Let me know which one you thought had the high pass filter at 4:20... I'm still waiting on a winner. Edit: still waiting on a correct answer. I will reveal the answer in my next video. If you enjoy my videos, you can buy me a coffee here buymeacoffee.com/michael.inthemix 📱 All my links: linktr.ee/inthemixlinks 0:00 - Introduction 0:20 - Good advice 1:12 - Bad advice 1:35 - Where the bad advice comes from (Audio Example) 4:20 - Can you hear the difference? Full Mix Example 5:20 - Plugin Doctor and some social media advice 6:45 - Thanks for watching
A & B sound the same to me for the most part. Which brings up an interesting point. Like 99% of the people listening to your music isn't gonna hear the difference either, or even care or know what a little phasing or pre-ringing is/sounds like. Listening on E8XT's
Literally exactly, I think its a trick question, but even if its not, it may aswell be, theres NO audible difference to 99% of people, on 90% of devices.
Yeah true. I have 10+ years playing piano, so my ears are pretty sensitive, but generally I don't hear the differences. I thought it's because I wasn't a pro in digital music.@@t3ch_n0
@@t3ch_n0there will be audible differences when you're mastering the track. You'll be able to punch through harder during limiter with phase aligned tracks
but you're listening on a low-to-mid-range pair of monitors. When people mix and master music it's designed to sound great on ALL speakers. Especially dance music. If the low-end sounds like shit, it will literally never get played in a club or by any DJ.
Audible differences are only one reason people do this. The second one is low end rumble from different tracks adding up and thus forcing you to mix at a lower level before you risk clipping. This problem by now only applies to the main bus since floating point computing allows us to exceed 0dB on individual tracks but there used to be times where this was different. The lower bit rate forced you to record as hot as possible to use up as much of the precious dynamic range as possible. But since you were not allowed to go past 0dB at any point in the chain a simple EQ boost on a single track could cause huge problems, because you already were as close to the ceiling as possible. So high pass filters were not only used to give every instrument it's own place in the mix but also to help to be deal with the limitations of early digital recordings.
Was a little worried when I saw the thumbnail, but I am glad you used the clickbait properly and got people to the “correct” conclusion. Lots of crazy info on this topic
This is weak sauce advice, why focus on the “DONT DO THIS” instead of show what to do? Look up alexmadeit he’s leaked audio school material thank me later
Yes, that! Thank you for saying it. This whole tight schedule content creation that eventually need to create problems to solve is constantly pushing beginners off track
I can only thank you. After a 15 year break from my hobby doing music with Orion Platinum that was discontinued. I bought FL Studio. Things had moved on alot. Fl studio certainly has. You've made my learning curve so much easier with your excellent presentation and style. I'm really enjoying my hobby once again. Thank you so much. Awesome tutorials. You've a great talent 👌
The end of this vid is so so true, I have found vids on RU-vid that question my whole music and then obsess over fixing it when it was actually fine and I spent hours wasting time.
2:57 This is a perfect example of what a highpass filter is - it differentiates a signal, just as a lowpass filter integrates a signal (I know, big scary calculus...) I generally agree with your conclusions and this is presented fairly, but the old advice I remember receiving was to highpass everything except for bass instruments, which is much more sensible than HPF everything. If you know that you are filtering below the lowest fundamental of a given instrument, it really doesn't do any harm, and you really can clean up some noise (rumble) that may have bled into the recording. Obviously, still a tool that should be used appropriately.
A is the original and B is the one where the high pass is applied, you feel that the kick is less strong, and the other elements, including the mids and highs, shine through better in the mix, this is because even though it wasn't a brutal cut but the fact that the kick is "cleaner" ends up making room for other frequencies, bringing a cleaner sound
This person is professional, as a beginner the video is very easy to follow. I managed to make 1 EP of lofi music with standard chords and following this guy's mixing tutorial, at least it worked in my case. Thank You
Great vid Michael! "A negative AUDIBLE difference" is a very important point. There's so many in-depth analytical tools available to us, this makes it easy for mixers to focus on the science way more than the sound.
The answer for me is B - I’m listening with the iphone earplugs on a train, so maybe not the best surroundings but i get the sense that B has the low cut on, in zero latency/ Natural phase mode - the reason being is because i can hear the 180 degree phase shift which usually translate in an enhancement of the mid-low frequencies, very frequent in what is called the ‘Gibbs effect’, not too audible but that’s what my ear tells me.
For ‘enhancement’ i mean an increase in amplitude in an aera really close to the applied Low Cut. Having a low cut/eq on, doesn’t necessarily mean that there’ll ve a reduction in amplitude of the overall dynamic, as in this case. Phase shift needs to be taken in consideration.
I think there's some advice online that is bad if it's not given with enough context or explanation for why or when to apply it. One example is the common technique offered to beginners to sweep an unnaturally boosted EQ band to "find" problem frequencies that are too loud and need to be cut. Without explaining that this is only useful if there is an obvious problem frequency and you need some help finding it, novices go around hunting for "bad" sounding frequencies. And guess what? If you've got an EQ boosting a frequency band by 10dB or something, you'll find plenty of them! It's a classic example of a solution in search of a problem. Inexperienced mixers need to learn to trust their ears, and not search for problems that may not exist. They are bound to find plenty of real problems and their ears will tell them about them. That's the time to figure out how to solve them, and hopefully how to avoid them in the first place. High passing is a different kind of case, because you're kind of dealing with things you CAN'T hear, but you can hear the effects when all the headroom is being eaten up inaudible of unneeded frequencies. However, you still need to know why and when it's actually useful to apply it. There's a lot of content online in which people say "Yeah, do this," without explaining why or when it's actually useful. They may not know themselves, their just making content out of things they've heard or seen from others. I'm barely more than a novice myself, and my advice based on my own mistakes is "Don't do something unless you really know why your doing it with the specific thing you're working on." Generally novices should be doing less and focusing on more fundamental things like simply getting a good balanced static mix. This is going to offer greater returns in the actual sound of the mix and also help them gain more useful experience than "three weird tricks" kind of stuff.
I've been wondering this forever. Ever since I started HP again I've been hitting -2 LUF easy. When I was just shelving it was never possible. Thank you for confirming this
You give easily some of the most informative and useful advice when it comes to this stuff, I'm glad you're making videos like this and just your videos in general, such a gold mine for a newbie like me 💙
I finally unlearned the habit of highpassing most sounds a couple years ago. Ill Gates has a fantastic seminar called "The 80 20 of Mixing" where I first learned about the phase cancelation problem. Glad to see you are also putting out this corrected version of the information 👍
He's not even discussing phase cancellation here, phase cancellation is only something you need to worry about if you're doing parallel EQ'ing. If using an EQ in serial/as an insert at 100% mix for example, you don't need to worry about phase cancellation.
I was on the fence about quitting music altogether because it just wasn't sounding how I wanted it, not to mention EQing I just couldn't figure out whether I was taking out too much, or too little. This video actually helped me to understand that I might have been focusing far too much on trying to eq the lows off of most tracks apart from bass etc. Thanks!
This is what we have spectrum analysis for: only cut away the _noise,_ not the *signal.* I use high passing on _every_ microphone track because there can always be low freq rumble from your steps on the studio floor - or from thumping the desk - reaching the mic via the mic stand.
Do you really know what noise is? I mean in terms of frequency spectrum. If we could easily filter it out noise when it is present in the signal we would not need low noise audio pre-amps and ADCs.
@@hbo001Does _who_ really know what noise is? Anyone who doesn't know their rumble from their hiss doesn't belong anywhere near a mixing console in the first place. Anyone who wants to *learn* can _solo_ and loop any noise to get a good long look at in a spectrum analyzer. Train your ears _and_ your eyes.
With my earliest efforts I knew little about EQ or compression, except that cutting some of the lower end in sounds made the resulting mix clearer. Along with attenuating at 16000, 8000 & 4000 Hz, It's still one of the most important things I have learned about music production. There is no single rule of thumb for where to cut low end. You need a good audio setup for monitoring low end & you have to use your ears. A gentle slope is best of course. And the linear phase trick is handy.
My guess is: [A] is highpassed, [B] is unprocessed It's more like a feeling since I got no Sub and no large monitors so I went with my ATH-M50x. [A] sounded more "airy" and "cleaned up" to me where [B] sounded "mushy" to my ears. Anyways: Thank's alot for your videos and the effort you put in. Your channel is solid gold 🙏
Thank you for this video. I consider this common sense. Or one would think it is. There is absolutely no reason, what so ever, to leave unwanted frequencies tornadoing through your mix. Especially subs and low end in general. Leave that area for the Kick and Bass. With the exceptions of specific productions where some percussion elements or other low end instruments wanting to play in that frequency range as well. More often than not, I see useless low end information in hihat samples, snare samples, vocals, strings..etc..etc. And highpass filters is one of the first things I do in a mix when mixing for others. In fact, quite often this fixes all the clarity issues.
I can't say anything with absolute certainty. However looking at the recorded audio through a spectrograph view has me thinking that both sides are identical and that any perceived difference may just be because we have been primed to believe there's a difference.
This isn't by any means a hate comment, but besides the clickbait-y title, there's a bit of ambiguity that may turn into, well, bad advice. The general takeaway of the video is that you shouldn’t overdo HPFs, but what it fails to mention is that cutting closer or even above the fundamentals not only is common practice, but usually makes sense in the context of the mix. This could make newer engineers hesitant, if not completely deterred to try it in the first place. It would also be great if you could help further de-mystify the phase coherence in filters, which I think you're in the right track for. Keep up the good work!
I believe A has a highpass on it. It felt like the kick and snare did lack just a little bit of power & Clickiness in the transients in comparison to B. This I believe would be due to the phase shift caused by the non linear cutoff messing with the signal phase :)
Great vid! My take is generally if you have phase issues, you'll hear 'em. New engineers are probably overthinking phase. I only worry about phase when placing mics, or when I hear a phase issue while mixing - which is actually quite rare. Also if you really want to ensure your phase is correctly aligned, there are plugins for that. I have used them, and no longer do.
i wish people would stop talking about phase generally …. there’s so much misunderstanding about what it actually means and there are so few situations where phase shift actually causes problems. obviously if you have two related sounds playing on top of eachother phase shift matters. but for example with kick and bass your kick and bass will never be in phase!!! they’re completely different and unrelated signals. in the same way like this video says the phase shift from highpassing a single track will literally never ever be an issue
Exactly, lowpass and highpass are used to transform the original shape of the sound to make space in the mix and get more control, NOT FOR RUIN anything... Take account if the master are commercially squashed at -5 dB LUFS, the filters are an essential process... thanks.
The issue I find is sometimes I won’t even notice something is causing an issue but I will know there is an issue somewhere or I won’t be able to hear it because it’s too subtle but when I remove it, it somehow makes the track cleaner and fixes major issues I didn’t know existed
I'm going to say B is the processed example, because I can hear the bass coming through cleaner and clearer, which only happens when you get the low frequencies out of each others way.
A had the high-passed filter and B hadn't because A sounded more balanced than B as B had some low rumble going on that might have been messing with the frequency's overall balance. A was much balanced compared to B after lowering some of the very low (almost inaudible but can be heard if you have the ears) frequencies from the mix. That gave the mix a bit more room too. A sounds high passed to me..🎧🎵
We need people to realise we don't hear phase, we only hear the problems and benefits when two or more sounds are played together. So 'phase issues' from filtering is always either too much or too little.
The waveform changes because, the shorter the waveform, the higher the frequency.... so filter does what filter does. The implication of a filter destroying the mix is we mix only with faders and everything else happens in front of the mic. A filter has trade-offs, everyone who mixes should know that. We could, of course lessen the bass response by release time in a gate or compressor, but unsurprisingly that also changes the waveform and phase. Edit: I initially missed 'allegedly' and assumed the video made a point that highpass ruins your mix...
:36 There's there's another thing to be aware of here though which is in the spectrographs the fft band spacing isn't equal. What does that mean? It means that because of overlap in the bands and the fact that you've got one band for all the low frequencies it looks like there's a bunch of energy down there but that's just the curve that it's drawing over the values which would be able to be displayed for example as a bar graph. Try making a sine wave maybe using something like operator in Ableton and do it on a low E1. It's a sine wave so you know by definition it's just one frequency. Now open up spectrum and instead of the standard view set it to just look at the fft bins. You will see what would effectively be a normal distribution If sound wasn't geometric in nature because of the fft bin overlap and you'll see that there are way less bins for lower frequencies. What does this all mean? It means that even if the spectrograph shows that there is stuff down there there often isn't stuff down there.
That is an important point that I didn't address. However, in my example at 0:36, I did actually make sure there was a low end down there by synthesizing a low pulse wave (for the sake of the visual only!).
They sound almost the exact same, but A felt a bit weaker. B just had a bit more beef to it, so I think A is high passed. Overall, the master would sound fine either way imo. Just separate your low, mid, and hi frequencies, crank them up with a super transparent Pro-L, then sidechain them using whatever method you like. I personally use Fruity Limiter for the Kick, followed by trackspacer to clean up the spectral space with the kick sidechained through it, then i add another Fruity Limiter for the snare after that. It'll be loud, but just bring the slider down to balance the volume however you want in the mix. It's pretty meticulous, but it's SUPER clean and works for me. Route that through a Post Sidechain, then trough a pre-master, clean up the mix in the premaster, then go through the master and do your thing. Tbh, as long as you clean up your low end mix, your mix is gonna turn out great. Just a ton of very transparent limiting. I don't really compress much unless I'm sound designing.
If you do not hear a problem, there is no problem. Especially if you have trained ears and good monitoring - also, listen to in your car, mix in your songs in playlists to get a good comparison.
It's worth adding that the high-pass filtering technique is shoved in our faces as though it's something that absolutely must be applied whenever making mixes otherwise your mix will end up being "wrong". Instead, highpassing too much can simply take away the "life" from a song because sometimes those low frequencies just add a bit of body to the sound, whatever it may be, and amateur music producers will end up having rather thin and weak mixes.
I think B has the filter, reasoning is it pushes the higher low frequencies to be louder and with some master chain compression those higher frequencies stand out more causing the mix to be a bit more imbalanced
I've only listened quickly via my smartphone inears, but B should be with a high-pass filter. At least it sounds less precise and "fatter" to me than without the highpass filter. I've known about this topic for a long time because it's also a big issue in loudspeaker development. A much more important point, in addition to the phase, is the time consistency. With loudspeakers, the so-called 'group delay' is measured. This is a frequency-dependent impulse response. With steep IIR filters, the problem is that this impulse response rises very sharply at the frequency of the filter. Put simply, this makes it sound muddier and less precise. This is not only an issue with high-pass filters, but a general problem with small-band EQs. Cool video - thanks for making the topic more transparent for the general public!
My guess is B was high passed, sounds louder and the clarity of the high end is better. Louder because there's more headroom for you to drive more dB's into your mix before it hits the limiter. Clarity of the high end is better because there's less destructive phasing occurring when all the waveforms are combined.
I know this is old, but just wanted to comment anyway. "B" is the mix that is high-passed. Can tell because less sub frequencies are hitting the master compression/limiter. The result is cleaning up the pumping effect coming from the kick drum, and allowing the overall mix to feel less heavy and giving more clarity. Though I could be wrong, as my open-back headphones aren't great for sub frequencies. Sometimes I get tricked by the low-bass, rather than sub because of that... The main issue I ran into with OVERDOING it with the high-pass on the master, was that due to the phase shift, the input on the limiter INCREASED, even though I was only trying to "remove" frequencies. The mix ended up being hit harder by the limiter and sounding more squashed, even though there was no audible difference in level coming out of the EQ... First time I'd come across that and man did that ever throw me for a loop. Because of this, the answer could be "A", especially if linear phase is not enabled, because the phase shift could be causing a higher level to go into the limiter, which would cause the pumping. Without a doubt, the highs are clearer on "B". Dang, sucks to be wrong. But even after viewing this track on a spectrum analyzer, don't even see a difference either!
Comapring A and B I feel that B would definitely be the Hipassed sample if you listen closely the Bass on the Kick seems to be more tamed neutral sounding compared to sample A where you feel the kick thumping in the low end.
Yes! Absolutely! Something I've recently changed in my production was ridiculous low cuts. Just cut the bassy part of a hihat bro, let that transient juice come out of that damn hat, if you want white noise hats just make it in 3oscx instead.
what i can learn from this video is: use linear phase mode, high pass less agressive when the sound you are eq'ing a nice sound with a usable low end, BUT when you compare those two kicks well, actually the phased one is what i wanted, reduce the power of the kick some times, is not always bad, sometimes you want to reduce some power or exesive low end, so its not always a bad thing.
"A" had the high pass filter applied. That mix sounded (and felt) slightly cleaner AND louder. Conversely, "B" had a bit less energy; and sounded "muddier", more diffuse.
Most speakers don't play below 35hz, however some giant festival speakers do from what I hear. So SOME producers actually include 20hz frequencies. But it's extremely rare. And for stuff like jazz, frequencies between 40-100hz add character to kicks, but they're not very important, whereas for dance they're the most important frequencies
Mix A has the highpass on it. Mix A sounds more clear/less muddy but also doesn't hit me in the chest the same way as Mix B. Mix B almost sounds like it has a slight pumping/sidechaining effect to me as the bass volumes dips when compared to Mix A.
I'm gonna go with B. It sounds like there's more low end, so the first guess is it's definitely wrong, but good high passing eliminates inaudible frequencies; if good high passing is done, you should actually hear *more* frequency content, because your limiters and such are only triggering when more content is there.
The phase shift is only an issue if you mix the HP'd signal w/ the dry signal. Or if you used multiple mics/channels on a source and don't HP all the channels together. If you understand how your tools work & how to use them properly there won't be any audible problems.
I think a huge misunderstanding is that multitrack recordings can be messed up (phase shift) quite easy using getting rid of lower frequencies. Imagine a fundamental which has harmonics on top of itself. Getting rid of that fundamental makes the harmonics shift in time but also compressed in time. This does not have to be catastrophic but you simply cannot cut the fundament without consequences. I need to make a visual maybe to explain this.
A is the high passed Mix as the kick thump is definitely more present there compared to B. This is because (when correctly high passed), removing unwanted low frequencies increase the perceived volume of our tracks and helps to remove the bass mud.
Yup. I had to learn this the hard way... I listened to this one youtubers advise on this and stopped using low pass filters... overtime I found that my mixes not translating well from speaker to speaker / were always 'off'. One day I decide 'what if I just don't care at all about the phase on the low end?' Boom. That solved so many problems. I hate that youtuber with a passion now.
Amateurs often stare at charts and meters and draw conclusions about what they are hearing. It's the usual insecurity and problem creation that separates them from creative and skilled pros.
Excelente reflexión y consejo. Estoy totalmente de acuerdo. Para mí el track original es el A, el B lo escucho con un ataque artificial, no solo hay una diferencia en el low end sino timbrica en los hi hats. En comparación el A suena más natural.
The Phase shift is also only a problem if you mix it with an unphased variant of the same (e.g. with kick bleed on other mics). Guess what. If you high-passed your other mics that can't happen anymore. Also, the waveform looking different: Yeah? That is kinda the idea of any effect - if the waveform looked the same before and after the effect did nothing. That being said you can use resonant Highpass filters to ADD low end in very specific spots, not uncommon e.g. in Eurorack world. As a professional freelance mixer for cinema, it is absolutely normal to slap a HP filter on literally every track and remove/shift it only when needed.
A has highpass enabled, because lowend on A has a tiny bit of delay and slight boost of 40-50hz, which is probably the result of slight phase shift, while B seems to have more balanced lowend and it's more on time
The correct answer is (and this is demonstrated by all the different answers and opinions here in the comments) - it doesn't matter. Real listeners of the music (not mixing folk) will likely never know the difference or care.
B is processed I believe since even with linear phase EQ you always get a little bit more punchiness in the sub region of your mix especially when it comes to Kick drum. The sub will have a bit more transient therefore it sounds punchier.
Mix B, the kick came through the mix more, there was definitely an auidible difference between A where the kick feels more muted. So I'm gonna go with A is highpass
end of the day what matters is what we hear, it doesn't matter what phasing are in background none of the listener will start investigate weather there is some phasing occurs in music, they will care with what they hear..
I may be wrong but this is what my gut's telling me... I believe A is the high passed mix, although it sounds at first the other way around. Many speakers struggle with anything under 20hz, and as well is around the limit of the human ear (around 20 hz). In a mix, too much energy in this area can destroy the mix, and even take away the audible frequencies around it. Speakers/headphones etc perform better at producing that nice bass response when those ultra low frequencies are removed. For your low end to clean up, you must first clean up your low end. Edit: I'm listening on open back headphones... not sure if there's audibly more sub sound on B as some others have pointed out listening to on their monitors, but A sounded more audibly deep and clean through my open backs.
Really nice video as usual🙌 As for the mixes, I feel both the Mixes A and B are the same. I couldn't hear any difference honestly. The Mix B seemed slightly different but I feel its just because of the different bass note. Keep up with such videos.🙌♥️
good info for troubleshooting when mixes dont sound boomy enough. tho, honestly. i would never call a 40hz highpass a "ridiculous" move. it is all about taste, and if u know what u do go for it. for me the highpass almost sounded like using a bit of a compressor. really, its really up to u what u do. still good info for troubleshooting whith mix issues.
Like pretty much everything in life, moderation is key here too. HPF is a tool, it's neither good nor bad inherently. A good golden mean approach is your best bet here too. Use this tool, like all other tools under your belt, with purpose. Don't just slap it on everything, and also don't avoid it just because it can mess things up when used incorrectly. Use it *if* you need it. When you start doing things without purpose, that's when the funky stuff happens.
A is the original while B is the high pass filter. Reason because the striking transients of the kick and sub is lesser than that of A. Which is just punching through anytime the kick comes up
The HighPass on mix A gives more room for the mids and highs to push through so the overall mix feels brighter and less muddy. This also makes it appear slightly louder.
B is high passed with a linear phase eq. The low end sounds "tighter"-- like what it sound like when you high pass something. And I say linear phase specifically because linear phase has a sort of "flattening" and "smoothing" effect that I don't hear with minimal phase :)
My unsolicited 2 cents: The general current canon (popular loud music) says to high-pass everything but your kick and bass at some frequency above the fundamental octave. this intersects with another popular maxim about removing any content outside of a selected frequency bandwidth for a given instrument. panning plus frequency sculpting are the primary tools for placing sounds in a mix. rendering applied processing to a visually alignable audio clip can make phase shifts obvious. phase incoherence, however, is an undesirable byproduct of extreme settings on inappropriate filter types. under the hood, filtering can be achieved by a variety of means, not only the traditional FIR and IIR filters, but by a wide variety of techniques ranging from frequency domain bin manipulations to wavelets to convolution, but i digress. the fundamentals of interference (constructive/destructive) and phase response exist and can be exploited, in any case... the appropriate usage of various kinds of filtering is based upon understanding the characteristics of the tools used. cargo-cult tool fetish, though rife in audio "engineering" is not an actual byproduct of engineering, which looks at the context, rather than using hammers on phillips head screws etc.
Lots cof old school mixers used unheard low end high boost with q paired with the same frequency and Q in a low cut to get a low end smear in phase relationships cuz they said it sounds good. Next gen went for krispy bass and the old way got lost. These days you can have it either way. I am personally preferring clean crisp bottom with not a lot of mud, but clients can order what ever they want and I eill mix / master their favorite sauce.
B should have the hi pass filter if im hearing it correctly. A gives that specific feel in the ear that very low frequencies give you (using earbuds btw). It's less present in B.
I suggest B is the one with the High pass Filter applied and here's why, So in sample A the low end is a lot more fuller and punchy, but when we compare it with sample B , sample B is punchy but lacks the fullness in the low end especially around 50 - 80 Hz. So that is what I have been able to conclude !
A = The lowend of the bases/kick roughly 0 -18hz you can hear it's a bit hard on the mix but very unnoticeable ,even the waveform there can be a slight difference as on B the kick is more subtle but effective together with the mix everything blends well shows the mix is high passed letting all the mids and highs more free and clear I stand to be corrected as I am using just ordinary headphones but i just closed my eyes and let the ear do the judgement
It is A, the high hats show more presence in the mix but, when B is played you can totally notice the slight and shuttle shift to more low end as it seems like the bass arp is louder and more noticeable so, in conclusion A has the high pass filter enabled and B does not.
