also after I squash my drums like that Then I mix it back with the original or I use compression to bring back some transients so they can hit harder You can have all the good things about Squashed highly compressed sound without any of the bad things if you just use compression Parallel compression and transient shapers effectively. You don't have to just mindlessly slam your music with a limiter Don't Can this catch focus? I think it can. I can it all right. I've been talking for way too long right now Thanks for watching But I'm gonna record something on this little bike trip. I Hope you've enjoyed this Thanks to everyone who is supporting me on patreon and libera pay these people are helping me immensely making more videos and Stay tuned for more actual open source music and audio production tutorials Oh, and if you would like some help with Linux and open source music production You can always check out my chat at chat.onfa.xy All right, I should get going. It's become pretty dark Bye