Not a boring video for me at least. Its these technical details I appreciate when watching your videos. These deep dives i dont really get from other YTer's and it's why i subscribed and click on your stuff. I find content on YT is saturated with beginner and intermediate level tips, which is understandable as it might hold the higher demographic for views. But there's very few sources on the more advanced/experienced side of these processes. Keep doing what you're doing.
I worked at two different radio stations in the 2000s, and I can confirm it's not a good idea to clip first because any EQ or phase process after the clipper can introduce what the broadcast industry refers to as "tilt." Bob Katz wrote an entire section of his book about broadcast processors and how they handle clipping, mentioning that the plateau on top of a clipped waveform can end up nearly anywhere post process, especially when a phase rotator is used. To combat this problem, the Telos Alliance developed a system called "Undo" in their Omnia.9 and later processors which algorithmically reconstructs a prediction of the missing portion of clipped waveforms so as to avoid things like intermodulation distortion and tilt later in the signal chain. It works very well in all but the worst cases (Metallica lol). I personally use it when a client asks for remastering to vinyl on CD or streaming files that are clipped badly, and the artist no longer has access to the multitrack.
The most recent approach I've set for myself is clipping in, clipping out of the master bus. Previously, i used to clip on the way out only. After Nicholas's video about clipping in, I've decided to give it a try and yeah it works for me as an additional check for overheating the mix, cause I mix loud and master up to -2LUFS short term, so I have multi-stage clipping and it's hard to maintain clean. And this 2-clipper approach gives me a hint when I'm starting fucking up my mix and still I'm free to do whatever crazy processing on the master I want, cause the last clipper will catch everything.
Something I saw from Luca Pretolesi. If you're going to high pass your master, use a low shelf to cut -12db instead. A lot less phase shift and achieves the same effect. You can also get away with a more aggressive Q factor.
I thought the criticism was that following a phase shift the clipped samples are no longer at the peak point of the waveform but are now rotated to an alternate place in the waveform, either the upward or downward transition, with the suggestion that the distortion created by the clipper would then become more audible. Most people are aware, I assume, that very aggressive filtering will change peak values but they generally aren't applied during mastering. Whether the clipped signals being rotated to elsewhere on the waveform is audible is another matter but I can see theoretically why that might be problematic.
Can absolutely confirm that clipped waveform shapes are changed by phase rotation and when audible clipping occurs the tone of that clipping is also changed. Can be measured easily with an oscilloscope and an all pass filter
With kind of old school EQs, most of the time the moves tend to be very broad, with digital EQ we have an option of doing ver sharp cuts, notch filters or boosts with very narrow Q and stuff like that. With broad moves the waveform stays relatively intact compared to how a couple of narrow bell cuts/boosts can mess up the waveform. And I believe thats one of the reasons why say 10k shelf boost on Neve 31102 actually starts at around 1kHz for example (the same with a lot of other analog/analog emulations EQs). The broader your moves are, less damage you do to the signal with EQ.
This is super informative. I'm gonna be honest, all this time I was simply putting the Clipper in my masterchain in the start to just control the peaks, not worrying about losing headroom even after some experimental processing, and also on some individual instruments. But this has given me an insight on how little things like ordering of the EQ into the clipper might affect the overall results.
Can see it when mixing too. Always group, clip my kicks and clap/snare. A lot of processing after, generally affects the peaks... never thought about too much tbh. Nice to know it's the phase shifting. Makes sense.
Crazy this came across my feed today. I’m new to mixing, currently mixing my first record (a band I recorded) and didn’t even know what clippers were. I’ve seen them in the plug-in menus, but just assumed they were some type of distortion effect, by the name probably some super stylistic digital distortion that would be useful for electronic music, and never looked into it. After watching this and reading the comments I’m realizing I need to incorporate this into my toolset. I take it you and most people here are working with electronic music? I’m recording and mixing live instruments, but drums in particular can be tough when getting dynamics under control while maintaining punch, and there are a couple songs I’m working on that are sounding great but the snare and kick are taking up a lot of headroom. I think some subtle clipping on the individual tracks and maybe the bus will help a lot, and this is also great to keep in mind when it comes to the final mix/master. Really glad i came across this right now!
Hey! Glad to hear this helped! I am working with all styles of mysic. As an intro to clippers, this is a good video :) No one told me this! Here's WHY clipper plugins make BIG LOUD MIXES & Masters! ua-cam.com/video/StDvZD_caM0/v-deo.html
Great video. It's good to see things tested rather than just have people repeat what they've heard as truth (and we all know how much misinformation has been galvanised via repetition without proof!) and as you say, this what we all need to do rather than waiting for someone to show us.
when people say things aren't like rocket science, I say mixing and mastering and recording actually is. it feels like a never ending journey to try and understand what the fucking, FUCKING fuck is going on with sound
I mix and master full-time for my living. I've found that what works the best in one scenario/song doesn't always work with other songs. Gotta be trying out different tools and combos. It's crazy how sometimes the limiter youve found the best may perform poorly in some scenarios. The other day, I did one of my loudest masters ever, for a client who wanted that hyper-loud modern DNB sound. I used two Ozone Maximizers in IRC3 mode, and had Standardclip between them. Ended up getting that -2.4 LUFS that the ref had (some Camo&Krooked + Mefjus stuff) and my master had better punch. I think limiters may be a little more forgiving, to an extent, when you have a bit too much bass, and clippers may start distorting too much there. Anyways, gotta have a large toolbox and try out different things. I handle a ton of house, tech house etc mixes and masters and work with people like John Summit, and for the past 10 years Ive been doing it, Ive rarely used clippers TBH.
And what would you say about clipping before AND after? Like clip as much as you can/like before, do your mastering techniques and then clip as much as you can/like after?
well said! Your thirst for improving yourself is also reflecting in your recent videos. This also tells me your gain stage really don't change much in the mixing process.
The clipper after a limiter will clip the limiting artifacts introduced by the limiter. I've always used a clipper like standardCLIP, and used a very high oversampling rate upon render (over 64x), which then runs in to a limiter. Oversampled clippers are wayyy more transparent than any limiter will ever be.
Yeah confusing comment - not clear what you're advocating for here. Clipping after limiter? I've never heard of doing this. It seems unnecessary since even +1db transients over 0 tend to be difficult to hear. Just look at the number of professional top 50 tracks on Spotify that have peak values above 0db.
@@panorama_mastering hey man I had a question. I keep beating my head over this so bad. Thank you for your reply. if I can hire sessions q and a please direct me. But I been making beats for a year now. I got the sounds I like. I get consistently good beats according to artist around me and if it’s a matter of making a beat putting a clipper and posting on UA-cam I’m happy. But when I mix them down to some random practice vocals. I’m stuck with my lufs around -10 db. Max. Truly -11 -12 lufs. And I keep getting down. I follow all these mixers and everybody seems to clip well over what I’m clipping no matter what 808 or kick I use. I can’t seem to contain the mix enough to push pass that in my mix buss practice master. Im not comparing songs in spectrum analyzers but im not sure what to look for fully. Or how to compare it to mine in the moment. I make hip hop trap mainly. Any advice?.
I see beat makers clip audio and put a clipper in mix bus. But since I mainly started recording artist first and tracking. That’s what I’m best at. And I track artist around -6 db. So I like to make my beats around that volume max so the session sits ready for an artist. I just use a gain plugin or the clipper itself to bring up the volume in two bus after I wanna mix it and such. I’m really not sure what I’m doing wrong I tried even just a basic fully piano with kick and 808 and some vocals. And just trying to see how loud I could push the the mix and still as stuck. I guess when your mixing is there any like level range you aim for. How do you know when a transient is doing too much or if something is too dynamic. I know it all ranges. But I gotta be doing something wrong if with a simple 4 track song I can’t push the volume even when the mix sounds great? :/
Would using a limiter instead of a clipper at the beginning of the chain make any difference? BTW, I love how passionate you are about everything down to the last single detail.
Normally i would not mention this but when you say negligent you mean negligible. Negligent is a bad thing, this makes some parts of the video sound very confusing if you don't realize that.
Post clipper processing, didn't you control the volume to stay the same? Because any further processing may increase the volume, did you reduce it at the same level it was from the clipper?
Natural phase only changes the response in the very highs, so it would probably change a bit with lowpass or filters in the higher part of the spectrum
@@iamjordanpass it corrects an artifact called eq cramping where in the highs the filters are not symmetrical anymore, by doing so it adds a very short latency and also changes slightly the impulse response (so transients, that’s why probably some people prefer it). Minimum is the most transparent to transient tho so if i’m not Eqing the top i use that👍🏻
Really liked this channel! Discovered recently! Started to follow! My personal comment regarding the question raised in the video about high passing in mastering. According to all material I researched and about what mastering itself is about, I think you never should need to high pass it, even in that harsh level. If there is an actual need in doing it so, it means that the pre-master/mixing phase was awfully made. Mastering is all about very small and subtle adjustments. And even doing such high passing after the clipping it wouldn't make sense because you will be distorting the audio wave after clipping the peaks, which would make the usage of the clipping kind of useless.
By any means make more of these geeky videos.. I've been watching your videos for a while but now just subscribed exactly because of these type of geeky contents ✌🏻
@@panorama_masteringThank you. Is the author Bob Katz. As an extension to this it would be very useful to have a video on the books you have in your library, for sound. To cover,,,, if you're comfortable about doing one? I was impressed you read the whole Pro Tools manual. Particularly when they provide NO physical manifestation of this resource. One has to print it!
Was wondering what your thoughts are about RX Adaptive Phase Rotation. This seems to give 1 to 3dB of headroom on the source material and shouldn’t impact the audio (see RX manual). Fun fact is that it add peaks when you use it on the final master. Wondering why.
To me making records is sometimes a game of inches. These bits of info (although time consuming on your end to put together) are quite useful, and even profound at times. In fact it's this kind of information that keeps me coming back to your channel.
I think with a lot of things, the theory almost never matches up with reality, because the theory is usually based upon absolute worse case scenarios. i.e. "What if you had a ton of phase shift and what if that phase shift causes a massive difference?" In practice and realistic situations, it's not the case. I had made a comment on another video regarding high-pass filtering and how it affects the shape of waveforms and the sound. I had to address my own theory again with some tests in real world scenarios. I found I was totally wrong about high-passing being nearly as bad as I thought, and my original tests were likely done with extremely sharp high-pass filters with unrealistic settings. Thanks for the video. It's always good to test things and find the truth. I love experiments like this!
I'm so appreciative of your advice on here. I know you aren't here to consult for us, but I wonder if anyone here can tell me if an aggressive low shelf filter after the clipper causes the same issues as a high pass filter after the clipper does?
Very interesting video! This topic it's very important but often overlooked. Just for curiosity: are there any mixing or mastering situation where you would use a hipass over a lowshelf cutting? And also, are there any situations where you hipass to purposely increase the peak (eg a snare sample with a weak transient)?
Clipper first is about to the low level compression in first slot, sucks out the low level information to the front with the transients. It halped me a lot this weekand on my work, thankyou. Ozon Dynamics on multi band can do the same on diffrent bands.
Thanks for the video. I struggle a bit with seeing how I would integrate this habit into my workflow. I tend to clip my snare and my kick individually, then clip my drum bus and then clip my mix bus because I (perhaps stupidly) assume that those stages of clipping will add up to give me plenty of headroom once I go to master. However, I also want to cut the unnecessary lows (sub 30 Hz) from my drum group and I usually do this by using a HPF on my drum bus. If I've already clipped my kick and snare at the track level, how would I fit in the HPF? In order to put the HPF before the clipping, I would have to apply it to every individual track in my drum group and that seems excessive and cumbersome.
@@panorama_mastering I suppose to remove sub amplitude that isn't necessary so that compressors and limiters have more room to create loudness in mastering before they start to distort. Regarding phase, I usually don't find there to be any phasing issues when I apply a HPF to a drum group (or any instrument for that matter). The only time that concerns me is if the phase is being shifted on one of two or more elements that really need to sound cohesive and in phase - like the overhead signal and the snare signal. I suppose a phase shift on an entire drum group could cause the phase relationship between the kick and the bass change but I usually don't strive to get perfect phase alignment between the kick and every individual bass note anyway. Have you a video on this?!
Great suggestion! But different result, limiters are controlling the amplification of the entire signal, clippers are only looking at the integer values of sample above a threshold and then adjusting those sample accordingly
3:30 "Negligent to have the HPF" could be very confusing to people watching with translated subtitles. I know you meant "negligible" because we're English speakers and I recognise the similarity of the words, but in other languages the translated words could be so different that people are left thinking you said the opposite of what you really said.
Sorry for writing multiple comments but I just realized that you recommend clipping before compressing so the compressor doesn't overreact but I understand that compressors, especially the multiband kind, also introduce phase shift. Does that mean we have to be careful with compressor settings, and mb crossfades otherwise we risk undoing the benefit of clipping? Do you think clipping between each compressor is a good idea? Would you, maybe even want a compressor with built in "pre-clipping" which would kind of make it a limiter? What about just slapping multiple limiters on with different settings?
Hi Nick, when mixing AND mastering a project YOURSELF, do you ever hard clip nearer the start of the MIX bus instead of the master bus (after aggressive phase shift of course)? With the goal of loudness, is the difference negligible?
What i would find interesting to see, and that to me caused me to keep moving my clipper towards the end of my chain, is what happens when you add parallel processing to it. For instance, if you use Clariphonic, stereo width in parallel, or even a vitalizer, having the clipper later might be advantageous if you're trying to get a sound out of it (and not just chop peaks). How do you see this?
Not sure, as I don’t use either of those processors, my thing would be take measurements placing them before and after the clipper to see how it’s manipulating the waveform and peak values
Curious your thoughts on clipping individual tracks/groups while also clipping on the master? Is that redundant/less clean? I’m talking about hard clipping for edm specifically
yes thanks to your video I'm playing in RX for the first time after having it for a while, very cool, I can't get the Loudness panel to go above -10 though@@panorama_mastering
amazing video as always ! i have a question, when i finish a project after mix and master i make sure the dynamics is closed to my reference similar track and the overall loudness is closed , its sounds good on my monitors and headphones and most systems. In the phone at middle volume its sounds good, but in full volume its like i puted 20 limiters and the mix is flat and lifeless. and the reference track sounds clean ive tested if im doing something wrong with my master and after checking - instagram audio or spotify its sounds quite clean loud and punchy at full volume but when i open the track trough soundcloud or the files folder, its sounds crushed in terms of dynamics at higher volumes. - the question - is modern phones speaker cannot pull out audio accurate at full volume when the track is louder than -14 lufs ? and what In your opinion could cause the problem if not this
Soundcloud has a shitty codec. I hope I am not misinterpreting here. But I am assuming those shittier codecs you’re hearing the degradation of the audio.
@@panorama_mastering Thank you for the answer, Appreciate, learn from you a lot ! it seems like what you said. from punchy sound and quite clean ( a bit clipping for transient harmonic enhancing purpose) on Spotify at full volume now about maybe misinterpreting The project details : The highest peak is 0.1 Maximum overall DR in the track - 8.7 Individual sub channels treatment Sounds as they should be in my experience The same audio on soundcloud sounds or even trough iphone files sounds like around -1.2 db of Dynamic range squashed and quiter At full volume Shitty Codec as you said is the main problem lead
Asking a complete noob to mastering (I primarily produce and mix) so this is primarily only for my understanding. You mentioned EQ, but how does clipping affect mix bus compression?
@@panorama_mastering Yes, definitely not a boring video. Defo need to be called out more, as it makes us delve deeper into the thought process of why something is a good choice or not... and this is something we wont satisfy until we know.
@@panorama_mastering you're not the problem dw lol. its just me being a noob. I watched your video on clipping you released prior to this one and its still a bit out of my depth but I never knew that clipping can also increase the signals below the threshold depending on the function of the curve, not just decreasing the signal like a limiter. (I also bought that book by Bob Katz "Mastering Audio" that was mentioned in the beginning of the video)
Of course geeky, ... but it adds to get things clear. Phase seems to be becoming more and more important. In EDM it's taking the shape of buy or die. For instance, DJs notice kicks that are hard to beat match.
@@panorama_mastering Cheers. I downloaded V-Clip just earlier today and have chucked it on each of my bus groups in Ableton... been able to chop off a bunch of sharp transients that - to MY ear - made zero difference in the resulting sound, but created more headroom. Is there a benefit to doing it with a clipper versus a limiter?
Since you seem to have a lot of confidence in your dynamics control, how would you make a signal louder with minimum stylistic effect (least coloring). I would go slow-attack-fast-release compressor (to even out things that are 1-2 beats long but keep transients in tact) => clipper (to lower the transients without smoothing them and prevent the next compressor from overreacting) => a fast-attack-medium-release compressor (to even out things that are about a bar long and get the louder "silences" in between) => limiter (-for safety- to maximize everything we squashed and get a bit of "extra"). I'm actually not sure about my chain because I heard people recommend doing fastest reacting things first and slowest last (so clipper would be before any compressors?) and I have no idea how after all that, a limiter is still able to squeeze out anything "extra", it feels like I am not doing my job enough and relying on a magic box to kind of do it well enough...
There are several points i don‘t get. First … I wouldn’t do a HP in the mastering stage with zero latency. Second … why is the overall level going up when you cut out low frequencies? That could only happen if you have activated auto gain compensation. This brings me to the third point … Gain staging. If the level goes up or down, no matter what type of EQing you do … compensate for the next plugin. I would use a clipper before the limiter to tame transients a bit more to avoid that the final limiter is working to much. If you need an extra EQ after the clipper the mix was already not balanced or you pushed the clipper to far. Or maybe i missed something important in your video 😊
This happens because the filter is non-linear phase. The phase shift causes increase in peak level as it “rearranges” the frequency spectrum. It’s a by product of the filter effect and the exact increase is dependent on signal content. The same happens with low pass filters but generally less so.
To respond to the HPF in mastering itself. You’re completely correct! That was demonstrative to meter the most extreme example of phase rotation from a processor and see what it did to peak value
Hyper nerd mode bro! Wow. Respect for both you and Ian. You're both ninja engineers. But the key point here is don't take any advice or opinions as gospel to be followed blindly. Get used to stress-testing anything you've been taught yourself. Form your own opinions from real-world data gleaned from real-world scenarios. Rarely in audio or acoustics is there one size fits all advice that applies to all genres, all workflows, and all projects.
Amazing work as usual Nick... The scientific/engineering (proper) separates theory from real world results.. Of course - while some people religiously high pass everything, I know you're of the thinking that it's a move that should be used only when necessary anyway ...
It was interesting and useful test results. I'm now using Softube Widener, Pulsar Audio 8200 in M/S, SSL de-esser, SSL G3, Softube Clipper on RMS, Schwabe Gold Clip, and SSL X-Limit on my 2 bus. EQ, compression, etc. only doing a little, such as the G3 is about 3 dB gain reduction, EQ just adding a low and high shelf of a few dB, then a few dB clipping first RMS then the peaks. Maybe it seems like I'm doing a lot, but they're really doing just a little shaping. I find I always clip then limit as the last 2 things.
You're someone who undoubtably knows his stuff but as someone who has never had any formal education in sound engineering, I sometimes hear you switching between mandarin and English because of how technical your explanations are.... Do you mind creating a second UA-cam channel where you explain the same concepts in a more simplified manner or at least provide an elementary course on the fundamentals of sound engineering??
This is the kind of video I pay internet for. thank you.
You’re welcome!
Not a boring video for me at least. Its these technical details I appreciate when watching your videos. These deep dives i dont really get from other YTer's and it's why i subscribed and click on your stuff.
I find content on YT is saturated with beginner and intermediate level tips, which is understandable as it might hold the higher demographic for views. But there's very few sources on the more advanced/experienced side of these processes.
Keep doing what you're doing.
Thanks man! Appreciate you!
I worked at two different radio stations in the 2000s, and I can confirm it's not a good idea to clip first because any EQ or phase process after the clipper can introduce what the broadcast industry refers to as "tilt." Bob Katz wrote an entire section of his book about broadcast processors and how they handle clipping, mentioning that the plateau on top of a clipped waveform can end up nearly anywhere post process, especially when a phase rotator is used. To combat this problem, the Telos Alliance developed a system called "Undo" in their Omnia.9 and later processors which algorithmically reconstructs a prediction of the missing portion of clipped waveforms so as to avoid things like intermodulation distortion and tilt later in the signal chain. It works very well in all but the worst cases (Metallica lol). I personally use it when a client asks for remastering to vinyl on CD or streaming files that are clipped badly, and the artist no longer has access to the multitrack.
Interesting!
Thanks for sharing this
The most recent approach I've set for myself is clipping in, clipping out of the master bus. Previously, i used to clip on the way out only. After Nicholas's video about clipping in, I've decided to give it a try and yeah it works for me as an additional check for overheating the mix, cause I mix loud and master up to -2LUFS short term, so I have multi-stage clipping and it's hard to maintain clean. And this 2-clipper approach gives me a hint when I'm starting fucking up my mix and still I'm free to do whatever crazy processing on the master I want, cause the last clipper will catch everything.
Nice! Thanks for sharing
Something I saw from Luca Pretolesi. If you're going to high pass your master, use a low shelf to cut -12db instead. A lot less phase shift and achieves the same effect. You can also get away with a more aggressive Q factor.
Honestly, I find myself using shelves more as it will maintain the integrity of the signal whilst controlling its volume.
Bang on! Shelves are the way!
Dan Worrall has a video on this too which was interesting.
Not boring at all! And it was a textbook lesson on how to process criticism positively. Bravo! I appreciate all you do.
Glad you enjoyed it! I appreciate that.
I thought the criticism was that following a phase shift the clipped samples are no longer at the peak point of the waveform but are now rotated to an alternate place in the waveform, either the upward or downward transition, with the suggestion that the distortion created by the clipper would then become more audible. Most people are aware, I assume, that very aggressive filtering will change peak values but they generally aren't applied during mastering. Whether the clipped signals being rotated to elsewhere on the waveform is audible is another matter but I can see theoretically why that might be problematic.
Can absolutely confirm that clipped waveform shapes are changed by phase rotation and when audible clipping occurs the tone of that clipping is also changed. Can be measured easily with an oscilloscope and an all pass filter
@@Oni_I_Dub just did that same test and came back to finish the video 😂
With kind of old school EQs, most of the time the moves tend to be very broad, with digital EQ we have an option of doing ver sharp cuts, notch filters or boosts with very narrow Q and stuff like that. With broad moves the waveform stays relatively intact compared to how a couple of narrow bell cuts/boosts can mess up the waveform.
And I believe thats one of the reasons why say 10k shelf boost on Neve 31102 actually starts at around 1kHz for example (the same with a lot of other analog/analog emulations EQs). The broader your moves are, less damage you do to the signal with EQ.
Bangon
Thank you, can you please tell me which software you use to analyze the audio? the blue one
IZotope RX audio editor
This is super informative. I'm gonna be honest, all this time I was simply putting the Clipper in my masterchain in the start to just control the peaks, not worrying about losing headroom even after some experimental processing, and also on some individual instruments. But this has given me an insight on how little things like ordering of the EQ into the clipper might affect the overall results.
Glad it was helpful!
Thanks for lesson🤝
Btw: all these small boring details at the end make huuuge difference 🔥
Spot in, devils in the details
thank you!
You're welcome!
Your videos are the best. Truly my favorite audio content on youtube right now ❤
Wow, thank you!
Can see it when mixing too. Always group, clip my kicks and clap/snare. A lot of processing after, generally affects the peaks... never thought about too much tbh. Nice to know it's the phase shifting. Makes sense.
Yeap! It was a great ovservation for Ian to make and us as engineers to be conscious about
Crazy this came across my feed today. I’m new to mixing, currently mixing my first record (a band I recorded) and didn’t even know what clippers were. I’ve seen them in the plug-in menus, but just assumed they were some type of distortion effect, by the name probably some super stylistic digital distortion that would be useful for electronic music, and never looked into it. After watching this and reading the comments I’m realizing I need to incorporate this into my toolset. I take it you and most people here are working with electronic music? I’m recording and mixing live instruments, but drums in particular can be tough when getting dynamics under control while maintaining punch, and there are a couple songs I’m working on that are sounding great but the snare and kick are taking up a lot of headroom. I think some subtle clipping on the individual tracks and maybe the bus will help a lot, and this is also great to keep in mind when it comes to the final mix/master. Really glad i came across this right now!
Hey! Glad to hear this helped!
I am working with all styles of mysic.
As an intro to clippers, this is a good video :)
No one told me this! Here's WHY clipper plugins make BIG LOUD MIXES & Masters!
ua-cam.com/video/StDvZD_caM0/v-deo.html
@@panorama_mastering thanks man! Very much appreciated.
Great video. It's good to see things tested rather than just have people repeat what they've heard as truth (and we all know how much misinformation has been galvanised via repetition without proof!) and as you say, this what we all need to do rather than waiting for someone to show us.
Bang on! And there’s nothing incorrect about the original comment! We kust need to flesh out concepts in practice!
when people say things aren't like rocket science, I say mixing and mastering and recording actually is. it feels like a never ending journey to try and understand what the fucking, FUCKING fuck is going on with sound
I mix and master full-time for my living. I've found that what works the best in one scenario/song doesn't always work with other songs.
Gotta be trying out different tools and combos.
It's crazy how sometimes the limiter youve found the best may perform poorly in some scenarios.
The other day, I did one of my loudest masters ever, for a client who wanted that hyper-loud modern DNB sound.
I used two Ozone Maximizers in IRC3 mode, and had Standardclip between them.
Ended up getting that -2.4 LUFS that the ref had (some Camo&Krooked + Mefjus stuff) and my master had better punch.
I think limiters may be a little more forgiving, to an extent, when you have a bit too much bass, and clippers may start distorting too much there.
Anyways, gotta have a large toolbox and try out different things.
I handle a ton of house, tech house etc mixes and masters and work with people like John Summit, and for the past 10 years Ive been doing it, Ive rarely used clippers TBH.
Spot on, thanks for sharing!
And what would you say about clipping before AND after? Like clip as much as you can/like before, do your mastering techniques and then clip as much as you can/like after?
Clip with purpose.
My purpose is control the transients that feed my processors so I clip first.
well said! Your thirst for improving yourself is also reflecting in your recent videos. This also tells me your gain stage really don't change much in the mixing process.
Thanks mate
The clipper after a limiter will clip the limiting artifacts introduced by the limiter. I've always used a clipper like standardCLIP, and used a very high oversampling rate upon render (over 64x), which then runs in to a limiter. Oversampled clippers are wayyy more transparent than any limiter will ever be.
So clipping before limiting is better than after?
Yeah confusing comment - not clear what you're advocating for here. Clipping after limiter? I've never heard of doing this. It seems unnecessary since even +1db transients over 0 tend to be difficult to hear. Just look at the number of professional top 50 tracks on Spotify that have peak values above 0db.
clipping before multiband compression sounds so good though. something about the way it interacts with the distortion
Thanks for doing the homework for us! I’ll sit by you in class today!
No sweat, steal my notes!
@@panorama_mastering hey man I had a question. I keep beating my head over this so bad. Thank you for your reply. if I can hire sessions q and a please direct me. But I been making beats for a year now. I got the sounds I like. I get consistently good beats according to artist around me and if it’s a matter of making a beat putting a clipper and posting on UA-cam I’m happy. But when I mix them down to some random practice vocals. I’m stuck with my lufs around -10 db. Max. Truly -11 -12 lufs. And I keep getting down. I follow all these mixers and everybody seems to clip well over what I’m clipping no matter what 808 or kick I use. I can’t seem to contain the mix enough to push pass that in my mix buss practice master. Im not comparing songs in spectrum analyzers but im not sure what to look for fully. Or how to compare it to mine in the moment. I make hip hop trap mainly. Any advice?.
I see beat makers clip audio and put a clipper in mix bus. But since I mainly started recording artist first and tracking. That’s what I’m best at. And I track artist around -6 db. So I like to make my beats around that volume max so the session sits ready for an artist. I just use a gain plugin or the clipper itself to bring up the volume in two bus after I wanna mix it and such. I’m really not sure what I’m doing wrong I tried even just a basic fully piano with kick and 808 and some vocals. And just trying to see how loud I could push the the mix and still as stuck.
I guess when your mixing is there any like level range you aim for. How do you know when a transient is doing too much or if something is too dynamic. I know it all ranges. But I gotta be doing something wrong if with a simple 4 track song I can’t push the volume even when the mix sounds great? :/
Would using a limiter instead of a clipper at the beginning of the chain make any difference? BTW, I love how passionate you are about everything down to the last single detail.
Normally i would not mention this but when you say negligent you mean negligible. Negligent is a bad thing, this makes some parts of the video sound very confusing if you don't realize that.
He fixed toward the end.
@@Wizardofvoz2 yeah, I wanted to write an update but couldn't come up with how to do it nicely so figured will just leave it
Sorry man!
Post clipper processing, didn't you control the volume to stay the same? Because any further processing may increase the volume, did you reduce it at the same level it was from the clipper?
Have you experimented with taking Pro-Q out of “zero latency” and switching it to “natural phase”?
Natural phase only changes the response in the very highs, so it would probably change a bit with lowpass or filters in the higher part of the spectrum
@@coffeemusicit interesting. I’ve seen a lot of mixing engineers running all their Pro-Qs in Natural over Zero
@@iamjordanpass it corrects an artifact called eq cramping where in the highs the filters are not symmetrical anymore, by doing so it adds a very short latency and also changes slightly the impulse response (so transients, that’s why probably some people prefer it). Minimum is the most transparent to transient tho so if i’m not Eqing the top i use that👍🏻
@@BROOKLYN_EMPIRE minimum is zero latency, i am used to it cause i use Kirchoff and it’s called like that ahah
Really liked this channel! Discovered recently! Started to follow!
My personal comment regarding the question raised in the video about high passing in mastering. According to all material I researched and about what mastering itself is about, I think you never should need to high pass it, even in that harsh level. If there is an actual need in doing it so, it means that the pre-master/mixing phase was awfully made. Mastering is all about very small and subtle adjustments. And even doing such high passing after the clipping it wouldn't make sense because you will be distorting the audio wave after clipping the peaks, which would make the usage of the clipping kind of useless.
Spot on!
By any means make more of these geeky videos.. I've been watching your videos for a while but now just subscribed exactly because of these type of geeky contents ✌🏻
Welcome aboard!
What is the book Bob K he’s referring too?
Mastering audio!
@@panorama_masteringThank you. Is the author Bob Katz. As an extension to this it would be very useful to have a video on the books you have in your library, for sound. To cover,,,, if you're comfortable about doing one?
I was impressed you read the whole Pro Tools manual. Particularly when they provide NO physical manifestation of this resource. One has to print it!
1 min in to one of your videos made me subscribe and dive into all of your stuff. It's that good
Thanks man!
Was wondering what your thoughts are about RX Adaptive Phase Rotation. This seems to give 1 to 3dB of headroom on the source material and shouldn’t impact the audio (see RX manual). Fun fact is that it add peaks when you use it on the final master. Wondering why.
Interesting topic, i want to spend a little more time with. I am well aware of it, and have used it in some instances before!!
@@panorama_mastering how is it going? 🤓
To me making records is sometimes a game of inches. These bits of info (although time consuming on your end to put together) are quite useful, and even profound at times. In fact it's this kind of information that keeps me coming back to your channel.
Thanks man!
I think with a lot of things, the theory almost never matches up with reality, because the theory is usually based upon absolute worse case scenarios. i.e. "What if you had a ton of phase shift and what if that phase shift causes a massive difference?" In practice and realistic situations, it's not the case.
I had made a comment on another video regarding high-pass filtering and how it affects the shape of waveforms and the sound. I had to address my own theory again with some tests in real world scenarios. I found I was totally wrong about high-passing being nearly as bad as I thought, and my original tests were likely done with extremely sharp high-pass filters with unrealistic settings.
Thanks for the video. It's always good to test things and find the truth. I love experiments like this!
Spot on! Glad to have you aboard and hear you trying new things
this is the content I'm looking for on yt!!!
I'm super happy to hear!
what are your thoughts on clipping a little upfront then clipping a little at the end of the chain?
Stumbled across your channel recently. Excellent insight and practical guides!
Glad you like them!
I'm so appreciative of your advice on here. I know you aren't here to consult for us, but I wonder if anyone here can tell me if an aggressive low shelf filter after the clipper causes the same issues as a high pass filter after the clipper does?
Not boring at all. Quality content.
Cheers
Very interesting video! This topic it's very important but often overlooked. Just for curiosity: are there any mixing or mastering situation where you would use a hipass over a lowshelf cutting? And also, are there any situations where you hipass to purposely increase the peak (eg a snare sample with a weak transient)?
Excessive gnarly sub flab below 30hz
5:58 THERE it is! 😂
It would have been interesting to see the values if you had used the Q3 in linear phase mode. Nevertheless, as always a very interesting video.
Of course the HPF was just to showcase EXTREMES. Given things are linear phade it shouldn’t move the sample value.
Great content as usual! ✌🏻
Appreciate it!
Negligent or negligible?
He means negligible, as in "not worth worrying about"
Negligible… sorry!
Not all heroes wear capes! 🙏
Not all heros eat crepes!
@@panorama_masteringcouldn't help yourself
Clipper first is about to the low level compression in first slot, sucks out the low level information to the front with the transients. It halped me a lot this weekand on my work, thankyou. Ozon Dynamics on multi band can do the same on diffrent bands.
Yeap! No worries!
Not a boring topic at all! Learning is fun lol
Ha awesome!
Thanks for the video. I struggle a bit with seeing how I would integrate this habit into my workflow. I tend to clip my snare and my kick individually, then clip my drum bus and then clip my mix bus because I (perhaps stupidly) assume that those stages of clipping will add up to give me plenty of headroom once I go to master. However, I also want to cut the unnecessary lows (sub 30 Hz) from my drum group and I usually do this by using a HPF on my drum bus. If I've already clipped my kick and snare at the track level, how would I fit in the HPF? In order to put the HPF before the clipping, I would have to apply it to every individual track in my drum group and that seems excessive and cumbersome.
Why MUST you hy default use a HPF on the drums. Even if there is NO information their it completely shifts the phase…
@@panorama_mastering I suppose to remove sub amplitude that isn't necessary so that compressors and limiters have more room to create loudness in mastering before they start to distort. Regarding phase, I usually don't find there to be any phasing issues when I apply a HPF to a drum group (or any instrument for that matter). The only time that concerns me is if the phase is being shifted on one of two or more elements that really need to sound cohesive and in phase - like the overhead signal and the snare signal. I suppose a phase shift on an entire drum group could cause the phase relationship between the kick and the bass change but I usually don't strive to get perfect phase alignment between the kick and every individual bass note anyway. Have you a video on this?!
Thanks! What about a limiter first in the chain to lower some peaks?. Same results?. Best/Mathias
Great suggestion! But different result, limiters are controlling the amplification of the entire signal, clippers are only looking at the integer values of sample above a threshold and then adjusting those sample accordingly
@@panorama_mastering Thanks, I see some mastering engineers do that.
3:30 "Negligent to have the HPF" could be very confusing to people watching with translated subtitles. I know you meant "negligible" because we're English speakers and I recognise the similarity of the words, but in other languages the translated words could be so different that people are left thinking you said the opposite of what you really said.
Yes! Sorry, you’re not th first person to mention. I’ll definitely take feedback on the syntax of my language on board moving forward. Thank you :)
Does the phase shift back to the original position if i put 2 identical hpf one after another? Could it be a way to correct the phase shifting?
No, you just double up the order (make steeper) the slope of the filter
@@panorama_mastering but what happens to the phase? if 1 hpf rotate the phase 180° doesn't 2 of them make 360°? (i'm just ignorantly wondering)
@@mttlsa686then it becomes a full cycle out, so it may be “in phase,” but it will technically not be “in time”
Sorry for writing multiple comments but I just realized that you recommend clipping before compressing so the compressor doesn't overreact but I understand that compressors, especially the multiband kind, also introduce phase shift. Does that mean we have to be careful with compressor settings, and mb crossfades otherwise we risk undoing the benefit of clipping? Do you think clipping between each compressor is a good idea? Would you, maybe even want a compressor with built in "pre-clipping" which would kind of make it a limiter? What about just slapping multiple limiters on with different settings?
A lot to unpack here.
Something like OTT will absolutely fuck with the peak value.
I never steess about phase/MB because I use leapwing dynone
Hi Nick, when mixing AND mastering a project YOURSELF, do you ever hard clip nearer the start of the MIX bus instead of the master bus (after aggressive phase shift of course)? With the goal of loudness, is the difference negligible?
Bro what about new softube clipper? Make a rewiev on it😊
Maybe down the line.
Awesome stuff
Cheers
What i would find interesting to see, and that to me caused me to keep moving my clipper towards the end of my chain, is what happens when you add parallel processing to it. For instance, if you use Clariphonic, stereo width in parallel, or even a vitalizer, having the clipper later might be advantageous if you're trying to get a sound out of it (and not just chop peaks). How do you see this?
Not sure, as I don’t use either of those processors, my thing would be take measurements placing them before and after the clipper to see how it’s manipulating the waveform and peak values
New sub here. Thanks for sharing this knowledge. It helps a lot to us that are new to mixing ❤
You’re welcome!
Love your videos
Cheers m808b
Curious your thoughts on clipping individual tracks/groups while also clipping on the master? Is that redundant/less clean? I’m talking about hard clipping for edm specifically
Great information! Thank you so much 🥳
You are so welcome!
what about clipping and then adding your eq moves with gain compensation so the tp would be the same
It’s moreso about how that shift in phase moves where the sample peak sits.
Awesome video! Thank you!
You’re welcome
Dumb question. What’s that audio editor?
Izotope rx audio editor
Thanks for the awesome content
Glad you enjoy it!
Do you understand how negligently I have added a 36db/oct hpf on ProQ3 on a master??!? 🤣🤣🤣🤣
Thanks!
No problem!
What plugin is that for Waveform Statistic?
nevermind, it's RX audio editor
Rx audio editor ;)
yes thanks to your video I'm playing in RX for the first time after having it for a while, very cool, I can't get the Loudness panel to go above -10 though@@panorama_mastering
I don't understand : Clipper or not, HPF will create phase shifting and will increase the peak value...
Yes, you’re right! The aim of having my clipper first is to controll peak values. I don’t want to compromise that with a HPF after the fact
amazing video as always !
i have a question, when i finish a project after mix and master i make sure the dynamics is closed to my reference similar track and the overall loudness is closed , its sounds good on my monitors and headphones and most systems. In the phone at middle volume its sounds good, but in full volume its like i puted 20 limiters and the mix is flat and lifeless. and the reference track sounds clean
ive tested if im doing something wrong with my master and after checking - instagram audio or spotify its sounds quite clean loud and punchy at full volume but when i open the track trough soundcloud or the files folder, its sounds crushed in terms of dynamics at higher volumes. - the question - is modern phones speaker cannot pull out audio accurate at full volume when the track is louder than -14 lufs ?
and what In your opinion could cause the problem if not this
Soundcloud has a shitty codec. I hope I am not misinterpreting here. But I am assuming those shittier codecs you’re hearing the degradation of the audio.
@@panorama_mastering
Thank you for the answer,
Appreciate, learn from you a lot !
it seems like what you said. from punchy sound and quite clean ( a bit clipping for transient harmonic enhancing purpose) on Spotify at full volume
now about maybe misinterpreting
The project details :
The highest peak is 0.1
Maximum overall DR in the track - 8.7
Individual sub channels treatment
Sounds as they should be in my experience
The same audio on soundcloud sounds or even trough iphone files sounds like around -1.2 db of Dynamic range squashed and quiter
At full volume
Shitty Codec as you said is the main problem lead
Asking a complete noob to mastering (I primarily produce and mix) so this is primarily only for my understanding.
You mentioned EQ, but how does clipping affect mix bus compression?
In what sense?
Sounds like you're asking the questions I ask myself all the time! does my music improve... no, but at least I know!!!
You’re not alone! We all go down these rabbit holes!
@@panorama_mastering Yes, definitely not a boring video. Defo need to be called out more, as it makes us delve deeper into the thought process of why something is a good choice or not... and this is something we wont satisfy until we know.
I just use the stock reaper clpper but i never have any luck with it as soon as i add just a little bit i can hear distortion
damn good
I have no idea whats going on here. where can I actually go to learn the foundations well enough to understand this stuff?
Was there anything in particular you'd like explained in better detail?
@@panorama_mastering you're not the problem dw lol. its just me being a noob. I watched your video on clipping you released prior to this one and its still a bit out of my depth but I never knew that clipping can also increase the signals below the threshold depending on the function of the curve, not just decreasing the signal like a limiter. (I also bought that book by Bob Katz "Mastering Audio" that was mentioned in the beginning of the video)
How would using an imager like leapwing StageOne2 before a clipper?
Not sure, we’d need to try it out!
Of course geeky, ... but it adds to get things clear.
Phase seems to be becoming more and more important. In EDM it's taking the shape of buy or die. For instance, DJs notice kicks that are hard to beat match.
Do you use a clipper on every channel or just on the premaster?
Rarely! I use clipping for managing sharp transient peak information. Whether thats on an individual element or the mix bus. It can change
@@panorama_mastering Cheers. I downloaded V-Clip just earlier today and have chucked it on each of my bus groups in Ableton... been able to chop off a bunch of sharp transients that - to MY ear - made zero difference in the resulting sound, but created more headroom. Is there a benefit to doing it with a clipper versus a limiter?
Both before and after limiter.
About to watch but immediate answer / Clip first AND last. (And sometimes in between) Very small amounts.
Since you seem to have a lot of confidence in your dynamics control, how would you make a signal louder with minimum stylistic effect (least coloring). I would go slow-attack-fast-release compressor (to even out things that are 1-2 beats long but keep transients in tact) => clipper (to lower the transients without smoothing them and prevent the next compressor from overreacting) => a fast-attack-medium-release compressor (to even out things that are about a bar long and get the louder "silences" in between) => limiter (-for safety- to maximize everything we squashed and get a bit of "extra").
I'm actually not sure about my chain because I heard people recommend doing fastest reacting things first and slowest last (so clipper would be before any compressors?) and I have no idea how after all that, a limiter is still able to squeeze out anything "extra", it feels like I am not doing my job enough and relying on a magic box to kind of do it well enough...
You buggin
@@StarOnCheek glue style compressor with 4db gain reduction 3ms attack, auto release and 4:1 ratio. I don't use 2 compressors on the master.
There are several points i don‘t get. First … I wouldn’t do a HP in the mastering stage with zero latency. Second … why is the overall level going up when you cut out low frequencies? That could only happen if you have activated auto gain compensation. This brings me to the third point … Gain staging. If the level goes up or down, no matter what type of EQing you do … compensate for the next plugin. I would use a clipper before the limiter to tame transients a bit more to avoid that the final limiter is working to much. If you need an extra EQ after the clipper the mix was already not balanced or you pushed the clipper to far. Or maybe i missed something important in your video 😊
This happens because the filter is non-linear phase. The phase shift causes increase in peak level as it “rearranges” the frequency spectrum. It’s a by product of the filter effect and the exact increase is dependent on signal content. The same happens with low pass filters but generally less so.
Proper lofi answering spot on about the peak level change with HPF.
To respond to the HPF in mastering itself. You’re completely correct! That was demonstrative to meter the most extreme example of phase rotation from a processor and see what it did to peak value
Keep in mind all options known to man. Rules are only relevant if it was the option that worked in that instance.
Spot on
Hyper nerd mode bro! Wow. Respect for both you and Ian. You're both ninja engineers. But the key point here is don't take any advice or opinions as gospel to be followed blindly. Get used to stress-testing anything you've been taught yourself. Form your own opinions from real-world data gleaned from real-world scenarios. Rarely in audio or acoustics is there one size fits all advice that applies to all genres, all workflows, and all projects.
Yeah nah, that was excellent. A bit boring, but great to have that confirmation. Clip last (Usually). Thank you!
Haha lovely! No sweat
can i get a TLRD for this video? Im just getting lost throughout it lmao
Amazing work as usual Nick... The scientific/engineering (proper) separates theory from real world results..
Of course - while some people religiously high pass everything, I know you're of the thinking that it's a move that should be used only when necessary anyway ...
Spot on!
3:32 do you mean “negligible”? Because that’s a COMPLETELY opposite meaning in the context of what you’re saying.
Negligible!
the answer is both.
Touchè
Exactly !
It was interesting and useful test results.
I'm now using Softube Widener, Pulsar Audio 8200 in M/S, SSL de-esser, SSL G3, Softube Clipper on RMS, Schwabe Gold Clip, and SSL X-Limit on my 2 bus.
EQ, compression, etc. only doing a little, such as the G3 is about 3 dB gain reduction, EQ just adding a low and high shelf of a few dB, then a few dB clipping first RMS then the peaks. Maybe it seems like I'm doing a lot, but they're really doing just a little shaping.
I find I always clip then limit as the last 2 things.
Yes! I used to do this too!
What made you choose the new softube stuff? Just the newest tools in the box, or did you a/b them against previous tools and they won?
If you clip first you're last.
YES!
You're someone who undoubtably knows his stuff but as someone who has never had any formal education in sound engineering, I sometimes hear you switching between mandarin and English because of how technical your explanations are.... Do you mind creating a second UA-cam channel where you explain the same concepts in a more simplified manner or at least provide an elementary course on the fundamentals of sound engineering??
There are other channels addressing these things, check them out
👍
👍
Aposto dez dólares que pessoas que falam inglês não entenderam. Podia ter dado exemplo prático...
Fair enough. Please do!
205th comment 4.10.2k24
If you have to make a cut on a master, the mix is BAD.
Also, I prefer to use a non-true-peak limiter and clip all the true peaks after.
Interesting… i might have something to say about limiting and clipping….
Clip firrrrrrsssssttt
at 6:21 Nicholas says the N-word
The most boring (not) informative video :)
Ha cheers!