if you have unanswered questions or feel outraged by this video, watch part 2 here: Almost Nobody Knows, Distortion Can Be UNDONE! ruclips.net/video/ldAyJCm8xAo/видео.html For anyone who didn't figure out the formula yet, all of my videos of this sort are in 3 parts and the first is just a warmup into the topic. My full reasoning and detailed explanation is always to be found in the second video. The 3rd is left open based on the comments for part 2.
I think there's a horrible amount of misrepresentation going on in this video. First, If you know what you're doing, clipping is an essential tool to getting a competitively loud mix with the least amount of coloring. This is achieved because no one is using clipping in such a broadband way that this video seems to imply that everyone does. You're not squashing a sound such that you can hear the distortion and you certainly aren't putting a single clipper on your master track and distorting the hell out of it*. You are using 10s if not 100s of clippers throughout your producing/recording process to shave off the inaudible transients so you can push the sound louder into the mix. If you just shave off these transients (aka know what you're doing) you can not hear the distortion because what's being clipped is too short (a few samples at most) to be heard by the human ear. Every producer of your favorite pop, hip-hop, or electronic music uses this technique and you haven't heard the distortion. Second, you very often don't pay for the algorithm but for the features that make the plugin more useful. I paid $30 for the clipper plugin I use and it is worth it to me for the feature of being able to audition the clipping at the same volume as the unclipped version so I can perform a true apples to apples comparison between the two to ensure clipping that is transparent to the ear. Sure there are free clippers out there but if they don't do what I'm talking about then you have to set up volume normalizing in your plugin chain and that's just a pain to do manually so I'm happy to pay for the convenience. It's not bullshit to sell these conveniences. They make producing music less tedious and more fun. If you want a full demonstration of how to use clipping properly, go over to Baphometrix's channel and watch his excellent Clip To Zero series. He gives many examples of this technique and even a breakdown of why clipping many times gives transparent results as opposed to trying to squash everything on the master channel. *You can (and should) put a clipper on your master but, again, that is only to catch the 1-5 sample peaks that would otherwise kick in your limiter when you don't really want it to kick in for those types of errant transients.
so you're the guy who makes all the loud mixes! but you're right in in that its a convenience factor and errant spikes can waste a lot of headroom, and you want to clip the tracks separately so the intermodulation produces harmonic hetereodynes.
@@pepe6666I remember when the loudness wars were at their peak! I'm glad things have settled down (except in EDM). But now we hear advice like 'you should aim for this LUFS level for Spotify or other platforms, or they'll turn it down.' A lot of non-pro mixing engineers just aim for that LUFS target, but then realize their track lacks energy and gets skipped in playlists where the volume drop is too noticeable. I think it's important to know how to make your mix louder, and to understand how loud your mix should be overall. But I'd love to hear your thoughts as well. I'm not sure how much you've experienced the changes in the industry-I've been mixing since 2003.
This guy only has 6K subs for a reason... a lot of these videos of his are just like this, a weird contrarian approach to audio that proves that he knows a lot, but is incredibly obtuse and eager to "debunk" things, even when it doesn't make sense.
@@RooftopKoreansMusic I’ll be honest with you, the reason I follow @AP Mastering is because I shared some of his videos with my team. It’s true that companies keep putting out plugins we don’t really need. In my producer WhatsApp group, people constantly post about new plugins, and I have to tell them, 'It’s the same thing as this other one!' lol. We even had a plugin workshop where I told them: if the plugin saves you time, get it; if not, you're just collecting EQs and compressors you'll never use. ):
1. Clipping (and aliasing) do sound great in some circumstances. "Should never be used" has always been a dumb suggestion in a craft half-defined by exploiting flaws in audio equipment. 2. "No such thing as soft clipping" is just semantics. You might have had an argument 20 years ago, before it became part of the audio engineering lexicon - but today, it's a widely-recognised term that's commonly understood to refer to a specific set of nonlinear processes. 3. There are very real differences between various types of distortion. Both measurable and audible. Transformer saturation is band limited, has different sensitivity depending on input frequency, and reacts with hysteresis - it's not the same as diode clipping, valve distortion, or FM overmodulation. Having different terms for these is useful - both academically and in practise.
1. never heard anything that sounds better with aliasing apart from maybe some weird shit from aphex twin but he uses spectral processing afaik not direct aliasing 2. soft clipping was a term way before plugins. i was allowed to disagree with it back then and i'm still allowed to not like the word. you are allowed to use it but i'm allowed to moan about it 3. i complete agree that distortion is diverse, if i came off like saying all distortion is identical then i did a bad job communicating that particular point in my video
Eh… Yes, as a creator, musician, composer, mixer… - Feel free. But, as an engineer doing mastering, transfers, archival, format conversions… - No, you would never (that’s the never-point) mindlessly mangle your clients’ work… …unless said client expressly demands you to do so, I guess…? 😅
@@Durkhead Exactly. Many people also use old 12 bit or even 8 bit samplers etc because that aliasing just sounds right for many things. Like for example the drums using DAT machines clipping on early drum and bass, lofi hip hop, chiptunes etc.
@@PerttuPiirto 12 bit has nothing to do with aliasing given correct reconstruction. 12 bit sounds cool because it is inherently a type of nonlinear processing and adds noise
@@APMastering saturation is more of a tanh type situation. soft clipping is linear with small tanh portion on top. If you don’t hit the transition point of soft clipper nothing happens but with saturation you are always generating harmonics.
@@APMastering One could argue that 'hard clipping' is an oxymoron compared to 'hard compression.' The concept represents that your sound is being shaped in an extreme way (for example, into a square wave), and maybe everything in between.(soft or softer shaped) So, just as compression can vary in intensity, clipping can also have degrees-hence the term 'soft clipping' to describe a less aggressive form of clipping. 🙌🏾
1. I think it is usefull to have different terms for different types of distorsion. Yes, clipping is a form of distorsion, BUT, using it usually has the purpose of reducing peak levels by a target amount. So in essence, the name of the type of distorsion point to its primary use. Same with soft clippers, it points to the purpose, you still want to basically cut the peaks at a certain threshold, but using a more gentle type of distorsion to achieve that. 2. Hard clipping sounds bad of course if you simply apply it over the full mix and cut into the main body of the song. But that is just showing you have no clue on how to use hard clipping. It can be extremely transparent if used smartly, for example, clipping individual channels or busses that have very sharp and short transients, in which you basically clip around 2...3 samples. I guarantee you cannot hear any distorsion over such a short time, while being able to reduce the peak level by 1...2...3 or even more dB, depending on the signal, basically for free. Added over many channels, this can improve the crest factor of the mix by a lot.
absolutely agree. Using hardclipping in a way that's audible (rather than as a tool to tame peaks) usually sounds bad unless you are going for a very specific sound (i.e. nasty digital distortion).
@@FelixLanzalacohe’s running frequency sweeps on clippers to check for aliasing, it’s not a useful test. Clippers, in this context, aren’t best applied to periodic signals. If you do, you’re just generating square waves, which effectively have infinite, fairly high amplitude harmonics, which will always have aliasing artifacts as illustrated. It should not be framed as a frequency domain issue. Clippers are intended to act on transient events and on a minimum number of samples.
@@FelixLanzalaco Maybe he does know about that, I can't know what he knows or not, I know what was presented in the video, and in the video it was clearly stated that clippers are a scam, they are bad and should not use them, no disclaimers, no exceptions mentioned.
I understand your point but I use hard clipping to get rid of drum sample peaks and they're so short that you cannot perceive any audible distortion. In mastering it helps me achieve the loudness my cleints are looking for & save me time from doing it manually. Aggressive clipping is also part of the energy & sound of some music genres!
I agree that on single sample long peaks it's fine but even on drum samples they are not single sample peaks. Not even a few samples. Even at low sample rates, I have a session open right in front of me and the transients are waves which rise and fall. There's not one single big sample and then the rest is low. That's not how acoustic instruments typically look at professional sample rates
First, you're mixing up methods with functions. Clipper is a function of a distortion plugin, and soft clipping is not a wrong name, it is correctly used to describe the purpose of such distortion - to clip peaks (in order to increase loudness while preserving dynamics). Second, aliasing is irrelevant if a clipper is used as intended - to clip transients. Transients are too short to have any distinguishable pitch, therfore, non-harmonic distortion (aliasing) does not sound disonant, it's just a short burst of noise, it can even be desireable - it's up to taste. While many "clippers" are waste of money, some are simply better than others. Many of us know why we use them, nobody scammed us, we're scamming listeners into thinking there's no distortion happening, and that songs are sounding better when they're actually louder.
I'm assuming you aren't talking about coding with your methods and functions bit. not sure what you mean... most distortion plugins don't do straight clipping unless they have a specific choice for it. When you AB clipping level matched generally it sounds worse. You probably just dont level match. Drum transients are not a single sample long, theres a lot of information loss incurred with clipping.
@@APMastering I'm not talking about coding, things have functions and methods of fulfilling their functions. Read what I wrote, clipping and distortion are not excluding each other, they're different things. When ABing, things sound better clipped if that is what you want. Of course I level match. Clipping can make drums sound more agressive and allows you to increase loudness, which in turn brings more value than what you lost by clipping. It is very simple. Transients can be infinitely long, the length is not what defines a transient, it's relative dynamics. Typically they don't resonate long enough to have percievable pitch if at all. I am not talking about single samples and transparency. The whole point is in getting rid of unneeded information that is eating up headroom, and having that energy back in form of overtones redistributed in less saturated parts of the spectrum.
@@duncan.o-vic yeaj but it sounds like shit. ive heard drums that clip. anyhere ya clip, a compressor can do better. it all needs to be done with finess.
@@pepe6666 I have heard many more compressed drums that sound like shit. People claiming they have never heard good sounding clipped drums are usually unaware of how many of those good "unclipped" drums are actually clipped.
Okay but you gotta acknowledge that these tools are used in sound design and production too, not just in mixing and mastering. Aliasing doesn't matter in some circumstances, like when it occurs in the transient of a kick drum. "It ruins the sound" not when used in the right way, and not if it sounds good anyway (no one's enjoyment of music hinges on what the PluginDoctor graph shows). Depending on how you clip, you can even put it on entire subgroups just to gain a few dBs of headroom. Try it on your drum buss, with a hard clipper as your first insert. Don't drive into it, set the ceiling to -0.1dB and just leave it. Check your LUFS with the clipper on and off to confirm the difference in headroom. Yes soft-clipping saturates, but it does so by rounding off peaks, leaving the rest of the signal untouched. That's why saying "there is no such thing as a soft-clipper, it's just saturation" is false, since a soft-clipper still behaves like a clipper. If anyone is interested in clippers: 1.)Standard Clip for hard clipping (some people prefer Kclip) 2.)Newfangled Saturate for soft clipping (their spectral algo makes it perfect for sound design) they're cheap and get the job done. free alternatives: G-Clip or PeakEater ....... No, you don't need the $300 Platinum Diamond Encrusted Super-Duper VIP Clip, or whatever nonsense it's called 😭and yes, they definitely paid the MixbusTV chad/guy to tell you otherwise(great hair btw 💪I have no doubt that he has lots of sex all the time(unlike us))
lol last sentence is all that matters. that said, i disagree that saturation doesn't have a limit.... if this was true i could play wembly stadium with a cigarette packet amp
In the '80s and early '90s, when mastering engineers were allowed to do a decent job, digital clipping/overloads were seen as a sign of a defective master. Clipping still shouldn't be considered the norm. How long will it take the music industry to realise that streaming normalises audio (in the case of Spotify, to a relatively conservative -14 LUFS) so you DON'T need those stupidly hot levels anymore?
Thanks! My RUclips suggestions are getting really uninteresting lately and FINALLY have I the chance to discover fresh, new, thrilling stuff for me to enjoy!
honey will ooze out of the speakers with that one! i'll maybe make a new plugin: the magical 3D analogue warmth generator. it will have a big glowing tube and simply boost the gain by 0.5db
I really like the stock saturation plugin in Ableton. I use it both to add harmonics and clip off transients on drums at the end of the individual drum processing chain.
Yes, there are so many terms people toss around to mean "distortion". People love overcomplicating things because they feel it makes them "smart". Distortion is any change made to a waveform. It's as simple as that.
The common point between a clipper and a soft clipper is that they have a limiting threshold. The word distortion doesn't necessarily imply a bounded transfer function.
Clipping IS distortion. You want to round those peaks off. Not snip them. Think of a guitar Fuzz Box. They clip the signal in a harsh way to get that edgy sound. Overdrive pedals, meanwhile, give you a more rounded distortion because they round off the signal more gently. It's driven pretty hard, of course, so in one sense it can hardly be called gentle. But it's a more gradual distortion you can ease into.
You’re confidently wrong about quite a lot in this video. For one, “Clipping is bad” yea if your using it to “sound good”. I use it to trim transient peaks so I can push a signal louder and to make dynamics processing behave more consistently. Not for colour. Also oversampling is overrated imo. We’re not clipping sine sweeps. We’re clipping complex waveforms. I find often that the non-oversampled clipping sounds better to me. Although granted I make heavy music. Your mileage may vary. Also saying “soft clipping” doesn’t exist when we all know exactly what people mean when they use the term seems pedantic. The signal still “clips” once it surpasses the soft knee. So why draw this “it’s not clipping distinction” because of a slight curve just before the impassible instant threshold (clipper) It’s a perfectly useful term 🤷♂️ While I agree that all clipper plugins are basically the same thing, and you don’t need to pay for anything fancy. They’re an extremely useful tool and I use them all the time.
by your definition then, tanh is a soft clipper. But it's used predominately as overdrive. In terms of drum transients, I use distortion all the time but not clipping. Clipping is literally digital information loss, like clipping the whites in a photo.
@@APMastering you do clip though. if your distortion transfer function / non-linearity graph has a hard limit (usually at 0) then it is also a clipper just with a non-linear in/out beforehand. You will delete information at the peaks with most saturators too. Just depends on how hot your hitting it.
Given how clipping is used when used properly, I think you were close when you were talking about the time domain. Clippers should be set so that the minimum number of consecutive samples are clipping. Ideally under 4. At this level, aliasing really isn’t a thing to consider, we’re not talking about periodic signals. Clippers should be set conservatively, and ideally at the individual mic/instrument level. It’s the same reason live sound folks would overdrive analog console channels for snare drums and other transient-rich sources. It saves speakers and prevents downstream compression from being overworked.
Id have to disagree a bit. in the 80s and especially 90s people really loved transients on hifi recordings and went to great lengths to preserve them with their tech at the time. In terms of
great explanation :) although i would like to say that your idea of not calling things soft clipping is a good idea. however it isn't that these things aren't soft clipping, they are, its that soft clipping is not descriptive enough to discriminate the different algorithms. thats the meat of your argument I think. someone here said you can use hard clipping just fine if you clip instruments on an individual track basis. its not wrong, as you probably know because the intermodulation will cause heterodynes that are a harmonic relationship of the individual instrument's harmonics. but i prefer to use a compressor on a per-track basis because clipping is polluting the spectrum. plus they want loud mixes, which suck.
@@APMastering I think you're right in how you classified stuff into time based vs not time based, linear vs non-linear grid. But there are soft stones though haha. I think people are giving you a bit too much shit. Your classification is logical. I think soft clipping is a measure of property but it's not an identifier. Like saying someone is tall - it's not descriptive enough to identify an individual because many people are tall. But it can still be true. A bit of a logical twist. Another issue is 'saturation' is used to describe mathematical addition/multiplication which hard clips at a maximum value. For example there is saturation multiplication in CPUs which will multiply two 8 bit integer numbers without the product going over 255. It's called saturation, but in audio it would be called clipping.
I enjoyed the video and the discussion! But I did not expect a lot of mean comments here, that's kinda sad after all the effort you put in your stuff. Suddenly everyone is an expert and calls you "scammer" for requesting an email adress. People: the code is public, no account needed, and you can always use a temp email. Thanks for uploading the source code! The world needs more open source plugins ;)
my thoughts too. email is to broadly keep track of usage and communicate with users in case of big bugfixes or announcements etc. but i offer a DELETE ME option and then i instantly delete the email and have no record so temp e-mail isn't even necessary
This is the biggest nonsense video where you're just talking in circles and contradicting yourself. Clipping is a special case of compression; all compression causes distortion which is half the reason for having multiple compressors. If a processor chops off the top of the waveform its a clipper, regardless of the specific limiting function. The reality is just about every plugin other than a gain plugin causes distortion, last time I checked we don't call eq's and compressors, and delays...... distortion.
@@APMastering Speaking/asking as a non-expert here but if a Gain utility gets pushed way above 0db, it will clip and cause distortion... same for an EQ if the output goes way 'into the red' or am I completely wrong?
StandardCLIP from SIR Audio Tools is the only clipper I use in the master bus, simply because it gives you control over all aspects of the process, and it's got loads of oversampling. In most cases, clippers can do more harm than good.
I think the main point worth emphasizing here is not so much that clippers sound bad (which is a matter of opinion) but that aside from oversampling, all clippers sound exactly identical and you are literally getting ripped off if you pay for one.
Many analog distortions do have a time component. Tape, transformers, diodes, transistors, etc, are all influenced by the past current/voltage. That means that they will respond differently depending on what signal has been passing through them. Soft clipping also exists, and introduces some saturation prior to hitting a hard ceiling. This is how most analog clippers actually function. The only “true” hard clipping would be a digital ceiling function though.
in my opinion it is more accurate to use the word saturation instead of soft clipping for what you describe. although hystereses is a thing in magnetic flux saturation, on this quadrant it is still considered static, the same way EQ is considered static yet has a time based aspect... time based effects are those with overt time based controls like reverb and chorus, and also compression for the sake of this quadrant. I hadn't heard of significant hystereses in diodes. although i'd tend to think you are wrong on that, i'm going to go away and research it because that would be interesting
@@APMastering Hysteresis is a magnetic phenomenon and would only apply, in this case, to tape and transformer saturation. That does not mean that diodes and transformers are not impacted by prior current/voltage, although you are correct that diodes would be the least affected of the mentioned components. I do not believe that saturation is a more accurate description than soft clipping in this case because saturation does not imply any sort of hard limit on the input amplitude. If the function or process does not allow signal over a certain threshold, that is a clipper. Edit: I’m wrong about hysteresis being only magnetic. That is one kind of hysteresis and the one I am most familiar with. However it is the broad term for any kind of “memory” effect, so disregard my first sentence.
yeah exactly, i raised my eyebrows there so cool that you corrected that. yeah i'm not claiming diodes have zero influence over past conditions, surely they do, i've also seen effects from irradiation, let alone temperature. but generally speaking they are not considered components which exhibit large effects in that specific regard
The reason why people call it soft clipping now, I think it's because of FL studio and their world famous fruity soft clipper. But I'm the newer versions, the soft clipper is a preset of the wave shaper... In the distortion folder 😂. So you have a point there. I primarily use FL studio and I find it a waste that people in FL tell people to buy third party clippers. If your DAW has a compressor that allows to turn off all the time constants, there you go... You have a clipper. Congratulations 👏🏿 🎉
exactly! although my compressor does this, many have non zero minimum times. but given fast enough times, you can do that. nobody should ever buy a third party non oversampled naïve clipper though! even before FL cubase had a soft clipper plug. i get why the name is attractive but it really is just saturation modelling
I remember the term "soft limit" from Apogee. Maybe soft clipping comes from that term, although that Apogee process was an analog circuit before the AD conversion.
I will agree that analog clipping always sounds better. That why I have so much analog gear. But i very much disagree that clipping never spunds good. Bass, Guitars, snare drums, and many other insturments sound amazing with clipping and depending on the style of clipping or distortion you can get amazing tones. So yes lots of clipping sounds amazing. But i agree its all just distortion. I think what you may be overlooking is that many styles of distortion or clipping is labeled as things thay they emulate. Tube, tape, transistor distortion emilatons are called clipping or saturation and given a label of what its imitating to make it easy for the user to know what to expect. Is it just a clipping algorithm? Yes yes it is.
Distortion is stupider word than any of the ones you listed. Crossover distortion, frequency modulation, phase distortion, ring modulation, sample and hold, bit crushing, sample rate reduction…None of these processes have anything to with each other except that they are called distortion. Also the video seems to skip over the point of using a clipper. If a snare hits a limiter its effects will be heard unit the limiter fully releases. If a snare hits a clipper the listen probably wont notice the clipping because tons or frequencies (including super hight up ones) are playing anyway during that one sample.
@@APMastering I’ve heard them called distortion in different contexts. Just because you wouldn’t call them that doesn’t mean that someone else might not. The list I gave is all things I seen refered to as distortion in textbook like materials.
Nice Video but there are some things to mention here: 1st - Yes Hard clipping causes Aliasing. However there really are Clippers that are better than others because they have better or worse Antialiasing. 2nd Softclipping exists. It's just a function that looks very close to a clamping function. So it's a (more or less understood) terminology. 3rd Every audio effect comes with a tradeoff. Sometimes using a hard clipper is the best tradeoff if you want to get sth. loud af. So you definitely should use a clipper if you know what you are doing. To me it seems like you didn't find out what value they can have ✌🏼
@@APMastering The meaning to me is: a function close to hard clipping. If you compare plugins you'll see that they more or less do the same thing. So to put it differently - It's a name for a certain kind of distortion. So I think there is a meaning yet it is kinda vague. However, just saying:"Distortion" would be even more vague. So why not use the TERMINOLOGY?
@@APMastering what do you want? Yes tanh is quite descriptive yet tells you nothing about the possible variables.. I don't get your argument here anyways? Do you suggest we should talk in mathematical terms instead of calling sth. a name that everyone understands?
@@metaspaceaudio if you want to be specific yes you would name the function. and there's no variables a tanh is a tanh. its a well defined specific function.
As a guitar player I was always confused by drive, fuzz, distortion, all these terms. I never got it, it's all just saturation and clipping right? Then eventually I figured out the terms don't indicate anything fundamental, it's just different adjectives to describe the quality of saturations. Of course in guitar fx world you also have heavy eq shifts and such in those pedals to shape their sounds but yes. Sometimes when I need a clipper I just use the stock pro tools air distortion plugin, I made a preset that just lets me use the clipping threshold as a clamp.
exactly. the stomp box guys were using all of these words well before the plugin world got hold of them. It's all just distortion and you can do that in any way you want. Put your finger nail against a speaker cone and you have distortion. Call it non-linear finger nails.
In the realm of circuitry, (meaning not programming), there are hard and soft clipping configurations. And then there is also saturation… usually achieved by gain staging, alone. (This can take you all the way to Fuzz). Which is even another kind of (clipping) The origin of these terms started there… and are quite real and distinctive. I’ve built each of these circuits using components configured as to accomplish each reality and sonic distinction. That is all I wish to add. Take care!
are you talking about gradual onset diodes then this is akin to saturation or saturation a mimicking design. If you think there is a substantive distinction between this and saturation please get in touch and I'd love to discuss further.
@@APMastering Yes I do, but it’s too much to go on about. Not nearly as fun as actually building each kind of the wave manipulating devices. Aaaaaaand then playing my guitar through them. Or Drums, vocal and so on. Thank you very much for the offer of the conversation, though. Maybe I should have not commented on your post. I’m not a great conversationalist, with a limited vocabulary for such things. Take care fellow Artist!
you are right that buying a clipper plugin is stupid because you can get a free one (like freeclip) or more likely already have one included in your DAW and it does exactly the same thing as any expensive one. But otherwise I think you are wrong about clippers necessarily "destroying your audio". Clippers (meaning hardclippers) are not meant to audibly change the sound of your signal but have very specific use cases when you have very high, very short peaks (which is much more likely in digital production). Having a hard clipper before a compressor to tame those completely useless peaks is a great technique. Of course hardclippers (most of the time) sound bad for coloring because that's not what they are meant to do.
It is audible though, even if its not insanely noticable and often even pleasant to my ears. If i clip the transient of a snare for example (like 3-4db) it makes it sound less snappy, try it
Your correct, clipping does have use cases where it makes absolute sense and it doesn't damage the "sound". As usual I don't think APMastering has the best grasp of mixing. Clipping is a choice with benefits and downsides, sometime the benefits outweigh the downside.
The only thing I'd disagree with is the intent behind the "conspiracy." There's probably no conspiracy. It's just a culture. I'm certain most marketers believe their own BS. The human mind is very prone to bias and to being fooled by its own perception. They say "seeing is believing," but in this case, it's more accurate to say that "believing is hearing." We can easily be tricked by our preconceptions into thinking we've heard something that isn't there. And those preconceptions are spread by everyone in the culture. There's no advertising mastermind needed for that to be true. Human bias alone is enough.
im adding a bit of sensationalist entertainment factor and caricaturing the audio industry. I agree there is no central department of propaganda for audio
Amazing video, i only got the Venn Clipper because its actually useful and lets u change the curve how you want plus it lets u select different curves for the positive and negative parts of the signal. PS: thanks for using my samples on the background music, that percussion loop from Audentity Records you just used is made by yours truly :D
hey, which music? I used a few different tracks but all are from the RUclips audio library. did you submit it there or did the guy in the RUclips library steal your loop ha ha?
Where's the scam? Following your "logic" there are no apples, oranges, lemons... There are only fruits. Different names exist to describe different "flavors" of wave shaping. You're using sensationalism and clickbait to promote your plugins. Nice way to finish your career before it even started. Clipping and saturation are building blocks, which you can combine with other components to achieve different types of distortion. If you know just of two simplest examples of distortion that doesn't mean that all other plugins are using just that. For instance, Scream 4 distortion unit in Reason has many different types of distortions that include much more than the two methods that you mentioned. Try replicating those types then make another video on how you didn't succeed and there's a lot more to learn before you start trashing other people's work.
you're accusing me of saying stuff im not saying. In this very video i say that subcategories of distortion are perfectly good, for example "tube modelling saturation" would be a perfectly fine description of a type of distortion. The rest of your comment is just random hate.
This video is a scam. Absolute nonsense. If you know what you are doing, clipping has many advantages and benefits. For many, also a killer sound shaping tool. But I hope you got some traffic to your plugins site.
sure, I sold a lot of free open source plugins today. Clipping has no advantages I am aware of. Generally speaking people who use clippers in their mixes improve their sound quality when they remove the clippers and learn how to mix better
@@APMastering Free courses as well? :). Stop spreading misinformation and confuse those new to the game is all I am asking. Using clippers has absolutely nothing to do with the mix itself or general mixing skills.
@@APMastering Clipping very much has an advantage if used in small enough doses where the THD it introduces is basically inaudible. By applying a small amount of practically inaudible distortion with clipping, you can use less limiting, which might have been more audible in that specific context. In a situation where adding yet another 1-2dB of further limiting is audible but adding your first 1-2dB of clipping is not meaningfully audible, you are effectively getting a sound that seems "cleaner" to the listener in blind listening tests, even though you are adding some THD. Yes, too much clipping can sound horrible unless you are using it as an effect. But the thing isn't to compare 12dB of clipping to 12 dB of limiting. That's not what people are using clipping for along with limiters. The idea is to compare, say, 1-2dB of clipping and 5-6 dB of limiting to 7-8dB of just limiting. In that context, adding clipping and removing limiting may sound cleaner-provided the added distortion is quiet enough as to be effectively inaudible. I hope that helps!
everything you say is possible with a standard overdrive. also, although you are taking strain off the limiter, the dynamic range reduction isn't free. when you do this incrementally across a whole mix you don't notice the damage it does until you AB level matched with the clippers removed. I was doing this stuff with clippers literally 25 years ago because i wanted to get loud mixes and then i was given some tips by sticky (was a big garage producer back in the day) and i stopped doing all that and started mixing more naturally and my tracks immediately started sounding substantially better. night and day. that was a quarter of a century ago and i've learned a lot since then
@@APMastering clipping is an integral part of getting D&B to the modern standard of 7LUFS or louder while retaining some dynamics, and we either run at 96Khz to prevent foldback at Nyquist or we oversample like crazy. I would even go as far as to say, it is impossible to get a "modern" mix up to a decent level without clipping (and properly gainstaging INTO the clippers at each stage) in many electronic genres, but if you can prove me wrong, be my guest. But STOP misinforming people, if you want to do oldschool CD masters at -12LUFS fine, but that goalpost has moved on (in my genre).
Makes Total Sense Now. I was experimenting with my WA73 Preamp (Transformers), looked at the meters and waveform and sure enough they were clipped off. Now that you explained it and have confirmed my suspicions, gives me more options to work with and try new things to see how other Saturators react when pushed. Love Your Channel And Your Plugin.
This controversial approach for clicks and comments is only going to get you so far in the end. I haven’t seen many audio channels consistently spreading misinfo
@@andreakleiner80 I've been there sure the expert was talking rubbish then as I got deeper into it realizing that it was me that me who did not understand, no shame in it see you at the other side.
Great video and I totally understand what you are saying with term usage, seems a bunch are not getting your point. For those of us that need to do louder/mixes and masters, what is the best form of clipping? going hot into a ADDA converter?
high end ADs have graceful overloading. this is the same as what I was saying with DAT recorders, because they are also high end ADs (for the high end models). However, you can accomplish the same in your DAW with oversampling and a distortion plugin then brick wall limiter.
@@APMastering "However, you can accomplish the same in your DAW with oversampling and a distortion plugin then brick wall limiter" Now this is something I need to understand, is not a brick wall limiter doing the same as hard clipping, it shaves the peak of. What changes if the peak is "soft clipped/distorted" before it goes to the brick wall limiter. I have mixes where i have used limiters on instruments and I have to go in and automate in dips to get rid of unpleasant "pop" like sounds when dong additional compression at the mastering stage. It is a given that I need to consider peaks at the sound selection/orchestration stage to avoid it being a problem in future compositions but still I'd like to understand what options I have to "transparently" tame peaks. Is there any plugins or applications where I could instruct the program to scale down all peaks by completely redraw the waveform scaling everything down in an offline mode? I do use the draw functionality in my editor to redraw problematic peaks but this is only viable if the number of peaks that needs to be reshaped is few. Great video Keep em coming!
Coding is not boring! On clipping, I tried it (don't judge me bro) and the result sounded bad, really bad. Maybe that was just my ineptitude, but I have since then not tried it again.
SOFT CLIPPING: You say it's "incorrectly" named, but if we all know exactly what it is and (more importantly) what it sounds like when we hear the name.. then, isn't that a really good name for it? (Point being: words aren't numbers. They change definition all the time.. in the pursuit of efficient +communication+.)
im not a linguistic prescriptivist so I fundamentally agree with you but I'm allowed to dislike certain words. For example, I don't like the words "loo" or "couch" either.
I think there's a reason I stick to free clipping and saturation plug ins. They sound plenty good (unless you don't like the sound of clipping I guess)
Hey some really good info thank you for the breakdowns. And it's good to see your plugin design is not yellow with a pink smiling sausage on it, catchy but really it's good to know some built in plugins can do the same
@@APMastering It's good for leads, makes easy to get sound quickly. It was suggested I use FL S Waveshaper to get similar sound, if I ever make money at my sound I'll buy that sausage
As with your other debunking videos, is this also going to go up to part 3? I can already see the comments are going crazy, but from the quality and depth of your EQ & Compressor videos, I'm ready for it!
Nice video! I agree with all the technical arguments, but I still like using clippers sometimes. So, I don't fully agree that they always sound worse than other distortions or limiters. However, it's important not to rely on them as if they are a magic solution to boost the loudness of your tracks.
I'm a big fan of UAD modeling the way particular units distort, that's pretty much my go to for color. From the pieces of gear accessible to me for comparisons, they do as good a job as I know of. I've never understood the whole "throw a clipper on it & it'll sound better" trend among producers. Ableton's Saturator has some good clipping algorithms though.
UAD is not bad, I just think it is overpriced and everything you can do with those plugins you can do elsewhere for free with more efficient plugins. they look good though and are good emulations
@@APMastering Certainly nowhere near everything. You get what you pay for at the end of the day, no points lost for being higher end. If you can point me to a free plugin that can sonically match the UAD LA-6176 plugin alone i'd eat my words but it's likely not out there. That one is a beast.
Clipping is rather a technical tool u use for specific purposes. Art and feeling happens way before the Clipper, it isnt created by the Clipper but rather altered according to technical necessities, such as reaching a certain LUFS value for ur client or making too snappy drums hurt ur ears slightly less.
@@Melloh-293 clippers are used for sound design so they are absolutely an artistic tool... this is idiotic. this guy thinks skrillex's music is garbage so i dont give a fuck what he has to say, he can stick his fossil videos up his bottom
it’s fun thinking about this for guitar pedals. just call “overdrive” saturation, just call “distortion” and “fuzz” distortion. i don’t think it’s ever happening but it would be entertaining to watch
many of those stomp boxes just have diodes and so either are analogue clippers in the naïve case or attempt to mimic amp saturation. but i really like stomp boxes! so many different kinds. but yeah, the pedal manufacturers were playing the terminology obfuscation marketing game well before the plugin manufacturers
@@APMastering i love them too and in that context its not hard to find distinct value in either a discrete transistor cascade circuit or an opamp circuit with negative feedback diodes. guitar signal flow is very accommodating with multiple stages of filtering and limited-range speaker (sim). so it seems like you could get away with distorting the sound in pretty untraditional ways in a more traditional context. is it going to be worth the extra time involved though? that’s the billion dollar question.
@@sweeterthananything yeah it's the best place to experiment with electronic circuits!! i don't understand the application of clipping diodes in the negative feedback path of an opamp though as that seems to me at least intuitively to promote open loop gain, or at least react a bit weird with loud signals. in fact this sounds like a rudimentary upwards expander circuit... maybe clipping after that stage would yield an interesting experimental kind of sound. that said im probably just misunderstanding your comment 🤣
@@APMastering diodes in the feedback path of an op amp is often called a “soft clipping” configuration, coincidentally. This is one of the most common ways to create overdrive circuits. Famously used in the Ibanez Tubescreamer.
@@spicechateau just looked up the schematic... really interesting design! They are doing the exact opposite there as what I got from the initial comment. I thought diode clipping (IE with zener diodes etc) was in the negative feedback loop, which I think would indeed cause wild gain. But if I understand what is going on correctly there in the tube screamer, the diodes are closed, creating a high gain scenario with low input levels and then they open up with larger voltages, closing the feedback loop and effectively attenuating the amplification (or clamping it to unity). But because the particular diodes chosen for that design had a gradual forward conduction, this circuit mimics amp saturation... but its further spiced up by the capacitor which creates a kind of tilt shelf, I guess making it more distorted in the highs?? Really cool design! I don't understand everything that's going on there as im not an electronics expert but I hope I got the general idea. I think I'd still refer to this kind of this as a tube saturation mimicking design or something... after all its called a tube screamer, not a soft clipper screamer.
I think at a certain point, the code itself is secondary, what you're paying for is the user experience. I found this by making a rudimentary amp sim by using an fx chain of free filters and saturator plugins. That and if the plugin company put actual r&d into coming up with more efficient DSP algorithms, but I doubt that's the norm
Very true but I think the video misses the idea that oversampling is really upsampling-> process-.> downsampling and there are various methods of upsampling and downsampling of varying quality and someone might be willing to toss a few $ for a better decimation filter.
Cool vid as always. Although u can actually clip the individual drum shells separately with an oversampled hard clipper. This way you get a very punchy sound that you can not get any other way. Also when you clip groups or busses you get horrific IM distortion if you r in equal temperament
A clipper has plenty of use. Sometimes a track may have super high digital peaks introduced by various processes, and you just want a non-fussy way of preventing those peaks from going into the master bus or sub-group. I don't want to be thinking about distortion plugins to get this job done, I just want a plain vanilla peak stopper. If there are soft clipping options then that's fine, I can explore those possibilities for saturating the signal as well, in which case I'm changing its "official" usage from mere peak stopper to distotion. I have no problem understanding what's going on. I don't feel conned, expecially as I mainly use a free clipper.
there no such thing as a soft clipper. unless you are making very unnatural experimental electronic music, clippers generally don't improve audio quality over a standard simple overdrive plugin, say from airwindows.
@@APMastering Indeed, soft clipping is distortion, but the term "distortion" is actually more vague. Is it phase distortion, overdrive or bit crushing? Is it a pedal or amp sim distortion? is there an eq stage? At least with a clipper you know it's just a signal hitting a wall, whether that's a straight line (hard clipping) or a transfer function curve (soft clipping, or if you prefer, distortion).
distortion is a signal hitting a wall. that's what saturation is. I have nothing against descriptions of distortion. I just dont like "soft clipper" because its stupid
This point that "all dirt can be classified as distortion or hard clipping" seems arbitrary; you might as well lump distortion and hard clipping together as well and say "they are all waveshapers, we don't need other names". But they *can* be meaningful if used responsibly. Instead of describing the technology, they describe the transfer function. "Saturation" usually means a subtle transfer function that bends towards but does not reach a hard ceiling. Soft clipping usually means a hard ceiling *and* some distortion of signal near the ceiling. Fuzz usually means that the transfer function has significant nonlinearity *away* from the ceiling such as a gate effect or rectification.
so we mostly agree but I want to clarify a bit... im not saying that we should remove any distinctions and call everything "distortion" and ban all subcategories. I totally get that fuzz and tube modelling saturation are different kinds of sounds and we SHOULD use appropriate names to describe what we are trying to achieve with our plugins- However I disagree with saturation not having a ceiling. Of course it does, otherwise all amps would have unlimited gain, At some point a 50w guitar amp doesnt go any louder than it goes. it has reached total saturation. and that's a ceiling. We can talk about sag, but that aside, the ceiling is a ceiling and saturation has a ceiling. What is tanh(5) and what is tanh(10000000000000000000000000000000000000000000) ?
@@APMastering It's dead easy to design a digital saturator that *doesn't* have a ceiling: just mix in some dry signal to that tanh function. Usually of course there *is* a ceiling... but if you aren't meant to actually hit it in regular operation, that's academic. People don't clip with Pultec transformers, but they still like how they sound. There is a bit of confusion around the word "saturate" I guess; in physics/electronics terms yes saturation means running out of headroom. I think there is some ambiguity around whether a "saturated" signal is fully saturated or just displaying some of the effects of that process.
@@FrequencySummoner reaching saturation means that it doesnt keep going. that's the point of the word saturation. obviously if you mix in dry signal then you will keep going past the ceiling but my point was that if you do tanh(5) or tanh(243578234598234572893457298345723984572893457234895234798) you basically get the same number out the other side
When people refer to 'clipping' in the context described in this video, what they mean is driving the input of a Lavry Gold ADC from the output of an analog mixing or mastering console running hot. It's a technique. There are variations on it with similar equipment. Whether it's a good or bad technique is up to you. There's a nice demo of the technique over here: ruclips.net/video/e8RT5c1b0wg/видео.htmlsi=CQHMFMMLbGFTGvoU&t=421. In the same video he explains the application of the technique. From the Lavry manual "the SOFT SATURATION feature provides a digital emulation of an ideal magnetic tape saturation characteristic." So there are two parts to the 'clipping' technique: driving the analog front end of the ADC and the digital soft saturation algorithm after conversion. The 'clipping' technique actually happens across both domains. According to Copilot “this technique is often used in mastering to increase the perceived loudness of a track without using traditional compression or limiting. It can add a certain warmth and punch to the audio, making it sound more impactful.” Pretty interesting if you ask me.
Dan Lavry knows what he is on about and I have owned Lavry convertors myself. Notice how it is called soft saturation and not clipping. In terms of streaky, I'm sure he's a great guy but I end up disagreeing with him more often than agreeing. I'd quite like to interview him one day and get to the bottom of the disagreements.
@@APMastering "StandardCLIP was designed to handle the clipping process as flexible as possible. You can adjust the way the clipping is done easily, like a hard-limiting brick wall or smooth soft-saturated." It also says soft saturation here. I'm not saying it's right, but it's commonly the case that when people say 'clipping' what they mean is making their mix or master louder with Lavry like saturation. Sometimes folks use a word that isn't the technically correct word for what they mean.
Good video, but to say that clipping without oversampling sounds “bad” as a blanket statement is a matter of opinion and not factual. There are times when clipping definitely sounds better.
I dont really know of *many* cases. I agree that blanket statements are normally a bad idea as you only require a single counter example to debunk them, but I can only really think of single sample spikes as being a use case for naive hard clipping without oversampling and that doesnt represent that many signals, unless you are recording close mic snare drums rim shots at a low sample rate
Yeah after many years of being very careful to avoid clipping, because it sounds bad, I'm not about to spend money on a plugin that does the thing I've spent years carefully avoiding because it sounds bad.
This is a great click-bait title - nice work! But basically it just boils down to semantics. I’ve watched the follow-up and mostly agree with the technical points you make. Personally I don’t ever hard clip - and if I do, I make sure it’s oversampled. BUT to say that soft clipping isn’t actually clipping… I understand the technical distinction but at a certain point the end result of saturation is a maxed-out value, which is effectively clipped. Saturation is never reversed in the real world in the way you demonstrate, so even though the information is technically still there, no-one will ever hear it. And to say that all hard clipping sounds bad is just as “inaccurate” as calling saturation distortion “clipping”. You say yourself in the follow-up that over a few samples, hard clipping can sound fine. I actually think it’s great that you’re getting people to think I’m more detail about this stuff, but doubling down on the idea that clipping is a “scam” and hard clipping should never be used because it always sound bad undermines your credibility. If you want to be super-precise and nuanced about the technical stuff, you need to do the same with your musical commentary as well, IMO. Joe gives some great examples of times when hard clipping can be useful and sound absolutely fine in the latest episode of my podcast, which came out today coincidentally (I found this video after someone in the comments there asked about it) ruclips.net/video/XT4g6fawjko/видео.htmlsi=_f6ArOhAokMDzikv As always in audio… it depends !
Thanks for your comment. Some clarification: 1. I disagree that it's click bait since that assumes it does not correspond to the message of the video, when in fact I say in the video that it is my on-the-record position that clipper plugins are a scam. 2. Terms: clipping is an extremely well defined term in multiple industries. I'm using this definition. Saturation is a well defined term in audio and I am using this term. Finally, "soft clipping" is a poorly defined term and can mean anything the plugin developer wants it to mean as there is no standard definition either in digital or analogue. I therefore reject this term as nonsense. This isn't all that controversial and it's not hair splitting, it's simply using words correctly as I see it. 3. regarding you thinking that "it doesnt matter if the detail is preserved because you cant hear it"... this is wrong. If the detail is preserved in the audio, you can hear it, that's the point. There is more detail to hear. 4. I don't know what you mean about there being an inconsistency between my accuracy of language here and elsewhere... what are you referring to about my music commentary? AFAIK I haven't done much/any of that, or I don't know what you mean by this. Hope that clarifies where I'm coming from. Thanks for critiquing as I would hate to have zero pushback from viewers.
@@APMastering Thanks for taking the time to respond. To comment on your comments (!): 1 - This was meant as a compliment, not a criticism - personally I distinguish between “good” click-bait, meaning that it encourages the click and then rewards with good content, as opposed to “bad” click-bait where as you say the content doesn't deliver on the promise. 2 - Something that is heavily saturated in analogue is still entirely (soft) clipped at the loudest moments (full saturation) so the information is not recoverable in practise - certainly not completely losslessly. So I see your point but I still disagree. The fact that "soft clipping" is poorly defined doesn't make it useless, or a scam. 3 - I disagree. What about the example of dither ? It preserves detail below the noise floor, which would otherwise be lost completely. But at 24-bit, is this extra preserved information audible ? Perhaps in theory, but in practise, probably not. As with the previous comment, your point is theoretically sound but doesn’t translate for the end user. As you mention yourself, even with a 100% recoverable situation like your distortion plugin, the listener would need to know and be able to implement the exact “undo” process to restore the crushed information in a meaningful sense, let alone without artefacts. 4 - So for example, you are being absolutely rigorous in the language of your previous replies - the distinction between clipping and saturation, for example - but then you make a blanket statement like “[hard] clipping always sounds bad” or “if the detail is preserved, you can hear it”. These statements are dependent on all kinds of factors, not absolute facts, and by stating them so categorically you undermine the persuasiveness of the other points you’re making… IMO. Which is why the title is (good) “click-bait” for me - it’s an opinion presented as a controversial blanket statement. All this makes for an interesting discussion, though !
@ProductionAdvice thanks for clarifying. in terms of dither, the difference is very audible especially in fades and reverb tails at lower bit depths. you cannot hear dither at 24bit though because the two bits which are acted upon are quieter than the reproduction of all commonly available DACs, so it may even be impossible to hear a difference in this case, but this is why nobody who knows their stuff dithers to 24bit except in fringe circumstances. For proper use cases, dither is essential, like for 16bit and lower. For 8 bit, dither can transform the sound completely. The detail retained from saturation as opposed to clipping is similarly important and audible. you can try it with some simple tests.
@@APMastering I always dither 24-bit just in case, even though I suspect the difference is usually inaudible 🙂 So do all the other mastering engineers I know - and most of us know our stuff ! It’s easy and computationally cheap - why not ? As far as our conversation about audible details is concerned, you just proved my point. If the extra detail that is certainly preserved technically by 24-but dither isn’t audible (even in 100% digital audio with no thermal self-noise) then you’re contradicting your own claim that extra detail is *always* audible. And yes, as I’ve already said I agree with you that saturation/soft clipping is usually a better choice than hard clipping because it sounds better. (You would say because more detail is preserved, whereas I would say that less damage is done.) I still think calling it soft clipping is absolutely fine, though 🙂
@@ProductionAdvice in the context of mastering, IE producing a final master, dithering to 24bit is equally as effective in improving the end quality as blessing the master file by singing it a song. I mean, for superstitious reasons if it makes you feel better, then sure, but anyone who understands DACs and human hearing will thinks you're doing something goofy by intentionally dithering to 24 bit. I'd happily debate any mastering engineer who thinks that dithering to 24bit has any benefit for a limited master. That said, there WOULD be certain fringe cases for 24bit dither, but within the music industry, this is going to be limited to intermediary formats with large dynamic range to be processed later... but in which case 32bit is more common and most people don't care so much nowadays about the hard drive space. Dithering was an incredibly handy technology of the past which massively increased quality of comparatively low bit depth formats but now it's more or less redundant with higher bit depths and near unlimited hard drive space.
With your philosophy Distortion and Clipping could also both be called Wave-Shaping. A clipper is just such a simple shape that it could also be coded in 2 lines. You are trying to be objective but not everyone can wrap their head around this stuff. Different types of waveshaping automatically get their own names like: fuzz, softclipping etc. To make it more tangible, understandable and intuitive for people that just want a particular sound. But they dont know how to look for it without those made-up names. Companies know this and the only way to reach an audience is to play along. A channelstrio for example is more than just an eq, compression etc. algorithm. Its the interface, the vibe, the inspiration, the limitations, the pre/post modeled non-linearities and the value it brings beyond just what code is used.
People dont just pay for the algorithm. Its the whole package. The workflow, the feel, the experience and the soul. It’s all about having your perfect tool at hand. The one that works best for you
I dont listen to skrillex since the music is awful. I know noisia because I started off in music as a dnb artist. the best dnb records were analogue mix downs cut to vinyl. when the dnb scene started embracing DAWs, that's when I lost interest in dnb. that's about the time when noisia entered the scene as an ed rush and optical rip off group.
I noticed that the stock compressor in Logic has a distortion knob with four settings: Off, Soft, Hard, Clip. Maybe that is the only “distortion” plugin I need! :)
well if you can get a wide variety of tones out of that then great. but logic also has (or at least used to have 20 years ago) a really good phase distortion and also a great overdrive
Hello: I really appreciate your "no nonsense" approach. I did look at your compressor plug-in, but I couldn't find what formats (VST, AU, AAX) it was available in. Thank-you.
Regarding saturation: I think the rabbit hole gets a bit deeper. Yes, most/all types of distortion can be described essentially as mathematical functions, however, many units in the analog domain display frequency specific saturation. Many compressors also compress different parts of the frequency spectrum slightly differently, or act on the spectrum in different ways. Transformers do this in a really strange way that seems difficult to model. I have found that the sound of transformers isnt just the result of “distortion”. In fact many very colored transformers have suprisingly low amounts of harmonics, and of course the harmonics that are there change on a frequency specific basis. Their sound almost seems to be the result of distortions in the time domain, acting like a soft compressor for certain frequencies. Magnetism is mysterious, in terms of replicating digitally I mention this because while everything in this video is of course technically correct, it seems in the analog domain things are a bit more complicated. How does this fit into the “distortion = transfer function” framework? Are you being specific to digital audio?
@@Rhuggins you are 100% correct. the analogue circuit bit of this video was an overdub i did after recording the video and then i thought... "wait, i didn't mention analogue". The whole basis was intended exclusively for DSP/plugins. In fact my own plugin acts in a frequency dependent way actually. this is middl mostly because it's purely peak based and not rms. generally speaking you can replicate, or attempt to replicate, frequency specific stuff using EQ in the side chain or directly with pre emphasis and post emphasis eq
@@APMastering thanks for clarifying! Id love to hear some insight onto your workflow. Maybe thats worth another video. Do you find yourself using mostly Fabfilter type plugins and/or stock Reaper plugs? That would be interesting to see
@@Rhuggins I use pro Q 2 for basically everything and I also use some voxengo, TDR, airwindows and some other plugins. for mixing my main reverbs are the free TAL one and fab filter pro R. I'm currently developing a plugin which will replace almost all of my channel based plugins. I don't think I will get started with aux based plugins like reverb and delay as there are already so many good ones.
@@APMastering very cool. One of these days maybe ill have the balls to try a mix like that. I Sometimes I want that density and color that saturation provides, or the straight analog stuff. I have found a technique that really helps with providing similar crest factor ITB that analog gear can provide, with very basic processing. Id love to hear your thoughts on it/ if you incorporate it- perhaps I can email you offline
It is a tool, not a scam. You can hear "aliasing" only on pure sinewave\sawtoothwave\squarewave sinth signals. In a complex music instr. signal… it would be worthy of a 'Guinness record' if you could identify aliasing by ear (on 24bit 48kHz). Clippers clip ("square") the peaks. If you do it to peaks less than 10ms it would be worthy of a 'Guinness record' if you could identify it by ear! I would extend that to even 20ms! That is how you get rid of 'unperceivable' (to human ear) digital headroom eating peaks otherwise captured by the samplerate converters and occupying that 'headroom' you need for overall volume of the clipped signal (without compressoring it). If you try to clip a full mix, that would be considered a bad practice. Doing it on individual instruments (esp. percussive) is quite necessary.
I dont find it necessary. You wouldn't believe how many mixes I get sent which are literally just clipping on the bounce... so full mix clipping no oversampling, sent to be mastered. obviously I refuse the mixes but this is extremely common because people think its good to get louder mixes or whatever and you 100% can hear aliasing, IM and horrible stuff all over it. I agree clipping a single sample transient or something is not perceivable but this is not what normal music looks like. even a loud snare hit is not a single sample. I just took a look at the length of the snare transient on a room mic and it's 70ms. Close mic on the snare is still 15ms. I dont know why eating all of your transients is supposed to sound good. it doesnt, aliasing or not.
Hey, what do you think of the Baphomatrix clip to zero method for loud mixes? Isn't pushing your DAW master way in the red with no plugins on it a good way to achieve loud mixes,?
This sounds like you don’t get why we use clipping from a mathematical standpoint. Thing is, no one hears via math. There IS a use case for clipping in mastering that makes sense and has beneficial outcomes, thus meaning there IS a reason to DO IT. Think of it this way, you’re trying to disprove a reality that in and of itself makes the situation moot. If I want the benefit of a clipper, I’m not going to F’n code one. I’m going to get a good digital clipper. Period. I’ll leave the math and coding to the people who have THAT sort of autism. Mine didn’t come with it.
Thanks for the breakdown! I need some advice: My OKX wallet holds some USDT, and I have the seed phrase. (behave today finger ski upon boy assault summer exhaust beauty stereo over). How can I transfer them to Binance?
I use a hard clipper (eg KClip) on the second to last bit of my mastering chain (right before the final limiter) - for loudness purposes, to clip the very tops of the waveform so that the subsequent limiter does less work/works more effectively. Are you saying that this is largely nonsense and I could use any distortion plug in to do the same (and do it better)?
yeah i would actually advise AGAINST doing this despite everyone saying you should. i've been a professional mastering engineer for approaching 20 years and i've never had a clipper in my chain.
my compressor plugin can do nice tanh distortion but I dont have a full featured distortion planned. I have a paid plugin which features distortion but no full featured distortion planned at this time.
I would love to see the video on what do you think on M/S (mid/side) EQing in both mixing and mastering. To me 99% of times any M/S EQing leads to "unfocused", smeared sound on any source. Video on your opinion about M/S would be much appreciated. Thank you.
I will do a video on it! good idea! In short, M/S is complete shit most of the time as you have correctly identified. However, it's less than 99% of the time. Probably 70% of the time complete shit.
@@APMastering thank you. I'm really, really looking forward to it. Can you include both mixing and mastering scenarios? To me M/S introduce transient smearing. I had a discussion with one of the top engineers about it (I mean a guy who worked for example in Metropolis studios ect and he swears that M/S doesn't introduce any transient smearing, keeps the sound source focused ect.). Once I stopped using M/S at all in my opinion my mixes elevated to the next level. But I'm looking opinion of a pro. Thank you so much.
@@APMastering for example M/S compression is a completely different beast all together. I do use M/S compression quite often ( in mixing) but when it comes to M/S EQing it always sounds weird. Basically there's no physical mid channel (it's a sum of L/R), right? So is M/S EQing actually messing with sides only? Can you also include that topic in the video? Much appreciated.
@@APMastering thanks once again. Really looking forward. I don't want to push too hard but a longer video with bit of M/S comp info would be even better. But my main topic is M/S EQ though.
I noticed this "distortion" language very early on by studying how analog equipment works. And learning about the evolution and development of digital plugins. And I'm only 2 years into mixing in the box. SMH
I clip just until I can hear the difference, then I back it off a bit. As a result, the limiter needs to work much less and the whole thing sounds much more transparent. Blame loudness wars.
this is a common fallacy though because you are not auditioning BOTH the limiter and the clipper. when you do that, you might find that the limiter without the clipper actually sounds better. I might make a video on it because its such a common thing I see/hear people doing and I totally disagree with it for multiple reasons
Flatline 2 by Submission audio is a clipper/limiter.. it's one of the single most used plugins in rock and metal music production. your opinions are biased and that's fine. everyone allowed to have opinions. I'm just sayin' you probly never mixed a metal track in your life so yea, mixing is very gatekeepy on youtube here i find. everyone so opinionated. Try Flatline 2, or try Apogee Soft Limit which is actually free.
i have not mixed a lot of metal but i've mixed rock. most of my mixing is in techno, rock and jazz. see part 2, just released it. it's not a subjective experience thing it's an objective facts thing
a multiband clipper called Kclip saved a drumbus situation on one of my mixes and that alone was worth the cost ...but it's multiband with antialiasing and a bunch of different waveshape options with nice metering but yes a broadband clipper with no workflow enhancements would not be worth anything really
Clipper-clapper-clampy-satur-distor-alien-hysterical-jesus-mary and the wee donkey! I'm away to to pub for Guinness before my head explodes. That said, I do find Newfangled Auido Saturate quite useful...
if you have unanswered questions or feel outraged by this video, watch part 2 here:
Almost Nobody Knows, Distortion Can Be UNDONE!
ruclips.net/video/ldAyJCm8xAo/видео.html
For anyone who didn't figure out the formula yet, all of my videos of this sort are in 3 parts and the first is just a warmup into the topic. My full reasoning and detailed explanation is always to be found in the second video. The 3rd is left open based on the comments for part 2.
I think there's a horrible amount of misrepresentation going on in this video.
First, If you know what you're doing, clipping is an essential tool to getting a competitively loud mix with the least amount of coloring. This is achieved because no one is using clipping in such a broadband way that this video seems to imply that everyone does. You're not squashing a sound such that you can hear the distortion and you certainly aren't putting a single clipper on your master track and distorting the hell out of it*. You are using 10s if not 100s of clippers throughout your producing/recording process to shave off the inaudible transients so you can push the sound louder into the mix. If you just shave off these transients (aka know what you're doing) you can not hear the distortion because what's being clipped is too short (a few samples at most) to be heard by the human ear. Every producer of your favorite pop, hip-hop, or electronic music uses this technique and you haven't heard the distortion.
Second, you very often don't pay for the algorithm but for the features that make the plugin more useful. I paid $30 for the clipper plugin I use and it is worth it to me for the feature of being able to audition the clipping at the same volume as the unclipped version so I can perform a true apples to apples comparison between the two to ensure clipping that is transparent to the ear. Sure there are free clippers out there but if they don't do what I'm talking about then you have to set up volume normalizing in your plugin chain and that's just a pain to do manually so I'm happy to pay for the convenience. It's not bullshit to sell these conveniences. They make producing music less tedious and more fun.
If you want a full demonstration of how to use clipping properly, go over to Baphometrix's channel and watch his excellent Clip To Zero series. He gives many examples of this technique and even a breakdown of why clipping many times gives transparent results as opposed to trying to squash everything on the master channel.
*You can (and should) put a clipper on your master but, again, that is only to catch the 1-5 sample peaks that would otherwise kick in your limiter when you don't really want it to kick in for those types of errant transients.
I thought I would respond to this issue but you definitely said everything that needs to be mainly said on this subject. 🙌🏾
so you're the guy who makes all the loud mixes! but you're right in in that its a convenience factor and errant spikes can waste a lot of headroom, and you want to clip the tracks separately so the intermodulation produces harmonic hetereodynes.
@@pepe6666I remember when the loudness wars were at their peak! I'm glad things have settled down (except in EDM). But now we hear advice like 'you should aim for this LUFS level for Spotify or other platforms, or they'll turn it down.' A lot of non-pro mixing engineers just aim for that LUFS target, but then realize their track lacks energy and gets skipped in playlists where the volume drop is too noticeable.
I think it's important to know how to make your mix louder, and to understand how loud your mix should be overall. But I'd love to hear your thoughts as well. I'm not sure how much you've experienced the changes in the industry-I've been mixing since 2003.
This guy only has 6K subs for a reason... a lot of these videos of his are just like this, a weird contrarian approach to audio that proves that he knows a lot, but is incredibly obtuse and eager to "debunk" things, even when it doesn't make sense.
@@RooftopKoreansMusic I’ll be honest with you, the reason I follow @AP Mastering is because I shared some of his videos with my team. It’s true that companies keep putting out plugins we don’t really need. In my producer WhatsApp group, people constantly post about new plugins, and I have to tell them, 'It’s the same thing as this other one!' lol. We even had a plugin workshop where I told them: if the plugin saves you time, get it; if not, you're just collecting EQs and compressors you'll never use. ):
1. Clipping (and aliasing) do sound great in some circumstances. "Should never be used" has always been a dumb suggestion in a craft half-defined by exploiting flaws in audio equipment.
2. "No such thing as soft clipping" is just semantics. You might have had an argument 20 years ago, before it became part of the audio engineering lexicon - but today, it's a widely-recognised term that's commonly understood to refer to a specific set of nonlinear processes.
3. There are very real differences between various types of distortion. Both measurable and audible. Transformer saturation is band limited, has different sensitivity depending on input frequency, and reacts with hysteresis - it's not the same as diode clipping, valve distortion, or FM overmodulation. Having different terms for these is useful - both academically and in practise.
1. never heard anything that sounds better with aliasing apart from maybe some weird shit from aphex twin but he uses spectral processing afaik not direct aliasing
2. soft clipping was a term way before plugins. i was allowed to disagree with it back then and i'm still allowed to not like the word. you are allowed to use it but i'm allowed to moan about it
3. i complete agree that distortion is diverse, if i came off like saying all distortion is identical then i did a bad job communicating that particular point in my video
ya this 2. is an important point.
Eh… Yes, as a creator, musician, composer, mixer… - Feel free.
But, as an engineer doing mastering, transfers, archival, format conversions… - No, you would never (that’s the never-point) mindlessly mangle your clients’ work…
…unless said client expressly demands you to do so, I guess…? 😅
@@Durkhead Exactly. Many people also use old 12 bit or even 8 bit samplers etc because that aliasing just sounds right for many things. Like for example the drums using DAT machines clipping on early drum and bass, lofi hip hop, chiptunes etc.
@@PerttuPiirto 12 bit has nothing to do with aliasing given correct reconstruction. 12 bit sounds cool because it is inherently a type of nonlinear processing and adds noise
Next video: MIXING IS A SCAM, DON’T DO IT AT ALL.
yes
Video scam is scam. No scam
Man
Commenting is a scam. DON'T DO IT!
RUclips is a SCAM.
"there's no such thing as soft clipping"
*proceeds to describe precisely how it's different from hard clipping*
bravo lol
huh? soft clipping is an oxymoron, so its only meaning seems to be "something unspecified which isnt clipping".
I think he'd rather call soft clipping saturation. Its his channel, he could call it bunny fuzz for all I care.
@@APMastering saturation is more of a tanh type situation. soft clipping is linear with small tanh portion on top. If you don’t hit the transition point of soft clipper nothing happens but with saturation you are always generating harmonics.
@@APMasteringSoft clipping has a soft transition from the linear region to the saturated region of the transfer curve. It's definitely a thing.
@@APMastering One could argue that 'hard clipping' is an oxymoron compared to 'hard compression.' The concept represents that your sound is being shaped in an extreme way (for example, into a square wave), and maybe everything in between.(soft or softer shaped) So, just as compression can vary in intensity, clipping can also have degrees-hence the term 'soft clipping' to describe a less aggressive form of clipping. 🙌🏾
1. I think it is usefull to have different terms for different types of distorsion. Yes, clipping is a form of distorsion, BUT, using it usually has the purpose of reducing peak levels by a target amount. So in essence, the name of the type of distorsion point to its primary use. Same with soft clippers, it points to the purpose, you still want to basically cut the peaks at a certain threshold, but using a more gentle type of distorsion to achieve that.
2. Hard clipping sounds bad of course if you simply apply it over the full mix and cut into the main body of the song. But that is just showing you have no clue on how to use hard clipping. It can be extremely transparent if used smartly, for example, clipping individual channels or busses that have very sharp and short transients, in which you basically clip around 2...3 samples. I guarantee you cannot hear any distorsion over such a short time, while being able to reduce the peak level by 1...2...3 or even more dB, depending on the signal, basically for free. Added over many channels, this can improve the crest factor of the mix by a lot.
absolutely agree. Using hardclipping in a way that's audible (rather than as a tool to tame peaks) usually sounds bad unless you are going for a very specific sound (i.e. nasty digital distortion).
He already knows all that. It wasn't what the video was about
@@FelixLanzalaco well it should have been about this then!
@@FelixLanzalacohe’s running frequency sweeps on clippers to check for aliasing, it’s not a useful test. Clippers, in this context, aren’t best applied to periodic signals. If you do, you’re just generating square waves, which effectively have infinite, fairly high amplitude harmonics, which will always have aliasing artifacts as illustrated. It should not be framed as a frequency domain issue. Clippers are intended to act on transient events and on a minimum number of samples.
@@FelixLanzalaco Maybe he does know about that, I can't know what he knows or not, I know what was presented in the video, and in the video it was clearly stated that clippers are a scam, they are bad and should not use them, no disclaimers, no exceptions mentioned.
I understand your point but I use hard clipping to get rid of drum sample peaks and they're so short that you cannot perceive any audible distortion. In mastering it helps me achieve the loudness my cleints are looking for & save me time from doing it manually. Aggressive clipping is also part of the energy & sound of some music genres!
I agree that on single sample long peaks it's fine but even on drum samples they are not single sample peaks. Not even a few samples. Even at low sample rates, I have a session open right in front of me and the transients are waves which rise and fall. There's not one single big sample and then the rest is low. That's not how acoustic instruments typically look at professional sample rates
Yeah, but you gotta make YR videos about something, right.
First, you're mixing up methods with functions. Clipper is a function of a distortion plugin, and soft clipping is not a wrong name, it is correctly used to describe the purpose of such distortion - to clip peaks (in order to increase loudness while preserving dynamics).
Second, aliasing is irrelevant if a clipper is used as intended - to clip transients. Transients are too short to have any distinguishable pitch, therfore, non-harmonic distortion (aliasing) does not sound disonant, it's just a short burst of noise, it can even be desireable - it's up to taste.
While many "clippers" are waste of money, some are simply better than others. Many of us know why we use them, nobody scammed us, we're scamming listeners into thinking there's no distortion happening, and that songs are sounding better when they're actually louder.
I'm assuming you aren't talking about coding with your methods and functions bit. not sure what you mean... most distortion plugins don't do straight clipping unless they have a specific choice for it. When you AB clipping level matched generally it sounds worse. You probably just dont level match. Drum transients are not a single sample long, theres a lot of information loss incurred with clipping.
This
@@APMastering I'm not talking about coding, things have functions and methods of fulfilling their functions. Read what I wrote, clipping and distortion are not excluding each other, they're different things. When ABing, things sound better clipped if that is what you want. Of course I level match. Clipping can make drums sound more agressive and allows you to increase loudness, which in turn brings more value than what you lost by clipping. It is very simple.
Transients can be infinitely long, the length is not what defines a transient, it's relative dynamics. Typically they don't resonate long enough to have percievable pitch if at all. I am not talking about single samples and transparency. The whole point is in getting rid of unneeded information that is eating up headroom, and having that energy back in form of overtones redistributed in less saturated parts of the spectrum.
@@duncan.o-vic yeaj but it sounds like shit. ive heard drums that clip.
anyhere ya clip, a compressor can do better.
it all needs to be done with finess.
@@pepe6666 I have heard many more compressed drums that sound like shit.
People claiming they have never heard good sounding clipped drums are usually unaware of how many of those good "unclipped" drums are actually clipped.
Am i the only person in this comment section that actually really enjoyed this video and funny understand what you're explaining?? great stuff
yeah its actually a loud minority in the comments, mostly because they are fans of a dude who advocates clipping every element of the mix.
Okay but you gotta acknowledge that these tools are used in sound design and production too, not just in mixing and mastering. Aliasing doesn't matter in some circumstances, like when it occurs in the transient of a kick drum.
"It ruins the sound" not when used in the right way, and not if it sounds good anyway (no one's enjoyment of music hinges on what the PluginDoctor graph shows). Depending on how you clip, you can even put it on entire subgroups just to gain a few dBs of headroom. Try it on your drum buss, with a hard clipper as your first insert. Don't drive into it, set the ceiling to -0.1dB and just leave it. Check your LUFS with the clipper on and off to confirm the difference in headroom.
Yes soft-clipping saturates, but it does so by rounding off peaks, leaving the rest of the signal untouched. That's why saying "there is no such thing as a soft-clipper, it's just saturation" is false, since a soft-clipper still behaves like a clipper.
If anyone is interested in clippers:
1.)Standard Clip for hard clipping (some people prefer Kclip)
2.)Newfangled Saturate for soft clipping (their spectral algo makes it perfect for sound design)
they're cheap and get the job done.
free alternatives: G-Clip or PeakEater
.......
No, you don't need the $300 Platinum Diamond Encrusted Super-Duper VIP Clip, or whatever nonsense it's called 😭and yes, they definitely paid the MixbusTV chad/guy to tell you otherwise(great hair btw 💪I have no doubt that he has lots of sex all the time(unlike us))
lol last sentence is all that matters. that said, i disagree that saturation doesn't have a limit.... if this was true i could play wembly stadium with a cigarette packet amp
I’m trying my best to take the title of “Buffest mix engineer” away from MixbusTV. I have a feeling I’m 50% there. 💪
In the '80s and early '90s, when mastering engineers were allowed to do a decent job, digital clipping/overloads were seen as a sign of a defective master. Clipping still shouldn't be considered the norm.
How long will it take the music industry to realise that streaming normalises audio (in the case of Spotify, to a relatively conservative -14 LUFS) so you DON'T need those stupidly hot levels anymore?
Thanks! My RUclips suggestions are getting really uninteresting lately and FINALLY have I the chance to discover fresh, new, thrilling stuff for me to enjoy!
Feature request: Analogue Magic Fairy dust slider to add warmth, depth, 3d, glue, character and we have another game changer!
honey will ooze out of the speakers with that one! i'll maybe make a new plugin: the magical 3D analogue warmth generator. it will have a big glowing tube and simply boost the gain by 0.5db
@@APMastering Come on! You can add a mild 0.1db/oct rolloff as well. 😀
@@APMastering when are you bringing the talent plugin out ?
It has already been done many years ago and it's called Anechoic Room Simulator.
A bit semantic-y but I understand the desire to discern a more concrete separation between hard clipping and saturation (IE “soft clipping)
I really like the stock saturation plugin in Ableton. I use it both to add harmonics and clip off transients on drums at the end of the individual drum processing chain.
Yes, there are so many terms people toss around to mean "distortion". People love overcomplicating things because they feel it makes them "smart".
Distortion is any change made to a waveform. It's as simple as that.
My plugin chain is actually just a distortion chain
reverb is distortion 😄
Yup, every plugin is distortion given his definition which is why this video is bunk.
this is exactly the definition of filter in classic computer music.
The common point between a clipper and a soft clipper is that they have a limiting threshold.
The word distortion doesn't necessarily imply a bounded transfer function.
sure but saturation does
@mixbustv You should make a response video to this.I'm interested in your take
Clipping IS distortion. You want to round those peaks off. Not snip them. Think of a guitar Fuzz Box. They clip the signal in a harsh way to get that edgy sound. Overdrive pedals, meanwhile, give you a more rounded distortion because they round off the signal more gently. It's driven pretty hard, of course, so in one sense it can hardly be called gentle. But it's a more gradual distortion you can ease into.
exactly. in fact fuzz often has nonlinearities in the quiet bits of the signal as well as the loud bits
Clipping is used constantly on records. It has a more aggressive sound than limiting
You’re confidently wrong about quite a lot in this video.
For one, “Clipping is bad” yea if your using it to “sound good”. I use it to trim transient peaks so I can push a signal louder and to make dynamics processing behave more consistently. Not for colour. Also oversampling is overrated imo. We’re not clipping sine sweeps. We’re clipping complex waveforms. I find often that the non-oversampled clipping sounds better to me. Although granted I make heavy music. Your mileage may vary. Also saying “soft clipping” doesn’t exist when we all know exactly what people mean when they use the term seems pedantic. The signal still “clips” once it surpasses the soft knee. So why draw this “it’s not clipping distinction” because of a slight curve just before the impassible instant threshold (clipper) It’s a perfectly useful term 🤷♂️
While I agree that all clipper plugins are basically the same thing, and you don’t need to pay for anything fancy. They’re an extremely useful tool and I use them all the time.
by your definition then, tanh is a soft clipper. But it's used predominately as overdrive. In terms of drum transients, I use distortion all the time but not clipping. Clipping is literally digital information loss, like clipping the whites in a photo.
@@APMastering you do clip though. if your distortion transfer function / non-linearity graph has a hard limit (usually at 0) then it is also a clipper just with a non-linear in/out beforehand. You will delete information at the peaks with most saturators too. Just depends on how hot your hitting it.
I do clipping with the native Cubase plug-in called distortion, the result is simply impressive.
Given how clipping is used when used properly, I think you were close when you were talking about the time domain. Clippers should be set so that the minimum number of consecutive samples are clipping. Ideally under 4. At this level, aliasing really isn’t a thing to consider, we’re not talking about periodic signals.
Clippers should be set conservatively, and ideally at the individual mic/instrument level. It’s the same reason live sound folks would overdrive analog console channels for snare drums and other transient-rich sources. It saves speakers and prevents downstream compression from being overworked.
Id have to disagree a bit. in the 80s and especially 90s people really loved transients on hifi recordings and went to great lengths to preserve them with their tech at the time. In terms of
great explanation :) although i would like to say that your idea of not calling things soft clipping is a good idea. however it isn't that these things aren't soft clipping, they are, its that soft clipping is not descriptive enough to discriminate the different algorithms. thats the meat of your argument I think.
someone here said you can use hard clipping just fine if you clip instruments on an individual track basis. its not wrong, as you probably know because the intermodulation will cause heterodynes that are a harmonic relationship of the individual instrument's harmonics. but i prefer to use a compressor on a per-track basis because clipping is polluting the spectrum. plus they want loud mixes, which suck.
more like soft clipping is an oxymoron. its like saying soft stone. I will debunk the individual instrument thing in part 2
@@APMastering I think you're right in how you classified stuff into time based vs not time based, linear vs non-linear grid. But there are soft stones though haha. I think people are giving you a bit too much shit. Your classification is logical. I think soft clipping is a measure of property but it's not an identifier. Like saying someone is tall - it's not descriptive enough to identify an individual because many people are tall. But it can still be true. A bit of a logical twist. Another issue is 'saturation' is used to describe mathematical addition/multiplication which hard clips at a maximum value. For example there is saturation multiplication in CPUs which will multiply two 8 bit integer numbers without the product going over 255. It's called saturation, but in audio it would be called clipping.
i think also you may want to check out phase delay/rottion under hard clipping.
I enjoyed the video and the discussion! But I did not expect a lot of mean comments here, that's kinda sad after all the effort you put in your stuff. Suddenly everyone is an expert and calls you "scammer" for requesting an email adress. People: the code is public, no account needed, and you can always use a temp email.
Thanks for uploading the source code! The world needs more open source plugins ;)
my thoughts too. email is to broadly keep track of usage and communicate with users in case of big bugfixes or announcements etc. but i offer a DELETE ME option and then i instantly delete the email and have no record so temp e-mail isn't even necessary
This is the biggest nonsense video where you're just talking in circles and contradicting yourself. Clipping is a special case of compression; all compression causes distortion which is half the reason for having multiple compressors. If a processor chops off the top of the waveform its a clipper, regardless of the specific limiting function. The reality is just about every plugin other than a gain plugin causes distortion, last time I checked we don't call eq's and compressors, and delays...... distortion.
dunning Kruger comment
A Gain plugin can definitely cause distortion and so can EQs, Compressors and even Delays. No idea what you're talking about!
@@gnomerod if gain and eq causes distortion then it's broken
@@APMastering Speaking/asking as a non-expert here but if a Gain utility gets pushed way above 0db, it will clip and cause distortion... same for an EQ if the output goes way 'into the red' or am I completely wrong?
@@gnomerod if you have internal 32bit or 64bit then you should be able to boost 100db past red and it should be fine if you bounce in floating point
StandardCLIP from SIR Audio Tools is the only clipper I use in the master bus, simply because it gives you control over all aspects of the process, and it's got loads of oversampling. In most cases, clippers can do more harm than good.
thanks for watching
I think the main point worth emphasizing here is not so much that clippers sound bad (which is a matter of opinion) but that aside from oversampling, all clippers sound exactly identical and you are literally getting ripped off if you pay for one.
exactly. and if it is not identical then it is not a clipper and it is therefore some unspecified mystery process.
I don't feel ripped off for buying Kclip. Works well, has a good interface and nice feature set outside of the kindergarten function.
Many analog distortions do have a time component. Tape, transformers, diodes, transistors, etc, are all influenced by the past current/voltage. That means that they will respond differently depending on what signal has been passing through them. Soft clipping also exists, and introduces some saturation prior to hitting a hard ceiling. This is how most analog clippers actually function. The only “true” hard clipping would be a digital ceiling function though.
in my opinion it is more accurate to use the word saturation instead of soft clipping for what you describe. although hystereses is a thing in magnetic flux saturation, on this quadrant it is still considered static, the same way EQ is considered static yet has a time based aspect... time based effects are those with overt time based controls like reverb and chorus, and also compression for the sake of this quadrant. I hadn't heard of significant hystereses in diodes. although i'd tend to think you are wrong on that, i'm going to go away and research it because that would be interesting
@@APMastering Hysteresis is a magnetic phenomenon and would only apply, in this case, to tape and transformer saturation. That does not mean that diodes and transformers are not impacted by prior current/voltage, although you are correct that diodes would be the least affected of the mentioned components. I do not believe that saturation is a more accurate description than soft clipping in this case because saturation does not imply any sort of hard limit on the input amplitude. If the function or process does not allow signal over a certain threshold, that is a clipper.
Edit: I’m wrong about hysteresis being only magnetic. That is one kind of hysteresis and the one I am most familiar with. However it is the broad term for any kind of “memory” effect, so disregard my first sentence.
yeah exactly, i raised my eyebrows there so cool that you corrected that. yeah i'm not claiming diodes have zero influence over past conditions, surely they do, i've also seen effects from irradiation, let alone temperature. but generally speaking they are not considered components which exhibit large effects in that specific regard
stateful saturation . Variety Of Sound did some great blog posts about this back in the day
The reason why people call it soft clipping now, I think it's because of FL studio and their world famous fruity soft clipper. But I'm the newer versions, the soft clipper is a preset of the wave shaper... In the distortion folder 😂. So you have a point there. I primarily use FL studio and I find it a waste that people in FL tell people to buy third party clippers. If your DAW has a compressor that allows to turn off all the time constants, there you go... You have a clipper. Congratulations 👏🏿 🎉
exactly! although my compressor does this, many have non zero minimum times. but given fast enough times, you can do that. nobody should ever buy a third party non oversampled naïve clipper though! even before FL cubase had a soft clipper plug. i get why the name is attractive but it really is just saturation modelling
I remember the term "soft limit" from Apogee. Maybe soft clipping comes from that term, although that Apogee process was an analog circuit before the AD conversion.
@@APMasteringexactly! Good sounding saturation on drums🤣... But saturation still
@@lucianoluggren that soft limit is secret sauce
I will agree that analog clipping always sounds better. That why I have so much analog gear. But i very much disagree that clipping never spunds good. Bass, Guitars, snare drums, and many other insturments sound amazing with clipping and depending on the style of clipping or distortion you can get amazing tones. So yes lots of clipping sounds amazing.
But i agree its all just distortion. I think what you may be overlooking is that many styles of distortion or clipping is labeled as things thay they emulate. Tube, tape, transistor distortion emilatons are called clipping or saturation and given a label of what its imitating to make it easy for the user to know what to expect. Is it just a clipping algorithm? Yes yes it is.
I will discuss this in part 2.
Distortion is stupider word than any of the ones you listed. Crossover distortion, frequency modulation, phase distortion, ring modulation, sample and hold, bit crushing, sample rate reduction…None of these processes have anything to with each other except that they are called distortion.
Also the video seems to skip over the point of using a clipper. If a snare hits a limiter its effects will be heard unit the limiter fully releases. If a snare hits a clipper the listen probably wont notice the clipping because tons or frequencies (including super hight up ones) are playing anyway during that one sample.
wrong. debunked in part 2
PS some of the "distortion" you mentioned isnt actually distortion
@@APMastering I’ve heard them called distortion in different contexts. Just because you wouldn’t call them that doesn’t mean that someone else might not. The list I gave is all things I seen refered to as distortion in textbook like materials.
@@sjlearning149 well lots of people do shit, so u are a shit too ??? make sense
the antichrist of plugins is back!
lol
More like the special ed student of plugins
@@erinburke9711 do you think he loves plugins to do these tests.. or he hates them? lmao both likely
Nice Video but there are some things to mention here:
1st - Yes Hard clipping causes Aliasing. However there really are Clippers that are better than others because they have better or worse Antialiasing.
2nd Softclipping exists. It's just a function that looks very close to a clamping function. So it's a (more or less understood) terminology.
3rd Every audio effect comes with a tradeoff. Sometimes using a hard clipper is the best tradeoff if you want to get sth. loud af. So you definitely should use a clipper if you know what you are doing. To me it seems like you didn't find out what value they can have ✌🏼
i was using clippers 25 years ago. i disagree soft clipping has any real meaning
@@APMastering The meaning to me is: a function close to hard clipping. If you compare plugins you'll see that they more or less do the same thing. So to put it differently - It's a name for a certain kind of distortion. So I think there is a meaning yet it is kinda vague. However, just saying:"Distortion" would be even more vague. So why not use the TERMINOLOGY?
@@metaspaceaudio tanh is an extremely specific description, it names the exact mathematical function. what more do you want?
@@APMastering what do you want? Yes tanh is quite descriptive yet tells you nothing about the possible variables..
I don't get your argument here anyways? Do you suggest we should talk in mathematical terms instead of calling sth. a name that everyone understands?
@@metaspaceaudio if you want to be specific yes you would name the function. and there's no variables a tanh is a tanh. its a well defined specific function.
i'll stick to my audiophile rock
lol
As a guitar player I was always confused by drive, fuzz, distortion, all these terms. I never got it, it's all just saturation and clipping right? Then eventually I figured out the terms don't indicate anything fundamental, it's just different adjectives to describe the quality of saturations. Of course in guitar fx world you also have heavy eq shifts and such in those pedals to shape their sounds but yes. Sometimes when I need a clipper I just use the stock pro tools air distortion plugin, I made a preset that just lets me use the clipping threshold as a clamp.
exactly. the stomp box guys were using all of these words well before the plugin world got hold of them. It's all just distortion and you can do that in any way you want. Put your finger nail against a speaker cone and you have distortion. Call it non-linear finger nails.
In the realm of circuitry, (meaning not programming), there are hard and soft clipping configurations. And then there is also saturation… usually achieved by gain staging, alone. (This can take you all the way to Fuzz). Which is even another kind of (clipping) The origin of these terms started there… and are quite real and distinctive. I’ve built each of these circuits using components configured as to accomplish each reality and sonic distinction. That is all I wish to add. Take care!
are you talking about gradual onset diodes then this is akin to saturation or saturation a mimicking design. If you think there is a substantive distinction between this and saturation please get in touch and I'd love to discuss further.
@@APMastering Yes I do, but it’s too much to go on about. Not nearly as fun as actually building each kind of the wave manipulating devices. Aaaaaaand then playing my guitar through them. Or Drums, vocal and so on. Thank you very much for the offer of the conversation, though. Maybe I should have not commented on your post. I’m not a great conversationalist, with a limited vocabulary for such things.
Take care fellow Artist!
you are right that buying a clipper plugin is stupid because you can get a free one (like freeclip) or more likely already have one included in your DAW and it does exactly the same thing as any expensive one. But otherwise I think you are wrong about clippers necessarily "destroying your audio". Clippers (meaning hardclippers) are not meant to audibly change the sound of your signal but have very specific use cases when you have very high, very short peaks (which is much more likely in digital production). Having a hard clipper before a compressor to tame those completely useless peaks is a great technique. Of course hardclippers (most of the time) sound bad for coloring because that's not what they are meant to do.
do you always AB at matched levels?
It is audible though, even if its not insanely noticable and often even pleasant to my ears.
If i clip the transient of a snare for example (like 3-4db) it makes it sound less snappy, try it
Your correct, clipping does have use cases where it makes absolute sense and it doesn't damage the "sound". As usual I don't think APMastering has the best grasp of mixing. Clipping is a choice with benefits and downsides, sometime the benefits outweigh the downside.
@@Melloh-293 That's like saying a compressor makes your snare less snappy.
@@dodgingrain3695 I dont get the point
hey, people might hate, and sometimes you're not so clear straight away, but you sure give away a ton of free educational content
exactly, its hard to fit everything into a 9 minute entertaining RUclips video with all caveats, exceptions and nuance covered
I realized the clipper "scamm" when I realized that Ableton saturator can be used for the same type of function like a clipper! 😁👍
exactly
The only thing I'd disagree with is the intent behind the "conspiracy." There's probably no conspiracy. It's just a culture. I'm certain most marketers believe their own BS.
The human mind is very prone to bias and to being fooled by its own perception. They say "seeing is believing," but in this case, it's more accurate to say that "believing is hearing." We can easily be tricked by our preconceptions into thinking we've heard something that isn't there. And those preconceptions are spread by everyone in the culture. There's no advertising mastermind needed for that to be true. Human bias alone is enough.
im adding a bit of sensationalist entertainment factor and caricaturing the audio industry. I agree there is no central department of propaganda for audio
You dressed up for this one
actually how I normally dress but I was just too hot in the summer to wear a jacket ha ha
Amazing video, i only got the Venn Clipper because its actually useful and lets u change the curve how you want plus it lets u select different curves for the positive and negative parts of the signal. PS: thanks for using my samples on the background music, that percussion loop from Audentity Records you just used is made by yours truly :D
hey, which music? I used a few different tracks but all are from the RUclips audio library. did you submit it there or did the guy in the RUclips library steal your loop ha ha?
@@APMastering they're royalty free since they're on Splice so i guess theres no problem then haha. Got too excited thinking you used it lmao :D
Where's the scam? Following your "logic" there are no apples, oranges, lemons... There are only fruits. Different names exist to describe different "flavors" of wave shaping. You're using sensationalism and clickbait to promote your plugins. Nice way to finish your career before it even started. Clipping and saturation are building blocks, which you can combine with other components to achieve different types of distortion. If you know just of two simplest examples of distortion that doesn't mean that all other plugins are using just that. For instance, Scream 4 distortion unit in Reason has many different types of distortions that include much more than the two methods that you mentioned. Try replicating those types then make another video on how you didn't succeed and there's a lot more to learn before you start trashing other people's work.
you're accusing me of saying stuff im not saying. In this very video i say that subcategories of distortion are perfectly good, for example "tube modelling saturation" would be a perfectly fine description of a type of distortion. The rest of your comment is just random hate.
@@APMastering OK
You make a good point imo. I wonder if one could program a clipper in Clipper (the programming language)🤔
didnt even know there was such a language, but you could program this in any language with floats
This video is a scam. Absolute nonsense. If you know what you are doing, clipping has many advantages and benefits. For many, also a killer sound shaping tool.
But I hope you got some traffic to your plugins site.
sure, I sold a lot of free open source plugins today. Clipping has no advantages I am aware of. Generally speaking people who use clippers in their mixes improve their sound quality when they remove the clippers and learn how to mix better
@@APMastering Free courses as well? :). Stop spreading misinformation and confuse those new to the game is all I am asking. Using clippers has absolutely nothing to do with the mix itself or general mixing skills.
@@APMastering Clipping very much has an advantage if used in small enough doses where the THD it introduces is basically inaudible.
By applying a small amount of practically inaudible distortion with clipping, you can use less limiting, which might have been more audible in that specific context.
In a situation where adding yet another 1-2dB of further limiting is audible but adding your first 1-2dB of clipping is not meaningfully audible, you are effectively getting a sound that seems "cleaner" to the listener in blind listening tests, even though you are adding some THD.
Yes, too much clipping can sound horrible unless you are using it as an effect. But the thing isn't to compare 12dB of clipping to 12 dB of limiting. That's not what people are using clipping for along with limiters.
The idea is to compare, say, 1-2dB of clipping and 5-6 dB of limiting to 7-8dB of just limiting.
In that context, adding clipping and removing limiting may sound cleaner-provided the added distortion is quiet enough as to be effectively inaudible.
I hope that helps!
everything you say is possible with a standard overdrive. also, although you are taking strain off the limiter, the dynamic range reduction isn't free. when you do this incrementally across a whole mix you don't notice the damage it does until you AB level matched with the clippers removed. I was doing this stuff with clippers literally 25 years ago because i wanted to get loud mixes and then i was given some tips by sticky (was a big garage producer back in the day) and i stopped doing all that and started mixing more naturally and my tracks immediately started sounding substantially better. night and day. that was a quarter of a century ago and i've learned a lot since then
@@APMastering clipping is an integral part of getting D&B to the modern standard of 7LUFS or louder while retaining some dynamics, and we either run at 96Khz to prevent foldback at Nyquist or we oversample like crazy. I would even go as far as to say, it is impossible to get a "modern" mix up to a decent level without clipping (and properly gainstaging INTO the clippers at each stage) in many electronic genres, but if you can prove me wrong, be my guest. But STOP misinforming people, if you want to do oldschool CD masters at -12LUFS fine, but that goalpost has moved on (in my genre).
Makes Total Sense Now. I was experimenting with my WA73 Preamp (Transformers), looked at the meters and waveform and sure enough they were clipped off. Now that you explained it and have confirmed my suspicions, gives me more options to work with and try new things to see how other Saturators react when pushed. Love Your Channel And Your Plugin.
nice one! more plugins coming later in the year 😀
This controversial approach for clicks and comments is only going to get you so far in the end. I haven’t seen many audio channels consistently spreading misinfo
This channel has no future. It’s a desperate try to stand out.
can you actually state what I have said which is misinformation and provide evidence that it is? Otherwise you are just a troll
@@APMastering the only reason to run out a channel like this is page landing to sell stuff. In other words, fishing.
@@andreakleiner80 I've been there sure the expert was talking rubbish then as I got deeper into it realizing that it was me that me who did not understand, no shame in it see you at the other side.
Great video and I totally understand what you are saying with term usage, seems a bunch are not getting your point. For those of us that need to do louder/mixes and masters, what is the best form of clipping? going hot into a ADDA converter?
high end ADs have graceful overloading. this is the same as what I was saying with DAT recorders, because they are also high end ADs (for the high end models). However, you can accomplish the same in your DAW with oversampling and a distortion plugin then brick wall limiter.
@@APMastering "However, you can accomplish the same in your DAW with oversampling and a distortion plugin then brick wall limiter"
Now this is something I need to understand, is not a brick wall limiter doing the same as hard clipping, it shaves the peak of. What changes if the peak is "soft clipped/distorted" before it goes to the brick wall limiter. I have mixes where i have used limiters on instruments and I have to go in and automate in dips to get rid of unpleasant "pop" like sounds when dong additional compression at the mastering stage.
It is a given that I need to consider peaks at the sound selection/orchestration stage to avoid it being a problem in future compositions but still I'd like to understand what options I have to "transparently" tame peaks. Is there any plugins or applications where I could instruct the program to scale down all peaks by completely redraw the waveform scaling everything down in an offline mode? I do use the draw functionality in my editor to redraw problematic peaks but this is only viable if the number of peaks that needs to be reshaped is few. Great video Keep em coming!
Coding is not boring! On clipping, I tried it (don't judge me bro) and the result sounded bad, really bad. Maybe that was just my ineptitude, but I have since then not tried it again.
SOFT CLIPPING: You say it's "incorrectly" named, but if we all know exactly what it is and (more importantly) what it sounds like when we hear the name.. then, isn't that a really good name for it?
(Point being: words aren't numbers. They change definition all the time.. in the pursuit of efficient +communication+.)
im not a linguistic prescriptivist so I fundamentally agree with you but I'm allowed to dislike certain words. For example, I don't like the words "loo" or "couch" either.
I really enjoy your videos (I don't wact them all) but the ones I do are brilliant - educational and interesting.
thanks!
I think there's a reason I stick to free clipping and saturation plug ins. They sound plenty good (unless you don't like the sound of clipping I guess)
Hey some really good info thank you for the breakdowns. And it's good to see your plugin design is not yellow with a pink smiling sausage on it, catchy but really it's good to know some built in plugins can do the same
sausage fattener is a classic
@@APMastering
It's good for leads, makes easy to get sound quickly. It was suggested I use FL S Waveshaper to get similar sound, if I ever make money at my sound I'll buy that sausage
Is this some sort of performative art where the scam video is the scam?
As with your other debunking videos, is this also going to go up to part 3? I can already see the comments are going crazy, but from the quality and depth of your EQ & Compressor videos, I'm ready for it!
Thanks. I've got part 2, will release later today. But part 3 is open, both in terms of content and whether I will actually do one or not.
@@APMastering Awesome! Very exciting. will have to check it out & peep what the comments are to push for a part 3!
Nice video! I agree with all the technical arguments, but I still like using clippers sometimes. So, I don't fully agree that they always sound worse than other distortions or limiters. However, it's important not to rely on them as if they are a magic solution to boost the loudness of your tracks.
does loud actually sound better? do you always AB loudness matched?
They are definitely part of the solution-what ever you want to call it.
@@phadrus make sure to ab loudness matched. often you can get fooled by loudness
@@APMastering always do, thanks
I'm a big fan of UAD modeling the way particular units distort, that's pretty much my go to for color. From the pieces of gear accessible to me for comparisons, they do as good a job as I know of. I've never understood the whole "throw a clipper on it & it'll sound better" trend among producers. Ableton's Saturator has some good clipping algorithms though.
UAD is not bad, I just think it is overpriced and everything you can do with those plugins you can do elsewhere for free with more efficient plugins. they look good though and are good emulations
@@APMastering Certainly nowhere near everything. You get what you pay for at the end of the day, no points lost for being higher end. If you can point me to a free plugin that can sonically match the UAD LA-6176 plugin alone i'd eat my words but it's likely not out there. That one is a beast.
Some A-D clipping sounds far better than others (on transients, ie: not long wavelengths), and it comes down to the analog design / implementation.
Speak in absolutes? Be ready for he masters to laugh you into obscurity. Not once did you mention art or feeling. Try again.
art and feeling of a clamp function?
@@APMastering Thanks for confirming you don't understand what I'm talking about.
Clipping is rather a technical tool u use for specific purposes.
Art and feeling happens way before the Clipper, it isnt created by the Clipper but rather altered according to technical necessities, such as reaching a certain LUFS value for ur client or making too snappy drums hurt ur ears slightly less.
@@Melloh-293 clippers are used for sound design so they are absolutely an artistic tool... this is idiotic. this guy thinks skrillex's music is garbage so i dont give a fuck what he has to say, he can stick his fossil videos up his bottom
it’s fun thinking about this for guitar pedals. just call “overdrive” saturation, just call “distortion” and “fuzz” distortion. i don’t think it’s ever happening but it would be entertaining to watch
many of those stomp boxes just have diodes and so either are analogue clippers in the naïve case or attempt to mimic amp saturation. but i really like stomp boxes! so many different kinds. but yeah, the pedal manufacturers were playing the terminology obfuscation marketing game well before the plugin manufacturers
@@APMastering i love them too and in that context its not hard to find distinct value in either a discrete transistor cascade circuit or an opamp circuit with negative feedback diodes. guitar signal flow is very accommodating with multiple stages of filtering and limited-range speaker (sim). so it seems like you could get away with distorting the sound in pretty untraditional ways in a more traditional context. is it going to be worth the extra time involved though? that’s the billion dollar question.
@@sweeterthananything yeah it's the best place to experiment with electronic circuits!! i don't understand the application of clipping diodes in the negative feedback path of an opamp though as that seems to me at least intuitively to promote open loop gain, or at least react a bit weird with loud signals. in fact this sounds like a rudimentary upwards expander circuit... maybe clipping after that stage would yield an interesting experimental kind of sound. that said im probably just misunderstanding your comment 🤣
@@APMastering diodes in the feedback path of an op amp is often called a “soft clipping” configuration, coincidentally. This is one of the most common ways to create overdrive circuits. Famously used in the Ibanez Tubescreamer.
@@spicechateau just looked up the schematic... really interesting design! They are doing the exact opposite there as what I got from the initial comment. I thought diode clipping (IE with zener diodes etc) was in the negative feedback loop, which I think would indeed cause wild gain. But if I understand what is going on correctly there in the tube screamer, the diodes are closed, creating a high gain scenario with low input levels and then they open up with larger voltages, closing the feedback loop and effectively attenuating the amplification (or clamping it to unity). But because the particular diodes chosen for that design had a gradual forward conduction, this circuit mimics amp saturation... but its further spiced up by the capacitor which creates a kind of tilt shelf, I guess making it more distorted in the highs?? Really cool design! I don't understand everything that's going on there as im not an electronics expert but I hope I got the general idea. I think I'd still refer to this kind of this as a tube saturation mimicking design or something... after all its called a tube screamer, not a soft clipper screamer.
I'm sensing a trend w/ this guy
I like to debunk what I see as bullshit but I also like to have positive videos like interviews with legend engineers and tips and tricks videos etc
You mean click bait title and then loads on semi-nonsense?
I think at a certain point, the code itself is secondary, what you're paying for is the user experience. I found this by making a rudimentary amp sim by using an fx chain of free filters and saturator plugins.
That and if the plugin company put actual r&d into coming up with more efficient DSP algorithms, but I doubt that's the norm
I mainly agree. I have a couple of paid plugins I'm working on and 80% of the magic is in UX, not in the code. However 20% is also in the code.
Love these videos keep them coming 👏
And by God if after watching this you still want a clipper plugin there are SO MANY free ones.
yes!
Venn audio for example
I love Kazrog KClip zero & it's free
Very true but I think the video misses the idea that oversampling is really upsampling-> process-.> downsampling and there are various methods of upsampling and downsampling of varying quality and someone might be willing to toss a few $ for a better decimation filter.
Cool vid as always. Although u can actually clip the individual drum shells separately with an oversampled hard clipper. This way you get a very punchy sound that you can not get any other way.
Also when you clip groups or busses you get horrific IM distortion if you r in equal temperament
yeah clipping can really ramp up IM good point
A clipper has plenty of use. Sometimes a track may have super high digital peaks introduced by various processes, and you just want a non-fussy way of preventing those peaks from going into the master bus or sub-group. I don't want to be thinking about distortion plugins to get this job done, I just want a plain vanilla peak stopper. If there are soft clipping options then that's fine, I can explore those possibilities for saturating the signal as well, in which case I'm changing its "official" usage from mere peak stopper to distotion. I have no problem understanding what's going on. I don't feel conned, expecially as I mainly use a free clipper.
there no such thing as a soft clipper. unless you are making very unnatural experimental electronic music, clippers generally don't improve audio quality over a standard simple overdrive plugin, say from airwindows.
@@APMastering Indeed, soft clipping is distortion, but the term "distortion" is actually more vague. Is it phase distortion, overdrive or bit crushing? Is it a pedal or amp sim distortion? is there an eq stage? At least with a clipper you know it's just a signal hitting a wall, whether that's a straight line (hard clipping) or a transfer function curve (soft clipping, or if you prefer, distortion).
distortion is a signal hitting a wall. that's what saturation is. I have nothing against descriptions of distortion. I just dont like "soft clipper" because its stupid
This point that "all dirt can be classified as distortion or hard clipping" seems arbitrary; you might as well lump distortion and hard clipping together as well and say "they are all waveshapers, we don't need other names".
But they *can* be meaningful if used responsibly. Instead of describing the technology, they describe the transfer function. "Saturation" usually means a subtle transfer function that bends towards but does not reach a hard ceiling. Soft clipping usually means a hard ceiling *and* some distortion of signal near the ceiling. Fuzz usually means that the transfer function has significant nonlinearity *away* from the ceiling such as a gate effect or rectification.
so we mostly agree but I want to clarify a bit... im not saying that we should remove any distinctions and call everything "distortion" and ban all subcategories. I totally get that fuzz and tube modelling saturation are different kinds of sounds and we SHOULD use appropriate names to describe what we are trying to achieve with our plugins-
However I disagree with saturation not having a ceiling. Of course it does, otherwise all amps would have unlimited gain, At some point a 50w guitar amp doesnt go any louder than it goes. it has reached total saturation. and that's a ceiling. We can talk about sag, but that aside, the ceiling is a ceiling and saturation has a ceiling. What is tanh(5) and what is tanh(10000000000000000000000000000000000000000000) ?
@@APMastering It's dead easy to design a digital saturator that *doesn't* have a ceiling: just mix in some dry signal to that tanh function. Usually of course there *is* a ceiling... but if you aren't meant to actually hit it in regular operation, that's academic. People don't clip with Pultec transformers, but they still like how they sound.
There is a bit of confusion around the word "saturate" I guess; in physics/electronics terms yes saturation means running out of headroom. I think there is some ambiguity around whether a "saturated" signal is fully saturated or just displaying some of the effects of that process.
@@FrequencySummoner reaching saturation means that it doesnt keep going. that's the point of the word saturation. obviously if you mix in dry signal then you will keep going past the ceiling but my point was that if you do tanh(5) or tanh(243578234598234572893457298345723984572893457234895234798) you basically get the same number out the other side
When people refer to 'clipping' in the context described in this video, what they mean is driving the input of a Lavry Gold ADC from the output of an analog mixing or mastering console running hot. It's a technique. There are variations on it with similar equipment. Whether it's a good or bad technique is up to you. There's a nice demo of the technique over here: ruclips.net/video/e8RT5c1b0wg/видео.htmlsi=CQHMFMMLbGFTGvoU&t=421. In the same video he explains the application of the technique. From the Lavry manual "the SOFT SATURATION feature provides a digital emulation of an ideal magnetic tape saturation characteristic." So there are two parts to the 'clipping' technique: driving the analog front end of the ADC and the digital soft saturation algorithm after conversion. The 'clipping' technique actually happens across both domains. According to Copilot “this technique is often used in mastering to increase the perceived loudness of a track without using traditional compression or limiting. It can add a certain warmth and punch to the audio, making it sound more impactful.” Pretty interesting if you ask me.
Dan Lavry knows what he is on about and I have owned Lavry convertors myself. Notice how it is called soft saturation and not clipping. In terms of streaky, I'm sure he's a great guy but I end up disagreeing with him more often than agreeing. I'd quite like to interview him one day and get to the bottom of the disagreements.
@@APMastering "StandardCLIP was designed to handle the clipping process as flexible as possible. You can adjust the way the clipping is done easily, like a hard-limiting brick wall or smooth soft-saturated." It also says soft saturation here. I'm not saying it's right, but it's commonly the case that when people say 'clipping' what they mean is making their mix or master louder with Lavry like saturation. Sometimes folks use a word that isn't the technically correct word for what they mean.
AS A PRO I LAUGH AT OUR NEW YOUNG GENERATION..FUCK CLIPPERS THAT DOESN'T MAKE YOU BRO .😂😂
Good video, but to say that clipping without oversampling sounds “bad” as a blanket statement is a matter of opinion and not factual. There are times when clipping definitely sounds better.
I dont really know of *many* cases. I agree that blanket statements are normally a bad idea as you only require a single counter example to debunk them, but I can only really think of single sample spikes as being a use case for naive hard clipping without oversampling and that doesnt represent that many signals, unless you are recording close mic snare drums rim shots at a low sample rate
Yeah after many years of being very careful to avoid clipping, because it sounds bad, I'm not about to spend money on a plugin that does the thing I've spent years carefully avoiding because it sounds bad.
This is a great click-bait title - nice work! But basically it just boils down to semantics. I’ve watched the follow-up and mostly agree with the technical points you make. Personally I don’t ever hard clip - and if I do, I make sure it’s oversampled.
BUT to say that soft clipping isn’t actually clipping… I understand the technical distinction but at a certain point the end result of saturation is a maxed-out value, which is effectively clipped. Saturation is never reversed in the real world in the way you demonstrate, so even though the information is technically still there, no-one will ever hear it.
And to say that all hard clipping sounds bad is just as “inaccurate” as calling saturation distortion “clipping”. You say yourself in the follow-up that over a few samples, hard clipping can sound fine.
I actually think it’s great that you’re getting people to think I’m more detail about this stuff, but doubling down on the idea that clipping is a “scam” and hard clipping should never be used because it always sound bad undermines your credibility. If you want to be super-precise and nuanced about the technical stuff, you need to do the same with your musical commentary as well, IMO.
Joe gives some great examples of times when hard clipping can be useful and sound absolutely fine in the latest episode of my podcast, which came out today coincidentally (I found this video after someone in the comments there asked about it)
ruclips.net/video/XT4g6fawjko/видео.htmlsi=_f6ArOhAokMDzikv
As always in audio… it depends !
Thanks for your comment. Some clarification:
1. I disagree that it's click bait since that assumes it does not correspond to the message of the video, when in fact I say in the video that it is my on-the-record position that clipper plugins are a scam.
2. Terms: clipping is an extremely well defined term in multiple industries. I'm using this definition. Saturation is a well defined term in audio and I am using this term. Finally, "soft clipping" is a poorly defined term and can mean anything the plugin developer wants it to mean as there is no standard definition either in digital or analogue. I therefore reject this term as nonsense. This isn't all that controversial and it's not hair splitting, it's simply using words correctly as I see it.
3. regarding you thinking that "it doesnt matter if the detail is preserved because you cant hear it"... this is wrong. If the detail is preserved in the audio, you can hear it, that's the point. There is more detail to hear.
4. I don't know what you mean about there being an inconsistency between my accuracy of language here and elsewhere... what are you referring to about my music commentary? AFAIK I haven't done much/any of that, or I don't know what you mean by this.
Hope that clarifies where I'm coming from. Thanks for critiquing as I would hate to have zero pushback from viewers.
@@APMastering Thanks for taking the time to respond.
To comment on your comments (!):
1 - This was meant as a compliment, not a criticism - personally I distinguish between “good” click-bait, meaning that it encourages the click and then rewards with good content, as opposed to “bad” click-bait where as you say the content doesn't deliver on the promise.
2 - Something that is heavily saturated in analogue is still entirely (soft) clipped at the loudest moments (full saturation) so the information is not recoverable in practise - certainly not completely losslessly. So I see your point but I still disagree. The fact that "soft clipping" is poorly defined doesn't make it useless, or a scam.
3 - I disagree. What about the example of dither ? It preserves detail below the noise floor, which would otherwise be lost completely. But at 24-bit, is this extra preserved information audible ? Perhaps in theory, but in practise, probably not. As with the previous comment, your point is theoretically sound but doesn’t translate for the end user. As you mention yourself, even with a 100% recoverable situation like your distortion plugin, the listener would need to know and be able to implement the exact “undo” process to restore the crushed information in a meaningful sense, let alone without artefacts.
4 - So for example, you are being absolutely rigorous in the language of your previous replies - the distinction between clipping and saturation, for example - but then you make a blanket statement like “[hard] clipping always sounds bad” or “if the detail is preserved, you can hear it”. These statements are dependent on all kinds of factors, not absolute facts, and by stating them so categorically you undermine the persuasiveness of the other points you’re making… IMO.
Which is why the title is (good) “click-bait” for me - it’s an opinion presented as a controversial blanket statement. All this makes for an interesting discussion, though !
@ProductionAdvice thanks for clarifying. in terms of dither, the difference is very audible especially in fades and reverb tails at lower bit depths. you cannot hear dither at 24bit though because the two bits which are acted upon are quieter than the reproduction of all commonly available DACs, so it may even be impossible to hear a difference in this case, but this is why nobody who knows their stuff dithers to 24bit except in fringe circumstances. For proper use cases, dither is essential, like for 16bit and lower. For 8 bit, dither can transform the sound completely. The detail retained from saturation as opposed to clipping is similarly important and audible. you can try it with some simple tests.
@@APMastering I always dither 24-bit just in case, even though I suspect the difference is usually inaudible 🙂 So do all the other mastering engineers I know - and most of us know our stuff ! It’s easy and computationally cheap - why not ?
As far as our conversation about audible details is concerned, you just proved my point. If the extra detail that is certainly preserved technically by 24-but dither isn’t audible (even in 100% digital audio with no thermal self-noise) then you’re contradicting your own claim that extra detail is *always* audible.
And yes, as I’ve already said I agree with you that saturation/soft clipping is usually a better choice than hard clipping because it sounds better. (You would say because more detail is preserved, whereas I would say that less damage is done.)
I still think calling it soft clipping is absolutely fine, though 🙂
@@ProductionAdvice in the context of mastering, IE producing a final master, dithering to 24bit is equally as effective in improving the end quality as blessing the master file by singing it a song. I mean, for superstitious reasons if it makes you feel better, then sure, but anyone who understands DACs and human hearing will thinks you're doing something goofy by intentionally dithering to 24 bit. I'd happily debate any mastering engineer who thinks that dithering to 24bit has any benefit for a limited master. That said, there WOULD be certain fringe cases for 24bit dither, but within the music industry, this is going to be limited to intermediary formats with large dynamic range to be processed later... but in which case 32bit is more common and most people don't care so much nowadays about the hard drive space. Dithering was an incredibly handy technology of the past which massively increased quality of comparatively low bit depth formats but now it's more or less redundant with higher bit depths and near unlimited hard drive space.
so i shouldnt buy a soft clipper because an incorrect name even if i love the sound of it?
buy it if you want
With your philosophy Distortion and Clipping could also both be called Wave-Shaping. A clipper is just such a simple shape that it could also be coded in 2 lines. You are trying to be objective but not everyone can wrap their head around this stuff. Different types of waveshaping automatically get their own names like: fuzz, softclipping etc. To make it more tangible, understandable and intuitive for people that just want a particular sound. But they dont know how to look for it without those made-up names. Companies know this and the only way to reach an audience is to play along. A channelstrio for example is more than just an eq, compression etc. algorithm. Its the interface, the vibe, the inspiration, the limitations, the pre/post modeled non-linearities and the value it brings beyond just what code is used.
People dont just pay for the algorithm. Its the whole package. The workflow, the feel, the experience and the soul. It’s all about having your perfect tool at hand. The one that works best for you
All that you are saying here is actually your hallucination. you should talk to dr. Skrillex about it.
I don’t think he has heard Skrillex or Noisia.
I dont listen to skrillex since the music is awful. I know noisia because I started off in music as a dnb artist. the best dnb records were analogue mix downs cut to vinyl. when the dnb scene started embracing DAWs, that's when I lost interest in dnb. that's about the time when noisia entered the scene as an ed rush and optical rip off group.
@@APMastering your musical taste, while personal, show that you haven’t felt the need to keep up with the times
Thats an unreasonable comment.
Skrillex often uses Clippers to make his stuff louder but nobody said anything against that, correct?
@@APMastering his brostep stuff is objectively terrible , but the two albums he released last year (or was it 2022) are quite good
More information is better than less. thanks for the conversation.
I noticed that the stock compressor in Logic has a distortion knob with four settings: Off, Soft, Hard, Clip. Maybe that is the only “distortion” plugin I need! :)
well if you can get a wide variety of tones out of that then great. but logic also has (or at least used to have 20 years ago) a really good phase distortion and also a great overdrive
Hello: I really appreciate your "no nonsense" approach. I did look at your compressor plug-in, but I couldn't find what formats (VST, AU, AAX) it was available in. Thank-you.
its available only in VST3 on windows and Mac and AU on Mac.
Regarding saturation: I think the rabbit hole gets a bit deeper. Yes, most/all types of distortion can be described essentially as mathematical functions, however, many units in the analog domain display frequency specific saturation. Many compressors also compress different parts of the frequency spectrum slightly differently, or act on the spectrum in different ways. Transformers do this in a really strange way that seems difficult to model. I have found that the sound of transformers isnt just the result of “distortion”. In fact many very colored transformers have suprisingly low amounts of harmonics, and of course the harmonics that are there change on a frequency specific basis. Their sound almost seems to be the result of distortions in the time domain, acting like a soft compressor for certain frequencies. Magnetism is mysterious, in terms of replicating digitally
I mention this because while everything in this video is of course technically correct, it seems in the analog domain things are a bit more complicated. How does this fit into the “distortion = transfer function” framework? Are you being specific to digital audio?
@@Rhuggins you are 100% correct. the analogue circuit bit of this video was an overdub i did after recording the video and then i thought... "wait, i didn't mention analogue". The whole basis was intended exclusively for DSP/plugins. In fact my own plugin acts in a frequency dependent way actually. this is middl mostly because it's purely peak based and not rms. generally speaking you can replicate, or attempt to replicate, frequency specific stuff using EQ in the side chain or directly with pre emphasis and post emphasis eq
@@APMastering thanks for clarifying! Id love to hear some insight onto your workflow. Maybe thats worth another video. Do you find yourself using mostly Fabfilter type plugins and/or stock Reaper plugs? That would be interesting to see
@@Rhuggins I use pro Q 2 for basically everything and I also use some voxengo, TDR, airwindows and some other plugins. for mixing my main reverbs are the free TAL one and fab filter pro R. I'm currently developing a plugin which will replace almost all of my channel based plugins. I don't think I will get started with aux based plugins like reverb and delay as there are already so many good ones.
@@APMastering very cool. One of these days maybe ill have the balls to try a mix like that. I Sometimes I want that density and color that saturation provides, or the straight analog stuff. I have found a technique that really helps with providing similar crest factor ITB that analog gear can provide, with very basic processing. Id love to hear your thoughts on it/ if you incorporate it- perhaps I can email you offline
It is a tool, not a scam. You can hear "aliasing" only on pure sinewave\sawtoothwave\squarewave sinth signals. In a complex music instr. signal… it would be worthy of a 'Guinness record' if you could identify aliasing by ear (on 24bit 48kHz).
Clippers clip ("square") the peaks. If you do it to peaks less than 10ms it would be worthy of a 'Guinness record' if you could identify it by ear! I would extend that to even 20ms!
That is how you get rid of 'unperceivable' (to human ear) digital headroom eating peaks otherwise captured by the samplerate converters and occupying that 'headroom' you need for overall volume of the clipped signal (without compressoring it). If you try to clip a full mix, that would be considered a bad practice. Doing it on individual instruments (esp. percussive) is quite necessary.
I dont find it necessary. You wouldn't believe how many mixes I get sent which are literally just clipping on the bounce... so full mix clipping no oversampling, sent to be mastered. obviously I refuse the mixes but this is extremely common because people think its good to get louder mixes or whatever and you 100% can hear aliasing, IM and horrible stuff all over it.
I agree clipping a single sample transient or something is not perceivable but this is not what normal music looks like. even a loud snare hit is not a single sample. I just took a look at the length of the snare transient on a room mic and it's 70ms. Close mic on the snare is still 15ms. I dont know why eating all of your transients is supposed to sound good. it doesnt, aliasing or not.
Hey, what do you think of the Baphomatrix clip to zero method for loud mixes?
Isn't pushing your DAW master way in the red with no plugins on it a good way to achieve loud mixes,?
Clip to zero is the real deal
Clip to zero is the truth
you can do what you want. I dont concentrate on loud, I concentrate on good.
@@APMastering Context. In order to be considered good when the music slaps hard in the club with no headroom on the mixer/pa - it has to be loud.
This sounds like you don’t get why we use clipping from a mathematical standpoint. Thing is, no one hears via math. There IS a use case for clipping in mastering that makes sense and has beneficial outcomes, thus meaning there IS a reason to DO IT.
Think of it this way, you’re trying to disprove a reality that in and of itself makes the situation moot.
If I want the benefit of a clipper, I’m not going to F’n code one. I’m going to get a good digital clipper. Period. I’ll leave the math and coding to the people who have THAT sort of autism. Mine didn’t come with it.
well i'm a mastering engineer and can code this stuff so not sure what you are trying to tell me
This
Great video mate 👏
Hello I would like to buy one hyperbolic tangent plugin please
its on special offer, $199
Thanks for the breakdown! I need some advice: My OKX wallet holds some USDT, and I have the seed phrase. (behave today finger ski upon boy assault summer exhaust beauty stereo over). How can I transfer them to Binance?
lol wut
loving the new style! 😁😁
I use a hard clipper (eg KClip) on the second to last bit of my mastering chain (right before the final limiter) - for loudness purposes, to clip the very tops of the waveform so that the subsequent limiter does less work/works more effectively. Are you saying that this is largely nonsense and I could use any distortion plug in to do the same (and do it better)?
yeah i would actually advise AGAINST doing this despite everyone saying you should. i've been a professional mastering engineer for approaching 20 years and i've never had a clipper in my chain.
I’m so glad I didn’t fall for this!
Alain, you're the Batman of the audio world. Keep up the good work!
Looking forward to your next, covers-all-bases-distortion plugin then ;-)
my compressor plugin can do nice tanh distortion but I dont have a full featured distortion planned. I have a paid plugin which features distortion but no full featured distortion planned at this time.
I would love to see the video on what do you think on M/S (mid/side) EQing in both mixing and mastering. To me 99% of times any M/S EQing leads to "unfocused", smeared sound on any source.
Video on your opinion about M/S would be much appreciated. Thank you.
I will do a video on it! good idea! In short, M/S is complete shit most of the time as you have correctly identified. However, it's less than 99% of the time. Probably 70% of the time complete shit.
@@APMastering thank you. I'm really, really looking forward to it. Can you include both mixing and mastering scenarios? To me M/S introduce transient smearing. I had a discussion with one of the top engineers about it (I mean a guy who worked for example in Metropolis studios ect and he swears that M/S doesn't introduce any transient smearing, keeps the sound source focused ect.). Once I stopped using M/S at all in my opinion my mixes elevated to the next level. But I'm looking opinion of a pro. Thank you so much.
@@APMastering for example M/S compression is a completely different beast all together. I do use M/S compression quite often ( in mixing) but when it comes to M/S EQing it always sounds weird. Basically there's no physical mid channel (it's a sum of L/R), right? So is M/S EQing actually messing with sides only?
Can you also include that topic in the video? Much appreciated.
I dont like MS comp. it's only useful in mastering in situations where you are trying to save a bad mix. ill put mixing and mastering in there.
@@APMastering thanks once again. Really looking forward. I don't want to push too hard but a longer video with bit of M/S comp info would be even better. But my main topic is M/S EQ though.
I noticed this "distortion" language very early on by studying how analog equipment works. And learning about the evolution and development of digital plugins. And I'm only 2 years into mixing in the box. SMH
Thats why i always get so much distortion while “clipping” 😂😂😂
I clip just until I can hear the difference, then I back it off a bit. As a result, the limiter needs to work much less and the whole thing sounds much more transparent. Blame loudness wars.
this is a common fallacy though because you are not auditioning BOTH the limiter and the clipper. when you do that, you might find that the limiter without the clipper actually sounds better. I might make a video on it because its such a common thing I see/hear people doing and I totally disagree with it for multiple reasons
Flatline 2 by Submission audio is a clipper/limiter.. it's one of the single most used plugins in rock and metal music production. your opinions are biased and that's fine. everyone allowed to have opinions. I'm just sayin' you probly never mixed a metal track in your life so yea, mixing is very gatekeepy on youtube here i find. everyone so opinionated. Try Flatline 2, or try Apogee Soft Limit which is actually free.
i have not mixed a lot of metal but i've mixed rock. most of my mixing is in techno, rock and jazz. see part 2, just released it. it's not a subjective experience thing it's an objective facts thing
Clipping is like overexposing (to my knowledge)
exactly. you even say clipping in photography
As a software developer, I recently wrote a clipper to practice making VSTs. Yup, that's the algorithm.
Me too, I started working on a distortion plugin about a year ago to learn. That's how I did it for the hard clipper.
Maybe producers would understand distortion better if they all tried to synthesize a hardcore kick at least once.
a multiband clipper called Kclip saved a drumbus situation on one of my mixes and that alone was worth the cost ...but it's multiband with antialiasing and a bunch of different waveshape options with nice metering but yes a broadband clipper with no workflow enhancements would not be worth anything really
that's a multiband distortion plugin incorrectly called a clipper. although i've never used it, i've heard it's great
Clipper-clapper-clampy-satur-distor-alien-hysterical-jesus-mary and the wee donkey! I'm away to to pub for Guinness before my head explodes. That said, I do find Newfangled Auido Saturate quite useful...
Judging any dynamic processing with a sine sweep is not possible and only done by hobbyists....
i guess you dont know what aliasing is then