A little trick with ProQ3: To boost the air band, move to 30Khz and do a ~30dB high shelf boost. The slope will gradually fall off and will have a smoother colouring effect when it reaches 20Khz and below than if you just boosted from 20Khz or at 12Khz or something like that.
Well, it works. I hear it and I like it. Explanation is excellent and pace of this show was perfect. I am an recording engineer with a couple decades experience and now I know something new. Subscribed.
My buddy is the drummer for this band!! Just trying to get some mixing tips, and I hear his band's singer bust out. Lol, what a trip. Planet Mercury is the name
The point of using the la2a is because it has a high frequency filter to where it only compresses the high frequencies. While normal compression can work sometimes if you don’t have that filter you can get really apparent and strange artifacts within the parallel. You’re better off using a compressor that has a SC filter built into it, doesn’t necessarily have to be a la2a but the 2a does work very well.
I know Andrew Scheps mentioned that where I saw him doing this trick, but then he didn't have the side chain filter on, so he was just using it as a typical compressor. And really, it makes sense that you wouldn't necessarily need the side chain filter because the whole idea of this trick is kinda sorta doing what the sidechain filter on a compressor does (just in a more flexible way). So using the side chain filter would help to exaggerate the effect, but you could also then just push your EQs farther to achieve more or less the same thing. And I'm assuming you know this, and it's just a wording thing, but I just wanted to mention this in case anyone else is reading this who might not understand how the sidechain filter works, but the compressor is still compressing the low end, it's just not looking at the low end when determining when/how much to compress. Again, I'm not trying to suggest that you didn't know that! Just a lot of people word that in a way that can be confusing to some people. But yes, I agree that the la2a does work well for this trick! And the sidechain feature can help to further exaggerate the effect. But I just like to try different flavors of compression.
Better Mixes I see where you’re coming from, I’m just speaking from personal experience. I’ve used 1176, 3As, Fairchild’s, boutique compressors, stock, etc. and the ones that didn’t have a SC would produce artifacts whereas the ones that had it didn’t. Again not all the time, but there were quite a few cases where this did happen and when I switched to a comp with a SC it immediately went away. This is more of just a warning to anybody looking in to this trick to watch out for that because you might not notice it at first and think it’s something else which causes you to have to trouble shoot a bunch of stuff and waste time
I believe this technique is referred to as, 'pre-emphasis - de-emphasis EQ', ie: whatever you do 'before' the compressor, you do the opposite 'after' the compressor. Waves Abbey Road Saturator has this feature built in and so does Kelvin from Tone Projects.
Yup! That's exactly what this is! I haven't tried those plugins, but it can be a really fun and powerful technique to use around saturation, so I might have to try those out!
Thanks man, appreciate the video :) i recently got a mixer & have had some rack units ive wanted to hook up & rock, i got a bunch of ideas but this helps me get an idea of some things i can do on vocals :) have a great week!!!!!
2:42 just on this alone it makes me think you're just replicating the "vocal trick" Mutt Lang did with the Dolby tape hiss remover. Audiothing has a plugin, Type A, that accomolishes this well.
It definitely could be a similar idea! The dolby trick (or at least the version I know) was basically just using the encoding stage of the dolby, and my understanding is the dolby (and most tape hiss removers) were based around a similar idea as this of using emphasis and de-emphasis curves to boost high end going into tape and then removing it after, taking a lot of the tape hiss with it. So yeah, definitely a lot of similarities! I'll have to try the Audiothing plugin at some point to see how I like it!
soo why don’t we just do this with a single multiband compressor in parallel? It makes me think that Shepp made this technique before multiband compressors were a thing in a DAW, because that’s essentially what we’re doing here.
While you could use a multiband compressor as another way to address the same issue, it's ultimately a different thing and would sound different. It could be better or could be worse depending on the situation, but still, it would be different. This is really closer to using a single band compressor with a highly tweakable sidechain (something like FabFilter ProC2 or I think Arouser has a pretty flexible sidechain as well).
@@eightrice Again, they're completely different things. Saying that one is straight up better than the other is like saying a saw is a better tool than a screwdriver. It depends what you're trying to do. I like using multiband compressors on vocals quite a bit. In fact, I'm pretty sure I was using one on this vocal, but it's a different thing entirely.
@John Pagale seems like it is compressing the range of frequencies that is left post eq, so the same thing. To me the salient difference is that it is easier to ride/automate when you have them as separate channels
I'm a huge fan of Spectre! Such a cool plugin! The results would be a little bit different than this method, but yeah, it would ultimately accomplish the same goal in a simpler way!
Wow That's so interesting and super. I'm getting close for what I'm looking for. But still couldn't figure it out what I really want when it's vocal mixing. Crispy vocals, lime saturated but not like old Cassete. Sound warm and crispy. And I'm getting closer by this trick you show. Thanks a lot
No problem! I'm glad I could help get you closer to what you're looking for. Just keep messing around with it, and you'll find the sound you're going for!
Not exactly. You absolutely could use Pro-Q3 in place of a multiband compressor in most cases, but in this case, we're not actually doing any multiband compression. Everything is being compressed together by a "normal" single-band compressor; it's just that the compressor is primarily being triggered by the upper midrange. If you have Pro-C2, you could use the sidechain section to accomplish this in a more straightforward way though!
how is the processing on the vox bright different from a multiband compression? I get it that the EQ control could be a lot more detailed but is there anything other than that? Thanks.
my question as well. he did mention that it's kinda like doing a parallel multi band compression but didn't talk about how they would differ at all. I would guess it would get close enough
@@pco2004 I figured that when the compressor is triggered by the bright part it also compresses the low end while multiband compression is only processing the high.... but still don't know how much of difference would this make
Like you said, when using a multiband compressor, you'd only be touching the highs (or whatever you have it set to compress) as opposed to with this method, it's still compressing the whole vocal together. This really helps to give it that super tight, in your face kind of sound. If you were looking for a simpler way of doing this, you could try using a compressor with a super flexible sidechain section (like FabFilter Pro C2 for instance). That would be much closer to this sound without the need for routing the vocal to an extra track. Multiband compression can still be really cool on a vocal though! In fact, I probably use it on vocals more often than I do this technique, but I use it for a different purpose. It can really help to keep a vocal more tonally consistent from one word or syllable to the next, so there are no words with too much low end or high end compared to the rest.
This is pretty different than just adding a bunch of treble. Like yeah, you could do that too, and it might do what you need it to do, but it won't sound the same as this method (for better or worse).
It's adding the vocal in parallel with some hi end. It's going to stand out because it's louder overall. As presented in the video. There are a thousand ways of doing this. I just hate how people are going to be duped into thinking this is some "secret" and it's not. It's basic audio. I get that it gets clicks but it's not anything special. @@BetterMixes
@@hinder10709 Yes, it's louder; that happens any time you do something in parallel (at least if it has to be done on a second fader. Obviously that's not the case if you're just using a plugin with a mix control). That's just kinda a side effect. You can do level-matched comparisons, which I've done, and I'd encourage everyone else to do when first trying out a new technique like this, but saying that it's just making it louder and brighter is entirely ignoring the compression aspect of this, not to mention the actual "trick" going on here which is using emphasis and deemphasis curves around the compressor (this can also be really cool when used around a saturation plugin!). As far as tricking people, I'm not saying anywhere in the video that this is some magic thing that automatically makes your tracks win Grammys, I just saw Andrew Scheps talking about it, thought it sounded interesting, tried it, thought it was cool (even after level matching), and figured I'd share it with some people. But again, if you think this is just EQ and volume, go play around with emphasis/deemphasis curves. Dan Worall has a great video on this subject where he explains it better than I ever could, so go check that out if you get the chance, but it can be a great technique to get some new and unique sounds
It's presented as a "secret" so, yeah, that's somewhat deceptive. Making something louder isn't "kinda a side effect" it's part of what is making this setup seem 'better' and makes it 'cut' through a mix more. It's not just the extra hi end that gets compressed and mixed. It's all the frequencies that get louder through constructive interference. More so in the hi end yeah, but overall. And yes, this is just eq and volume. That's all we as audio engineers really have to play with. Levels (overall and specific bands) and time. I'm aware of Worall's videos. @@BetterMixes
@@hinder10709 "And yes, this is just eq and volume" Again...ignoring the compression entirely. And more level IS a side effect of doing parallel processing. Before plugins compressors had mix knobs, if you wanted to do parallel compression, you'd have it on a second fader. By doing so, it would get louder. Does that mean that every engineer who ever used that technique was an idiot who was being tricked by level differences? Probably not. Same thing here. Try setting this up for yourself, and then make a third channel with a level boost to match the volume of the combined 2 tracks in this technique (oh, and make your deemphasis curve perfectly negate your emphasis curve so you're not getting the extra top end). Flip between the two and hear for yourself, they won't sound the same even when level-matched (and without the top end boost). I did this for myself when first trying out this method, and it's not the same. Heck, go a step further, just do a level matched AB of just a compressor and then one with emphasis/deemphasis EQ. Once again, they won't be the same. You may or may not like the technique, that's entirely up to opinion, I personally don't use it too often anymore, but every once in a while I will, but saying that this is just EQ and a volume boost is incorrect.
Sounds even better and easy to fit into a mix if you have duplicate vocal tracks and have one with some bottom and top taken away and the other with all the bottom removed and eq’d at the top end for the sharpness and clarity. Then balance the two tracks to taste.
This is not a good idea because you are modifying your original vocal track, creating phase issues with the duplicated processed copy and you are not adding anything to your signal to make it louder. So not a good alternative to this technique :(
To record these videos, I use a program called ishowu. It's a little buggy, and I think there are better versions out there, but it gets the job done. Then I edit with Final Cut Pro.
alternatively use TDR Nova ;) it's a free vst plug in equalizer (just like fabfilter) which contains a build in compresser. you can create frequency assigned compressors with it. great piece of software :)
I love Nova! But I assume you're referring to the dynamic EQ built into it? Cause if so, that's something a bit different (although it could be used to address the same problem in a different way). This is really more similar to a compressor with a highly tweakable sidechain. Unless that's what you meant and Nova can do that, in which case that's awesome, and I did not know it could do that!
Reaper's stock compressor, "Reacomp," has low- and high-pass filters. Why not just use a filter? Put the high pass at like 1100 and the low pass at like 10k? Not to mention, "ReaXcomp" has multiband . . .
That could definitely get you close to the same effect! Obviously, it's more tweakable using a full-on EQ (or something like Pro C2 which has a full EQ in the sidechain section), but the sidechain filters might be all you need for the sound you're going for! I wanted to show how you can do it with any EQ and compressor for people who don't have Reaper, or just if you want to use a different flavor of compression, so you're not just stuck with the sound of Reacomp. This way, you could use an 1176 or an la3a or whatever!
@@BetterMixes Hey, while I have your attention, do you think you could maybe critique my mixing? I've had a studio for a little over a year now and am still learning, and all the time I ask people to come listen to my mixes but so far nobody ever will. I'm trying to figure out why, after eight years of uploading and a year of actual studio quality production value, I still have only 263 subscribers . . . because I doubt it's my playing. So, I figure the better I get at mixing and mastering, the more generally appealing my channel will be. Anyway, thanks for the reply.
Hmmmm sorry but I probably missed something there. It seems to me that the double eq + comp you are doing is simply a multiband compressor or a compressor with eq inside,without the need for a second eq, which is better. You send your signal to a mb compressor with just highs band then you use the out level of this to mix it with the original sound. Right ?
Not exactly! The key difference between this and multiband compression is that with this technique, the whole vocal is being compressed, it's just primarily the highs that are triggering the compression. With a multiband compressor, you'd ONLY be compressing the highs (or compressing them independently from the rest of the signal). You could get a very similar result by using a compressor that has a highly tweakable sidechain filter (like Fabfilter Pro C2 for instance).
@@BetterMixes Thank you very much for your answer ! I like to check the tricks with my ears but also to understand it from an engineering point of view. Here I understand that you try to enrich the highs of the spectral response of the vocals, by adding the original track and an alterred or modified version of the highs. This could have been high pass + saturation, high pass + slight reverb, high pass + tape, even maybe with a simple high pass analog console emulation would maybe do the job like a pulltec or SSL, or as you did a high pass + compressor (I still think you could use a mb compressor, tuning several bands, I simply said 1 high end band to make my message short for YT, though in each band this would have been quite flat which is less precise and maybe less natural that what you did. You could also use a compressor with a high pass in it such as the Kotelnikov, I do not think you need to bump the highs in an eq, you just need to high pass and increase the compressor input, or a dynamic eq). What I miss I think is why you add a second eq to try to so to speak "restore" the low ends you removed. You will not restore it as it was, because nearly removed this part with eq and added a compression on it, which means you lost most of the information in the low end, so when you try to restore it you basically just add noisy signal, you cannot restore signal you have lost. And it seems you do not need to restore that part, you just need to blend your highs track with your original sound, you do not want to add back a modified low end to your original track because in this case you will also enhance the low end which does not seem to be your purpose. Or maybe you do not do a parallel track, which would explain why you try to restore the low end ? But I do not thing this can be what you do because too much information is lost after high pass+comp so, you certainly do that process in a parrallel track and blending after. Sorry to bother you I really want to get the insights, this is how I remember things the besst ! :)
@@johanjof5613 I add the low end back in because I still want the compressed parallel chain to be full frequency. It's kind of like cranking the high pass side chain on a compressor. The compressor is then only reaticing to the highs, but the actual signal still retains the lows. You certainly could just leave the low end out in this trick, but I think it would start getting harsh pretty quickly. And one thing to keep in mind is that the low end isn't gone when you EQ some out. In this particular example, it looks like I used a shelf that maxes out at about -12dB somewhere around 100Hz, so you could easily restore that low end with a low shelf boost at the same frequency and slope. You can try that for yourself, try cutting 20 dB with a clean digital EQ followed by another instance of the EQ doing an equal but opposite boost. It should sound exactly the same as it did without either EQ. And that's the power of this trick. Now try sticking a plugin between those two EQs. Could be a compressor like in this example, or try different saturation plugins. You might find some really cool sonuds by affecting what that middle plugin is reacting to withoutreally messign with the overall EQ balance of the source. But going back to restoring the lost low end, even if you were to use a high pass filter instead of a shelf, there's still plenty of low end that can be brought back. If you were to use a 12dB/octave slope and set it somewhere around 500Hz, let's say, you'd only be cutting 12dB at 250Hz and 24 at 125, which is certainly a lot, but not so much that that information is gone forever. If you were using something like a 96dB/octave slope, well that would be a different story!
You don't necessarily need to print it (you can if you want), but doing some EQing, compressing, de-essing, etc. on the main channel first is probably a good idea!
Ultimately, this method is only ever compressing the vocal all together as a single band. The compressor is just more so listening to that upper midrange, but it's still compressing the full frequency range together. With a multiband compressor, you'd be compressing the upper midrange completely independently from the rest of the vocal. So this method is really more like using a compressor with a tweakable side chain EQ (like Fabfilter Pro C 2 or Arouser or something like that). Both methods could get you cool results! They're just different.
It's really much more similar to using a compressor with a flexible sidechain section (like Pro C2 or Arouser or something), as the whole signal is being compressed together (as a single band). The upper mid-range isn't actually being compressed any more than the rest like it would be with a multiband compressor, it's just that the upper mid-range is primarily triggering the full band compression.
Always curious if this causes any type of flanging. I know with analog gear it would due to the time delays caused by the wire and console, but with digital not sure
It is dependent on the plugin algorithms you use I think some plugins that attempt to emulate the subtleties of analog gear will cause some phase in-coherency. One option is to render the “scheps” chain on a track and then bounce it in place, so you can actually see the waveform of the main vocal and the parallel vocal next to each other and accommodate for phase differences
In general, as long as you have delay compensation turned on in your DAW, it shouldn't cause any issues. There are some plugins that don't play well with parallel processing (multiband compressors typically don't as well as some random other plugins), so just try it, and if you happen to hear it getting phasey and weird, just try swapping out the plugin you're using.
@@BetterMixes You should compare it to using linear phase EQ. Theres the Mike Shipley trick of HPF at 8-9k, compressing the crap out of it and adding it back it - and the difference between linear and non-linear EQ is immense - (to the point where non-lin is unusable). This trick is fantastic on dull acc gtrs too.
@@minimoog4236 Yeah, when using high pass of low pass filters, linear phase mode will almost always sound better. With bells and shelves, it doesn't get nearly as weird in non-linear mode, but you're right, it would still be interesting to compare the two! I'll try that on a mix I'm working on tomorrow!
Can someone extrapolate on the idea this would cause possible phase issues with analogue gear. Since you are just eqing and then your compressing why would this cause any time issues? You would be eqing first and then compressing two audio tracks, why would they cause an issue? Maybe I missed something in the vid. I would assume there would be no delay worse than a double vocal take. But yeah I’m confused to the issue of delay with analogue gear , seems overstated to me? In general am I going to run into delay issues with going More hardware based? I honestly wouldn’t think this would be a real issue but this is something I’d like to clarify
Ahh, cool technique! I wonder if this was the inspiration for Denise audio’s most recent plugin Dragon Fire, as the “push/pull” application is essentially a side chain eq pre and post compression. Very interesting
@@BetterMixes it’s a very unorthodox compressor and breaks away from traditional analog-style emulations or even more modern UI/IX compressors on the market, but I am personally loving it! Highly recommend
I assume you mean using a compressor that can take a sidechain input and creating a mult of the vocal, EQing that, and sending that to the sidechain? Cause that would absolutely work as well! I just like this version because you can use any compressor at all, even if it doesn't have a sidechain input. Plus, for the sidechain version, I'd need 3 tracks, the main vocal, the EQ'ed mult, and the bright vox track as opposed to this that only needs two. But either way would totally work!
@@BetterMixes No, that's not what I mean. This technique dates back to early pop radio and every engineer I've ever shared it with has gone on to use on their mixes all the time. It's quite simple actually. In the final stage of your band track mix, insert a stereo compressor - which has a sidechain input - just before the band track hits the master faders. But send the vocal(s) DIRECTLY to the master faders. ALSO... use the vocal track as the input signal to the side chain of the band's stereo compressor. (using an aux send tapped off of the vocal) i.e. the band track should NOT drive the side chain - ONLY the vocal signal. Now, the vocal track will drive the compression of the stereo band mix - but not of the vocal itself since it's going directly to the 2 mix. With about a 3 to 1 ratio on the compressor, and relatively fast attack and release, the band track will dip invisibly underneath the vocal but still give a full sound. Set correctly, your ear won't be able to tell that the band track has been effected at all. Let me know if that's clear.
@@ronnyjay100 Oh yeah, that's a great trick! I actually thought I already made a video about that, but I just looked and it looks like in that video I was just sidechaining the vocals to the guitar bus, not the full mix. But I especially like that trick for pop songs where you want the vocal really out front.
It's still accomplishing pretty much the same thing. The main idea is the emphasis/de-emphasis. It's about causing the compressor to react differently, but then restoring more or less the original frequency balance after (or in this case, I left it a little extra bright, but it'd be the same idea with an exact opposite curve). It'll sound a bit different with a pultec or a neve or an api, but it's still the same concept. Not to mention, in most cases, pultecs are quite a bit cleaner than a lot of people think. There's certainly some harmonic activity, but it's not super crazy and not enough to fundamentally change what's happening in this technique.
@BetterMixes word 👍 yeah I hope you didn't take that as a criticism, it wasn't meant that way. This was a great video 🤘 And would love to hear what you think about Type A! Changed my world haha.
You could definitely use a multiband compressor as another way of addressing the same issue, but it's ultimately a completely different technique and will therefore sound different (maybe better, maybe worse, depends on the situation!).
Right around the 9 minute mark you can hear it with everything. Or, if you look up the song Honest by Planet Mercury, you can hear how it was used in the final mix!
This is really closer to using a compressor that has a highly tweakable sidechain filter (something like FabFilter ProC2 for instance) than it is to multiband compression. With a multiband compressor, it would ONLY be compressing the upper midrange area while leaving the rest of the vocal alone. With this method, it's still compressing the entire frequency spectrum, it's just REACTING more to the upper midrange. That being said, multiband compressors can sound great on vocals too! I like to use them to sort of even out the tone of the voice from one word or syllable to the next. But like I said, just a different thing.
Not sure that counts as super compressed exactly, it was more kind of turned down in volume. You needed to get that needle moving to reduce dynamic range.
Yes and no. As I'm sure you're aware, the meter doesn't show even close to the whole story when it comes to compression. For one, once you're getting to the extreme end of the meter, it can be really hard to see what's going on. Then on top of that, VU style meters, like the one in MJUC, are really pretty slow, so the actual compression is moving around much faster than what's reflected in the meter. You can hear pretty clearly that the vocal is way more compressed with the plugin on (and the autogain in MJUC really helps to hear it since it's doing a pretty good job of making up for the volume change in real time).
If it's a highly tweakable sidechain (like in Fabfilter Pro C2 or Arouser or something like that), then yeah, it would be very similar. If the compressor just has a basic sidechain filter, you could do something in the general vein of this, but not quite. Plus, this way, anyone can do it, even with stock plugins if they don't have a compressor with a tweakable sidechain!
This is a great technique and you have shown it's usefulness. I would caution trying to describe it in the way that you have though. You keep using the sidechain compression vocabulary where you describe it as what the compressor reacts to and what it does not. But using this technique literally removes frequencies from the processing itself. So that they are not getting compressed at all. In the case of sidechain compression the entire signal is being affected by what you are telling the compressor to do. This technique is much more surgical in that it does specify what will be compressed and what will not. The drawback is that you literally have to take a scalpel to it twice in order to accomplish this. That's kind of a win lose based on the situation
You're right about this being much more surgical, and also that it's kinda a bit more of a pain since you have to EQ twice, but that's not true that the frequencies we're ducking down aren't getting compressed. The entire frequency spectrum of the vocal is being compressed together as one. So when we sort of "undo" the first EQ after compression, that low end is still compressed, the compressor just wasn't reacting as much to it. Now if we didn't use the second EQ and left the low end out, when we blend that back with the main vocal, then you could say that we're pretty much only compressing the top end in parallel (which could totally be a useful technique as well), but that's not the case here.
@@BetterMixes actually, the entire purpose as I understand it, is to take those frequencies you don't want above the threshold down to an out of reach level, so that when the Compressor engages, it will only be attenuating the part of the spectrum that is still loud enough to be affected by the volume ratio cut. Something that cannot be restored after the fact. But the volume change of the EQ can be
@@rhemaman That's not how a "normal" (single band) compressor works. It sounds like you're describing a multiband compressor or maybe something kinda like the Dynamic Spectrum Mapper plugin (or another spectral compression kinda thing). A typical compressor has no idea if it's seeing low end or high end or midrange or anything else. It just sees one single level, and if it's above the threshold, it compresses. So even if a signal is like 95% high frequency stuff with just a hint of low end, the compressor just sees the combined level of the entire signal and compresses all of it together as a single thing.
@@BetterMixes whatever is above the threshold is the only thing that is loud enough to get crunched. The other stuff is generally untouched, with exception of when they're happening simultaneously. And even in that case, the headroom ratio will also translate to much less being compressed of this frequencies, because they didn't hold a large footprint in the mix when everything got crunched. This is why the softer parts of a track always get louder after compression. Because they actually benefit of more headroom by the taking away of the other stuff. This relationship of above and below the threshold is what's being exploited by the technique. And to take advantage of it you have to be extreme with all the moves; as you have shown.
@@rhemaman Yes, whatever is above the threshold is what get's compressed, so, in theory, if there was a part of the vocal that was ONLY low end, it likely wouldn't get compressed with this setup, but that's never the case with a vocal (unless it's got some crazy filter effects or something obviously). The low end and high end and every other part of the vocal are all happening simultaneously, so when you say "with the exception of when they're happening simultaneously," that's literally all the time. Some words will have a little more or less low end than others, but that doesn't change the fact that the compressor is only looking at the combined level of the signal. Think about a compressor as you with your finger on a fader (this is far from a perfect analogy, but close enough). Someone tells you that any time the level on the meter goes above -10, you pull the fader down, turning the volume down. You don't know whether it's mostly high end or low end or what, you just see the level on the meter (the combination of all the frequencies present in the signal). So when you pull that fader down, it's turning down the lows, mids, and highs all equally. Now let's say you cut 10dB of low end out of the signal. That will change when the signal goes above that -10 mark, but when it does go above, you're still turning down the lows, mids, and highs, all equally, not just the highs. That's what's happening inside the compressor. It's reacting more to the highs, but when it compresses, it's still compressing everything together.
Nope, not exactly. While you could use a multiband compressor as another way to address the same issue, it's ultimately a different thing and would sound different. This method is really closer to using a single band compressor with a highly tweakable sidechain (something like FabFilter ProC2) as it's compressing the entire signal together as a single band. It's just reacting primarily to the upper midrange. Using a multiband compressor, it would ONLY compress the upper midrange independently of the rest of the signal.
You didn’t play the song long enough or A - B it enough or turn the vocals up loud enough for the listener ( us ) to really get a full grasp of the effect . It’s hard to hear the full effect during a 6 second playback . I’ll stay tuned , thanks 🙏
Nope! It's not gone, it's just turned down. You can try it for yourself, do a massive cut somewhere, like 20 or 30 dB or something, and then put a second EQ plugin right after boosting the same amount at the same frequency. It should sound identical to how it started (if you're using a an EQ with symmetrical boost/cut shapes). Even if you used a high pass filter on something, the low end PROBABLY isn't gone (unless you used a very steep slope or set it to a super super high frequency).
Yes and no. You could absolutely use a multiband compressor as another way of tackling the same issue, but it would ultimately lead to a different sounding result. This is more similar to having a super tweakable sidechain on a compressor (although not exactly that either). You can use this emphasis / de-emphasis technique with a lot of things too, not just compression. Try putting a saturator in between two opposite EQs for instance! You can get some cool sounds.
There are a few ways you can do it. A lot of the time, I like to but my verse vocals and my chorus vocals on different tracks so I can have different settings and sounds for the different parts of the song, so that way, I can only send the chorus vocal track to the send. Alternatively, if you're in Pro Tools, you can control-command-click on the send's mute button, and then in the edit window, you can write in the automation on the vocal track to turn the send on and off throughout the song.
You could use a multiband compressor as another way to address the same issue in a mix, but it would ultimately sound different. With a multiband compressor, you'd ONLY be compressing the top end or upper midrange or whatever you set it to. With this method, the whole vocal is being compressed together, it's just that the compressor is reacting much more to the top end. This would be much more similar to using a compressor with a highly tweakable sidechain section (like FabFilter Pro C2 for instance) to determine what the compressor is reacting to.
It certainly COULD cause some issues! One of the main reasons it doesn't, in this case, is because the second eq is kind of the opposite of the first, so the phase shift caused by the first eq is largely undone by the second (although not perfectly since they're not exact opposite EQ moves). The little bit of phase shift that remains wasn't enough to bug me. If there was enough to mess with the sound, like Evan said, you could use a linear phase EQ.
@@BetterMixes thanks! Phase and eqs can get kinda complicated. I musta watched that Fabfilter video about it 5 times still wrapping my head around it completely
@@JellyBags80 No problem! All the FabFilter videos are great, but yeah, sometimes they require a couple watches! In my experience, it's pretty rare that phase shift from EQing causes any real issues. It definitely can happen (especially if you do a ton of extreme parallel processing), but ultimately, I'd say only worry about it if you notice it being an issue. That's another bonus of using ProQ2 (and I'm sure there are others that do this as well), on the rare occasion that I DO run into an issue, I can just pop it into linear phase mode. But 99% of the time the zero-latency mode is totally fine for me.
Great and well-explained tutorial. You earned a Sub! A question though... When I do parallel compression like this, I feel like you have to send Pre-fader because otherwise you're going to change how the compressor is being hit using Post-fader sends with any volume automation on the main vocal fader. Also, doing this with drums is easy because I'm always only sending kick, snare and toms to the parallel comp, so the compressor will stay consistent. With vocals though, if I want to add a Vox double track, or fill, that's going to hit the compressor even harder. So in that case, I have to automate the sends from the 2 Vox tracks so I'm getting consistent gain reduction. Any thoughts on this and the proper way to route/automate?
Either way can totally work! My vocal automation typically involves a whole bunch of pretty small moves, so it doesn't throw off the amount of gain reduction too much, and I kind of like the sound of it getting a little more compressed as I push the volume up. But again, that's for small changes. If I had to make some pretty massive volume moves, then it could be an issue, but in that case, I'd probably split that part of the vocal onto it's own track to process a bit differently, and then I could easily adjust the level being sent to the parallel track. I usually only do this kind of thing to my lead vocals, but if you want your doubles to have the same processing, I'd say try it both ways! It might sound cool with the compression smashing when the double comes in, but if not, I'd do exactly what you said and just automate the send down when the double comes in.
You could use a multiband compressor as another way to address the same issue in a mix, but it would ultimately sound different. With a multiband compressor, you'd ONLY be compressing the top end or upper midrange or whatever you set it to. With this method, the whole vocal is being compressed together, it's just that the compressor is reacting much more to the top end. This would be much more similar to using a compressor with a highly tweakable sidechain section (like FabFilter Pro C2 for instance) to determine what the compressor is reacting to.
You could absolutely do this with the stock plugins in any DAW, but not with a multiband compressor. That's doing something fairly different from this technique. Using a compressor with a tweakable sidechain (like Pro C2, or Arouser, or I'm sure there are plenty more) would be much more similar to this method.
@@JairajSinghPatil Yeah, absolutely! Using a multiband compressor, you're essentially separating the vocal into 2 or 3 or 4 different frequency ranges and compressing each of those independently from each other. Using this technique (or the tweakable sidechain alternative I mentioned), the entire frequency spectrum of the vocal is being compressed together. The compression is REACTING more to the upper midrange, in other words, the upper midrange is controlling the compression much more so than the low end, but ultimately, if the upper midrange is being compressed 5dB, so is the low end and every other part of the vocal since it's all being compressed together.
@@BetterMixes ah I see I'd be interested to see how the effect changes if you put the compressor first in the chain. Since the input signal is already quite even across the frequency range (the compressor reacts to any band that crosses the threshold, and an even signal say white noise would make no difference if we make the highs trigger the compressor), and the compressor isn't really 'boosting' the high mids/highs, perhaps it's the eq boost that's doing most of the brightening.
Not exactly. You could definitely use multiband compression as another way to address the same issue! But it'd end up a bit different (maybe better, maybe worse!). This technique is more similar to using a compressor with a tweakable sidechain (something kinda like having both the sidechain highpass and peak buttons in on a Distressor or something) and then mixing that in in parallel.
I am confused. Fabfilter's compressor plugin can do exactly the samen right? You have control over the bands (multiband compression) and the sidechain is tweakable.. or am I wrong?
Grab The World's Greatest FREE Drum Samples Here:
www.bettermixes.com/drumsamples
that song is pure genius
A little trick with ProQ3: To boost the air band, move to 30Khz and do a ~30dB high shelf boost. The slope will gradually fall off and will have a smoother colouring effect when it reaches 20Khz and below than if you just boosted from 20Khz or at 12Khz or something like that.
That makes sense, I'll have to give it a shot!
Well, it works. I hear it and I like it. Explanation is excellent and pace of this show was perfect. I am an recording engineer with a couple decades experience and now I know something new. Subscribed.
Awesome! I'm glad you liked the video!
My buddy is the drummer for this band!! Just trying to get some mixing tips, and I hear his band's singer bust out. Lol, what a trip. Planet Mercury is the name
Haha that's awesome! Chris is the man! And an absolutely killer drummer!
Using this from now on!
Heck yeah! I'm glad you like this one!
Very clear and efficient explanation! Nice work, thanks
I'm glad you like it! Thanks for watching!
The point of using the la2a is because it has a high frequency filter to where it only compresses the high frequencies. While normal compression can work sometimes if you don’t have that filter you can get really apparent and strange artifacts within the parallel. You’re better off using a compressor that has a SC filter built into it, doesn’t necessarily have to be a la2a but the 2a does work very well.
I know Andrew Scheps mentioned that where I saw him doing this trick, but then he didn't have the side chain filter on, so he was just using it as a typical compressor. And really, it makes sense that you wouldn't necessarily need the side chain filter because the whole idea of this trick is kinda sorta doing what the sidechain filter on a compressor does (just in a more flexible way). So using the side chain filter would help to exaggerate the effect, but you could also then just push your EQs farther to achieve more or less the same thing.
And I'm assuming you know this, and it's just a wording thing, but I just wanted to mention this in case anyone else is reading this who might not understand how the sidechain filter works, but the compressor is still compressing the low end, it's just not looking at the low end when determining when/how much to compress. Again, I'm not trying to suggest that you didn't know that! Just a lot of people word that in a way that can be confusing to some people.
But yes, I agree that the la2a does work well for this trick! And the sidechain feature can help to further exaggerate the effect. But I just like to try different flavors of compression.
Better Mixes I see where you’re coming from, I’m just speaking from personal experience. I’ve used 1176, 3As, Fairchild’s, boutique compressors, stock, etc. and the ones that didn’t have a SC would produce artifacts whereas the ones that had it didn’t. Again not all the time, but there were quite a few cases where this did happen and when I switched to a comp with a SC it immediately went away. This is more of just a warning to anybody looking in to this trick to watch out for that because you might not notice it at first and think it’s something else which causes you to have to trouble shoot a bunch of stuff and waste time
This technique is similar to Andrea Schep's Parallel Compress which Andrew Scheps is famous for! Nice application!
Great, this sounds like exactly what the parallel particles plugin is doing. Good to understand what is happening in the backend
Ah, interesting! I had never looked all the closely at that plugin, but it would make sense that this is what it's doing!
Thanks for sharing the trick with us. The Band/Song sounds aaaawesome. Blessings!
Thanks so much for watching! I'm glad you liked it!
Dang! This is going to completely change vocal clarity in some mixes!
It really does bring vocals forward in a great way!
thank you man, you saved my mix with this
Awesome! I'm glad I was able to help!
great ! thanks and greetings from berlin
I'm glad you liked the video! Thanks for watching!
Easily the clearest explanation I've ever heard 🙏
Thanks so much! I'm glad you like it!
Great way to be creative with signal flow in the box!
Love the tricks
I believe this technique is referred to as, 'pre-emphasis - de-emphasis EQ', ie: whatever you do 'before' the compressor, you do the opposite 'after' the compressor.
Waves Abbey Road Saturator has this feature built in and so does Kelvin from Tone Projects.
Yup! That's exactly what this is! I haven't tried those plugins, but it can be a really fun and powerful technique to use around saturation, so I might have to try those out!
Thanks man, appreciate the video :) i recently got a mixer & have had some rack units ive wanted to hook up & rock, i got a bunch of ideas but this helps me get an idea of some things i can do on vocals :) have a great week!!!!!
I'm glad you liked the video! I hope it helps you get an awesome vocal sound!
Right on time brother, just finishing the vocals on the mix, thanks wonderful explination
Oh awesome! I'm glad you liked it!
This trick is awesome, thanks for sharing!
I'm glad you like it! Thanks for watching!
@@BetterMixes my pleasure, you got a new subscriber!
very cool thanks for the tips
No problem! I'm happy to help!
Awesome and immediately applicable to the mix I'm working on! Thanks - and subscribed ;-)
Nice! I'm glad I was able to help! Thanks for watching!
Wow thanks man you just taught me a great tip
I'm glad you found it helpful!
Thank you! Great info!
No problem! I'm always happy to help!
Your voiceover is Perfect vocal chain in my opinion for hiphop in my country though. What u use? I mean hardware and software to get that sound like?
Jus wanted to say thanks!
No problem! I'm always happy to help. I'm glad you liked the video!
Such a great tutorial! thank you!
No problem! I'm glad you liked it!
Good demo and basically a Compander.
Pretty darn cool. Thanks
It reminds me the Waves Abbey Road Saturation. Very cool tip
Nice! I never got around to trying that plugin out. I had been meaning to for the longest time. But I'm glad you liked the video!
i gotta try this
Let me know how it works for ya!
Thanks for that great tip!
No problem! I'm happy to help!
2:42 just on this alone it makes me think you're just replicating the "vocal trick" Mutt Lang did with the Dolby tape hiss remover. Audiothing has a plugin, Type A, that accomolishes this well.
It definitely could be a similar idea! The dolby trick (or at least the version I know) was basically just using the encoding stage of the dolby, and my understanding is the dolby (and most tape hiss removers) were based around a similar idea as this of using emphasis and de-emphasis curves to boost high end going into tape and then removing it after, taking a lot of the tape hiss with it. So yeah, definitely a lot of similarities! I'll have to try the Audiothing plugin at some point to see how I like it!
loved this!
Awesome! I'm glad you liked it!
Thanks for sharing 👍
Really helpful lesson. Thanks man.
I'm glad you liked it! Thanks for watching!
Interesting technique, thanks for the video
No problem! I'm glad you liked it!
soo why don’t we just do this with a single multiband compressor in parallel? It makes me think that Shepp made this technique before multiband compressors were a thing in a DAW, because that’s essentially what we’re doing here.
While you could use a multiband compressor as another way to address the same issue, it's ultimately a different thing and would sound different. It could be better or could be worse depending on the situation, but still, it would be different. This is really closer to using a single band compressor with a highly tweakable sidechain (something like FabFilter ProC2 or I think Arouser has a pretty flexible sidechain as well).
Indeed, as Schepps himself explains, e.g., in his template video on PureMix, that's precisely what he's doing.
@@BetterMixes nah fam.. the multiband does all these three things and it's less degradation.
@@eightrice Again, they're completely different things. Saying that one is straight up better than the other is like saying a saw is a better tool than a screwdriver. It depends what you're trying to do. I like using multiband compressors on vocals quite a bit. In fact, I'm pretty sure I was using one on this vocal, but it's a different thing entirely.
@John Pagale seems like it is compressing the range of frequencies that is left post eq, so the same thing. To me the salient difference is that it is easier to ride/automate when you have them as separate channels
Great trick
Glad you found it helpful! Thanks for watching!
Very cool tip. Thanks!
Thanks! I'm glad you liked it!
Great TRICK this is a cram session...no time for full screen tonight...EQ & Compression...no name.
Great stuff here. First video of yours that I've watched. Immediate subscribe.
Thanks so much! I'm really glad you liked the video!
Wavesfactory Spectre you can do this without 5s, just saturate the high mids
I'm a huge fan of Spectre! Such a cool plugin! The results would be a little bit different than this method, but yeah, it would ultimately accomplish the same goal in a simpler way!
damn when the song came in that was sick
Wow
That's so interesting and super. I'm getting close for what I'm looking for. But still couldn't figure it out what I really want when it's vocal mixing. Crispy vocals, lime saturated but not like old Cassete. Sound warm and crispy. And I'm getting closer by this trick you show. Thanks a lot
No problem! I'm glad I could help get you closer to what you're looking for. Just keep messing around with it, and you'll find the sound you're going for!
I wonder if you could accomplish the same effect just using the dynamic EQ on the Pro-Q3 for the upper midrange. Or a multi band comp like you said.
Not exactly. You absolutely could use Pro-Q3 in place of a multiband compressor in most cases, but in this case, we're not actually doing any multiband compression. Everything is being compressed together by a "normal" single-band compressor; it's just that the compressor is primarily being triggered by the upper midrange. If you have Pro-C2, you could use the sidechain section to accomplish this in a more straightforward way though!
@@BetterMixes I agree with you on this as your compressing the eq curve once you flavor to taste. For sure a tool of the trade! Well done.
how is the processing on the vox bright different from a multiband compression? I get it that the EQ control could be a lot more detailed but is there anything other than that? Thanks.
my question as well. he did mention that it's kinda like doing a parallel multi band compression but didn't talk about how they would differ at all. I would guess it would get close enough
@@pco2004 I figured that when the compressor is triggered by the bright part it also compresses the low end while multiband compression is only processing the high.... but still don't know how much of difference would this make
Like you said, when using a multiband compressor, you'd only be touching the highs (or whatever you have it set to compress) as opposed to with this method, it's still compressing the whole vocal together. This really helps to give it that super tight, in your face kind of sound. If you were looking for a simpler way of doing this, you could try using a compressor with a super flexible sidechain section (like FabFilter Pro C2 for instance). That would be much closer to this sound without the need for routing the vocal to an extra track.
Multiband compression can still be really cool on a vocal though! In fact, I probably use it on vocals more often than I do this technique, but I use it for a different purpose. It can really help to keep a vocal more tonally consistent from one word or syllable to the next, so there are no words with too much low end or high end compared to the rest.
Yeah, the vocal is going to pop through when you essentially bring it up by 6dB and add some hi-end.
This is pretty different than just adding a bunch of treble. Like yeah, you could do that too, and it might do what you need it to do, but it won't sound the same as this method (for better or worse).
It's adding the vocal in parallel with some hi end. It's going to stand out because it's louder overall. As presented in the video. There are a thousand ways of doing this. I just hate how people are going to be duped into thinking this is some "secret" and it's not. It's basic audio. I get that it gets clicks but it's not anything special. @@BetterMixes
@@hinder10709 Yes, it's louder; that happens any time you do something in parallel (at least if it has to be done on a second fader. Obviously that's not the case if you're just using a plugin with a mix control). That's just kinda a side effect. You can do level-matched comparisons, which I've done, and I'd encourage everyone else to do when first trying out a new technique like this, but saying that it's just making it louder and brighter is entirely ignoring the compression aspect of this, not to mention the actual "trick" going on here which is using emphasis and deemphasis curves around the compressor (this can also be really cool when used around a saturation plugin!).
As far as tricking people, I'm not saying anywhere in the video that this is some magic thing that automatically makes your tracks win Grammys, I just saw Andrew Scheps talking about it, thought it sounded interesting, tried it, thought it was cool (even after level matching), and figured I'd share it with some people.
But again, if you think this is just EQ and volume, go play around with emphasis/deemphasis curves. Dan Worall has a great video on this subject where he explains it better than I ever could, so go check that out if you get the chance, but it can be a great technique to get some new and unique sounds
It's presented as a "secret" so, yeah, that's somewhat deceptive.
Making something louder isn't "kinda a side effect" it's part of what is making this setup seem 'better' and makes it 'cut' through a mix more. It's not just the extra hi end that gets compressed and mixed. It's all the frequencies that get louder through constructive interference. More so in the hi end yeah, but overall.
And yes, this is just eq and volume. That's all we as audio engineers really have to play with. Levels (overall and specific bands) and time.
I'm aware of Worall's videos.
@@BetterMixes
@@hinder10709 "And yes, this is just eq and volume" Again...ignoring the compression entirely. And more level IS a side effect of doing parallel processing. Before plugins compressors had mix knobs, if you wanted to do parallel compression, you'd have it on a second fader. By doing so, it would get louder. Does that mean that every engineer who ever used that technique was an idiot who was being tricked by level differences? Probably not. Same thing here. Try setting this up for yourself, and then make a third channel with a level boost to match the volume of the combined 2 tracks in this technique (oh, and make your deemphasis curve perfectly negate your emphasis curve so you're not getting the extra top end). Flip between the two and hear for yourself, they won't sound the same even when level-matched (and without the top end boost). I did this for myself when first trying out this method, and it's not the same. Heck, go a step further, just do a level matched AB of just a compressor and then one with emphasis/deemphasis EQ. Once again, they won't be the same. You may or may not like the technique, that's entirely up to opinion, I personally don't use it too often anymore, but every once in a while I will, but saying that this is just EQ and a volume boost is incorrect.
Amazing. Thank you!!!
No problem! I'm glad you liked it!
Sounds even better and easy to fit into a mix if you have duplicate vocal tracks and have one with some bottom and top taken away and the other with all the bottom removed and eq’d at the top end for the sharpness and clarity. Then balance the two tracks to taste.
This is not a good idea because you are modifying your original vocal track, creating phase issues with the duplicated processed copy and you are not adding anything to your signal to make it louder. So not a good alternative to this technique :(
Yooo dude thanks.
thats great, impressive explanation.. can you also tell me which software do you use to make such vdo's recording to editing and publishing, thanks
To record these videos, I use a program called ishowu. It's a little buggy, and I think there are better versions out there, but it gets the job done. Then I edit with Final Cut Pro.
@@BetterMixes thanks brother
@@gaganbakana No problem!
I use to use camtasia but I might switch to OBS . I building a channel just been working on all of my graphics in Adobe after effects
alternatively use TDR Nova ;)
it's a free vst plug in equalizer (just like fabfilter) which contains a build in compresser.
you can create frequency assigned compressors with it. great piece of software :)
Yes! TDR Nova is nearly in every Master and or Mix in our Studio. Its a magic thing
Or go for Melda Production Bundle.
Bunch of mixer and master bus vsts for free
I love Nova! But I assume you're referring to the dynamic EQ built into it? Cause if so, that's something a bit different (although it could be used to address the same problem in a different way). This is really more similar to a compressor with a highly tweakable sidechain. Unless that's what you meant and Nova can do that, in which case that's awesome, and I did not know it could do that!
Subscribed. Good work my man :)
Great video. Just excellent information and not constant shots of the speaker's face taking up the screen and obnoxious jump-cutting.
Thanks! I'm glad you liked it!
I hear you!
Love this!
Thanks! I'm glad you liked it!
Reaper's stock compressor, "Reacomp," has low- and high-pass filters. Why not just use a filter? Put the high pass at like 1100 and the low pass at like 10k? Not to mention, "ReaXcomp" has multiband . . .
That could definitely get you close to the same effect! Obviously, it's more tweakable using a full-on EQ (or something like Pro C2 which has a full EQ in the sidechain section), but the sidechain filters might be all you need for the sound you're going for! I wanted to show how you can do it with any EQ and compressor for people who don't have Reaper, or just if you want to use a different flavor of compression, so you're not just stuck with the sound of Reacomp. This way, you could use an 1176 or an la3a or whatever!
@@BetterMixes Hey, while I have your attention, do you think you could maybe critique my mixing? I've had a studio for a little over a year now and am still learning, and all the time I ask people to come listen to my mixes but so far nobody ever will. I'm trying to figure out why, after eight years of uploading and a year of actual studio quality production value, I still have only 263 subscribers . . . because I doubt it's my playing. So, I figure the better I get at mixing and mastering, the more generally appealing my channel will be. Anyway, thanks for the reply.
Hmmmm sorry but I probably missed something there. It seems to me that the double eq + comp you are doing is simply a multiband compressor or a compressor with eq inside,without the need for a second eq, which is better. You send your signal to a mb compressor with just highs band then you use the out level of this to mix it with the original sound. Right ?
Not exactly! The key difference between this and multiband compression is that with this technique, the whole vocal is being compressed, it's just primarily the highs that are triggering the compression. With a multiband compressor, you'd ONLY be compressing the highs (or compressing them independently from the rest of the signal). You could get a very similar result by using a compressor that has a highly tweakable sidechain filter (like Fabfilter Pro C2 for instance).
@@BetterMixes Thank you very much for your answer ! I like to check the tricks with my ears but also to understand it from an engineering point of view. Here I understand that you try to enrich the highs of the spectral response of the vocals, by adding the original track and an alterred or modified version of the highs. This could have been high pass + saturation, high pass + slight reverb, high pass + tape, even maybe with a simple high pass analog console emulation would maybe do the job like a pulltec or SSL, or as you did a high pass + compressor (I still think you could use a mb compressor, tuning several bands, I simply said 1 high end band to make my message short for YT, though in each band this would have been quite flat which is less precise and maybe less natural that what you did. You could also use a compressor with a high pass in it such as the Kotelnikov, I do not think you need to bump the highs in an eq, you just need to high pass and increase the compressor input, or a dynamic eq). What I miss I think is why you add a second eq to try to so to speak "restore" the low ends you removed. You will not restore it as it was, because nearly removed this part with eq and added a compression on it, which means you lost most of the information in the low end, so when you try to restore it you basically just add noisy signal, you cannot restore signal you have lost. And it seems you do not need to restore that part, you just need to blend your highs track with your original sound, you do not want to add back a modified low end to your original track because in this case you will also enhance the low end which does not seem to be your purpose. Or maybe you do not do a parallel track, which would explain why you try to restore the low end ? But I do not thing this can be what you do because too much information is lost after high pass+comp so, you certainly do that process in a parrallel track and blending after. Sorry to bother you I really want to get the insights, this is how I remember things the besst ! :)
@@johanjof5613 I add the low end back in because I still want the compressed parallel chain to be full frequency. It's kind of like cranking the high pass side chain on a compressor. The compressor is then only reaticing to the highs, but the actual signal still retains the lows. You certainly could just leave the low end out in this trick, but I think it would start getting harsh pretty quickly. And one thing to keep in mind is that the low end isn't gone when you EQ some out. In this particular example, it looks like I used a shelf that maxes out at about -12dB somewhere around 100Hz, so you could easily restore that low end with a low shelf boost at the same frequency and slope. You can try that for yourself, try cutting 20 dB with a clean digital EQ followed by another instance of the EQ doing an equal but opposite boost. It should sound exactly the same as it did without either EQ. And that's the power of this trick. Now try sticking a plugin between those two EQs. Could be a compressor like in this example, or try different saturation plugins. You might find some really cool sonuds by affecting what that middle plugin is reacting to withoutreally messign with the overall EQ balance of the source. But going back to restoring the lost low end, even if you were to use a high pass filter instead of a shelf, there's still plenty of low end that can be brought back. If you were to use a 12dB/octave slope and set it somewhere around 500Hz, let's say, you'd only be cutting 12dB at 250Hz and 24 at 125, which is certainly a lot, but not so much that that information is gone forever. If you were using something like a 96dB/octave slope, well that would be a different story!
Your klanghelm has that sidechain highpass btw
So is it likely that I would want to eq out bad frequencies and print that on to the track/file before going into this?
You don't necessarily need to print it (you can if you want), but doing some EQing, compressing, de-essing, etc. on the main channel first is probably a good idea!
that's the motown vocal trick
thanks and goodness
Thank you for this! Do you know what mic and pre were used to record the vocal? I have a similar voice and I like the sound of it.
I do! This was recorded with a Shure sm7b through an Audient ASP800. I hope that helps. Thanks for watching!
Better Mixes man, I’ve had my eyes on the asp800 for a while now as an expansion. Thank you!
This reminds me of greenday, the song and sound.
Yeah, this band almost definitely grew up listening to Green Day! They're called Planet Mercury if you want to check them out.
great technique!! how is this different than using a multi band compressor like the C6?
Ultimately, this method is only ever compressing the vocal all together as a single band. The compressor is just more so listening to that upper midrange, but it's still compressing the full frequency range together. With a multiband compressor, you'd be compressing the upper midrange completely independently from the rest of the vocal. So this method is really more like using a compressor with a tweakable side chain EQ (like Fabfilter Pro C 2 or Arouser or something like that). Both methods could get you cool results! They're just different.
How is this approach different from a multi-frequency compressor?
It's really much more similar to using a compressor with a flexible sidechain section (like Pro C2 or Arouser or something), as the whole signal is being compressed together (as a single band). The upper mid-range isn't actually being compressed any more than the rest like it would be with a multiband compressor, it's just that the upper mid-range is primarily triggering the full band compression.
Always curious if this causes any type of flanging. I know with analog gear it would due to the time delays caused by the wire and console, but with digital not sure
It is dependent on the plugin algorithms you use I think some plugins that attempt to emulate the subtleties of analog gear will cause some phase in-coherency. One option is to render the “scheps” chain on a track and then bounce it in place, so you can actually see the waveform of the main vocal and the parallel vocal next to each other and accommodate for phase differences
In general, as long as you have delay compensation turned on in your DAW, it shouldn't cause any issues. There are some plugins that don't play well with parallel processing (multiband compressors typically don't as well as some random other plugins), so just try it, and if you happen to hear it getting phasey and weird, just try swapping out the plugin you're using.
@@BetterMixes You should compare it to using linear phase EQ. Theres the Mike Shipley trick of HPF at 8-9k, compressing the crap out of it and adding it back it - and the difference between linear and non-linear EQ is immense - (to the point where non-lin is unusable). This trick is fantastic on dull acc gtrs too.
@@minimoog4236 Yeah, when using high pass of low pass filters, linear phase mode will almost always sound better. With bells and shelves, it doesn't get nearly as weird in non-linear mode, but you're right, it would still be interesting to compare the two! I'll try that on a mix I'm working on tomorrow!
Can someone extrapolate on the idea this would cause possible phase issues with analogue gear.
Since you are just eqing and then your compressing why would this cause any time issues? You would be eqing first and then compressing two audio tracks, why would they cause an issue? Maybe I missed something in the vid.
I would assume there would be no delay worse than a double vocal take. But yeah I’m confused to the issue of delay with analogue gear , seems overstated to me? In general am I going to run into delay issues with going More hardware based? I honestly wouldn’t think this would be a real issue but this is something I’d like to clarify
Ahh, cool technique! I wonder if this was the inspiration for Denise audio’s most recent plugin Dragon Fire, as the “push/pull” application is essentially a side chain eq pre and post compression. Very interesting
Oh interesting! I haven't heard of that plugin; I'll have to check it out!
@@BetterMixes it’s a very unorthodox compressor and breaks away from traditional analog-style emulations or even more modern UI/IX compressors on the market, but I am personally loving it! Highly recommend
Should I assume you've tried sidechain instead?
I assume you mean using a compressor that can take a sidechain input and creating a mult of the vocal, EQing that, and sending that to the sidechain? Cause that would absolutely work as well! I just like this version because you can use any compressor at all, even if it doesn't have a sidechain input. Plus, for the sidechain version, I'd need 3 tracks, the main vocal, the EQ'ed mult, and the bright vox track as opposed to this that only needs two. But either way would totally work!
@@BetterMixes No, that's not what I mean. This technique dates back to early pop radio and every engineer I've ever shared it with has gone on to use on their mixes all the time. It's quite simple actually. In the final stage of your band track mix, insert a stereo compressor - which has a sidechain input - just before the band track hits the master faders. But send the vocal(s) DIRECTLY to the master faders. ALSO... use the vocal track as the input signal to the side chain of the band's stereo compressor. (using an aux send tapped off of the vocal) i.e. the band track should NOT drive the side chain - ONLY the vocal signal. Now, the vocal track will drive the compression of the stereo band mix - but not of the vocal itself since it's going directly to the 2 mix. With about a 3 to 1 ratio on the compressor, and relatively fast attack and release, the band track will dip invisibly underneath the vocal but still give a full sound. Set correctly, your ear won't be able to tell that the band track has been effected at all. Let me know if that's clear.
@@ronnyjay100 Oh yeah, that's a great trick! I actually thought I already made a video about that, but I just looked and it looks like in that video I was just sidechaining the vocals to the guitar bus, not the full mix. But I especially like that trick for pop songs where you want the vocal really out front.
4:04 ok this technique makes a lot more sense with a Pultec since it's harmonic. I'm not sure if Pro Q3 does the same thing or not.
It's still accomplishing pretty much the same thing. The main idea is the emphasis/de-emphasis. It's about causing the compressor to react differently, but then restoring more or less the original frequency balance after (or in this case, I left it a little extra bright, but it'd be the same idea with an exact opposite curve). It'll sound a bit different with a pultec or a neve or an api, but it's still the same concept. Not to mention, in most cases, pultecs are quite a bit cleaner than a lot of people think. There's certainly some harmonic activity, but it's not super crazy and not enough to fundamentally change what's happening in this technique.
@BetterMixes word 👍 yeah I hope you didn't take that as a criticism, it wasn't meant that way. This was a great video 🤘
And would love to hear what you think about Type A! Changed my world haha.
@@guesswho22peekaboo no, didn't take it badly at all! I'm glad you liked the video!
Great!
Hey Man, did you set up that send from the already mixed Vocal-Track or is there no previous procession going on? Cheers
Yeah, the send is on the already processed vocal track.
Thanks for watching!
... Or a good multiband compressor do the job, certainly
, this for that, it is a nice and éducative by the way
You could definitely use a multiband compressor as another way of addressing the same issue, but it's ultimately a completely different technique and will therefore sound different (maybe better, maybe worse, depends on the situation!).
Can we hear it in context with the rest of the song?
Right around the 9 minute mark you can hear it with everything. Or, if you look up the song Honest by Planet Mercury, you can hear how it was used in the final mix!
Have you put any thought into how this translates into simply using a multi-band compressor in your vocal?
This is really closer to using a compressor that has a highly tweakable sidechain filter (something like FabFilter ProC2 for instance) than it is to multiband compression. With a multiband compressor, it would ONLY be compressing the upper midrange area while leaving the rest of the vocal alone. With this method, it's still compressing the entire frequency spectrum, it's just REACTING more to the upper midrange. That being said, multiband compressors can sound great on vocals too! I like to use them to sort of even out the tone of the voice from one word or syllable to the next. But like I said, just a different thing.
welp,I know what I'm doing today
Not sure that counts as super compressed exactly, it was more kind of turned down in volume. You needed to get that needle moving to reduce dynamic range.
Yes and no. As I'm sure you're aware, the meter doesn't show even close to the whole story when it comes to compression. For one, once you're getting to the extreme end of the meter, it can be really hard to see what's going on. Then on top of that, VU style meters, like the one in MJUC, are really pretty slow, so the actual compression is moving around much faster than what's reflected in the meter. You can hear pretty clearly that the vocal is way more compressed with the plugin on (and the autogain in MJUC really helps to hear it since it's doing a pretty good job of making up for the volume change in real time).
Can't you do this in pro C with a sidechain filtered?
You could definitely get something really similar with the sidechain section in Pro C! Plus you don't need to set up an extra aux doing it that way.
So wouldn't a compressor with a sidechain filter and EQ after that give you the same result?
If it's a highly tweakable sidechain (like in Fabfilter Pro C2 or Arouser or something like that), then yeah, it would be very similar. If the compressor just has a basic sidechain filter, you could do something in the general vein of this, but not quite. Plus, this way, anyone can do it, even with stock plugins if they don't have a compressor with a tweakable sidechain!
thanks G
I'm happy to help!
Woah
This is a great technique and you have shown it's usefulness. I would caution trying to describe it in the way that you have though. You keep using the sidechain compression vocabulary where you describe it as what the compressor reacts to and what it does not. But using this technique literally removes frequencies from the processing itself. So that they are not getting compressed at all. In the case of sidechain compression the entire signal is being affected by what you are telling the compressor to do. This technique is much more surgical in that it does specify what will be compressed and what will not. The drawback is that you literally have to take a scalpel to it twice in order to accomplish this. That's kind of a win lose based on the situation
You're right about this being much more surgical, and also that it's kinda a bit more of a pain since you have to EQ twice, but that's not true that the frequencies we're ducking down aren't getting compressed. The entire frequency spectrum of the vocal is being compressed together as one. So when we sort of "undo" the first EQ after compression, that low end is still compressed, the compressor just wasn't reacting as much to it. Now if we didn't use the second EQ and left the low end out, when we blend that back with the main vocal, then you could say that we're pretty much only compressing the top end in parallel (which could totally be a useful technique as well), but that's not the case here.
@@BetterMixes actually, the entire purpose as I understand it, is to take those frequencies you don't want above the threshold down to an out of reach level, so that when the Compressor engages, it will only be attenuating the part of the spectrum that is still loud enough to be affected by the volume ratio cut. Something that cannot be restored after the fact. But the volume change of the EQ can be
@@rhemaman That's not how a "normal" (single band) compressor works. It sounds like you're describing a multiband compressor or maybe something kinda like the Dynamic Spectrum Mapper plugin (or another spectral compression kinda thing). A typical compressor has no idea if it's seeing low end or high end or midrange or anything else. It just sees one single level, and if it's above the threshold, it compresses. So even if a signal is like 95% high frequency stuff with just a hint of low end, the compressor just sees the combined level of the entire signal and compresses all of it together as a single thing.
@@BetterMixes whatever is above the threshold is the only thing that is loud enough to get crunched. The other stuff is generally untouched, with exception of when they're happening simultaneously. And even in that case, the headroom ratio will also translate to much less being compressed of this frequencies, because they didn't hold a large footprint in the mix when everything got crunched. This is why the softer parts of a track always get louder after compression. Because they actually benefit of more headroom by the taking away of the other stuff. This relationship of above and below the threshold is what's being exploited by the technique. And to take advantage of it you have to be extreme with all the moves; as you have shown.
@@rhemaman Yes, whatever is above the threshold is what get's compressed, so, in theory, if there was a part of the vocal that was ONLY low end, it likely wouldn't get compressed with this setup, but that's never the case with a vocal (unless it's got some crazy filter effects or something obviously). The low end and high end and every other part of the vocal are all happening simultaneously, so when you say "with the exception of when they're happening simultaneously," that's literally all the time. Some words will have a little more or less low end than others, but that doesn't change the fact that the compressor is only looking at the combined level of the signal.
Think about a compressor as you with your finger on a fader (this is far from a perfect analogy, but close enough). Someone tells you that any time the level on the meter goes above -10, you pull the fader down, turning the volume down. You don't know whether it's mostly high end or low end or what, you just see the level on the meter (the combination of all the frequencies present in the signal). So when you pull that fader down, it's turning down the lows, mids, and highs all equally. Now let's say you cut 10dB of low end out of the signal. That will change when the signal goes above that -10 mark, but when it does go above, you're still turning down the lows, mids, and highs, all equally, not just the highs. That's what's happening inside the compressor. It's reacting more to the highs, but when it compresses, it's still compressing everything together.
Are you Kenny Beats?
I am almost definitely not Kenny Beats.
Could you no just save plugin space and use a multiband?
Nope, not exactly. While you could use a multiband compressor as another way to address the same issue, it's ultimately a different thing and would sound different. This method is really closer to using a single band compressor with a highly tweakable sidechain (something like FabFilter ProC2) as it's compressing the entire signal together as a single band. It's just reacting primarily to the upper midrange. Using a multiband compressor, it would ONLY compress the upper midrange independently of the rest of the signal.
You didn’t play the song long enough or A - B it enough or turn the vocals up loud enough for the listener ( us ) to really get a full grasp of the effect . It’s hard to hear the full effect during a 6 second playback . I’ll stay tuned , thanks 🙏
I'll keep that in mind and try to play the examples longer for future videos. Thanks for watching!
How can you boost low end that you cut out? Isn’t it gone?
Nope! It's not gone, it's just turned down. You can try it for yourself, do a massive cut somewhere, like 20 or 30 dB or something, and then put a second EQ plugin right after boosting the same amount at the same frequency. It should sound identical to how it started (if you're using a an EQ with symmetrical boost/cut shapes). Even if you used a high pass filter on something, the low end PROBABLY isn't gone (unless you used a very steep slope or set it to a super super high frequency).
Couldn't you just take a multiband compressor for that kind of stuff? Like a c4 or something?
Yes and no. You could absolutely use a multiband compressor as another way of tackling the same issue, but it would ultimately lead to a different sounding result. This is more similar to having a super tweakable sidechain on a compressor (although not exactly that either). You can use this emphasis / de-emphasis technique with a lot of things too, not just compression. Try putting a saturator in between two opposite EQs for instance! You can get some cool sounds.
@@BetterMixes Thank you so much for answering, I will try it out this week for shure!
How do you automate that in the chorus?
There are a few ways you can do it. A lot of the time, I like to but my verse vocals and my chorus vocals on different tracks so I can have different settings and sounds for the different parts of the song, so that way, I can only send the chorus vocal track to the send. Alternatively, if you're in Pro Tools, you can control-command-click on the send's mute button, and then in the edit window, you can write in the automation on the vocal track to turn the send on and off throughout the song.
what song in video?
It's called Honest by Planet Mercury!
why not use multiband compression?
You could use a multiband compressor as another way to address the same issue in a mix, but it would ultimately sound different. With a multiband compressor, you'd ONLY be compressing the top end or upper midrange or whatever you set it to. With this method, the whole vocal is being compressed together, it's just that the compressor is reacting much more to the top end. This would be much more similar to using a compressor with a highly tweakable sidechain section (like FabFilter Pro C2 for instance) to determine what the compressor is reacting to.
How do all those eq moves in parallel not cause phase shift or preringing? (Not being snarky, really asking)
Pro q 3 specifically is linear phase, so phase shift would be minimal with that eq.
It certainly COULD cause some issues! One of the main reasons it doesn't, in this case, is because the second eq is kind of the opposite of the first, so the phase shift caused by the first eq is largely undone by the second (although not perfectly since they're not exact opposite EQ moves). The little bit of phase shift that remains wasn't enough to bug me. If there was enough to mess with the sound, like Evan said, you could use a linear phase EQ.
@@BetterMixes thanks! Phase and eqs can get kinda complicated. I musta watched that Fabfilter video about it 5 times still wrapping my head around it completely
@@JellyBags80 No problem! All the FabFilter videos are great, but yeah, sometimes they require a couple watches!
In my experience, it's pretty rare that phase shift from EQing causes any real issues. It definitely can happen (especially if you do a ton of extreme parallel processing), but ultimately, I'd say only worry about it if you notice it being an issue. That's another bonus of using ProQ2 (and I'm sure there are others that do this as well), on the rare occasion that I DO run into an issue, I can just pop it into linear phase mode. But 99% of the time the zero-latency mode is totally fine for me.
you're supposed to mirror the EQ bands exactly to avoid the phase issues (as well as compensating a few ms w/ the channel you're sending from)
Great and well-explained tutorial. You earned a Sub! A question though... When I do parallel compression like this, I feel like you have to send Pre-fader because otherwise you're going to change how the compressor is being hit using Post-fader sends with any volume automation on the main vocal fader. Also, doing this with drums is easy because I'm always only sending kick, snare and toms to the parallel comp, so the compressor will stay consistent. With vocals though, if I want to add a Vox double track, or fill, that's going to hit the compressor even harder. So in that case, I have to automate the sends from the 2 Vox tracks so I'm getting consistent gain reduction. Any thoughts on this and the proper way to route/automate?
You can see at the end how his main vox fader is automated going up and down, I think thats why he does it post fader.
Either way can totally work! My vocal automation typically involves a whole bunch of pretty small moves, so it doesn't throw off the amount of gain reduction too much, and I kind of like the sound of it getting a little more compressed as I push the volume up. But again, that's for small changes. If I had to make some pretty massive volume moves, then it could be an issue, but in that case, I'd probably split that part of the vocal onto it's own track to process a bit differently, and then I could easily adjust the level being sent to the parallel track. I usually only do this kind of thing to my lead vocals, but if you want your doubles to have the same processing, I'd say try it both ways! It might sound cool with the compression smashing when the double comes in, but if not, I'd do exactly what you said and just automate the send down when the double comes in.
Did Andrew Scheps go to Berkeley in the early 90s?
Hmm, I'm not sure!
Dont think so. He went to University in Miami and is a bit older than to be going to college in the 90’s
Why not use a multi band compressor instead?
You could use a multiband compressor as another way to address the same issue in a mix, but it would ultimately sound different. With a multiband compressor, you'd ONLY be compressing the top end or upper midrange or whatever you set it to. With this method, the whole vocal is being compressed together, it's just that the compressor is reacting much more to the top end. This would be much more similar to using a compressor with a highly tweakable sidechain section (like FabFilter Pro C2 for instance) to determine what the compressor is reacting to.
@@BetterMixes thanks for the thorough explanation 👍🍻
@@timrex4175 No problem!
Lol, just download Reaper and use the stock multi band compressor/eq. Great tool on an affordable DAW
You could absolutely do this with the stock plugins in any DAW, but not with a multiband compressor. That's doing something fairly different from this technique. Using a compressor with a tweakable sidechain (like Pro C2, or Arouser, or I'm sure there are plenty more) would be much more similar to this method.
@@BetterMixes could you explain how that's different?
@@JairajSinghPatil Yeah, absolutely! Using a multiband compressor, you're essentially separating the vocal into 2 or 3 or 4 different frequency ranges and compressing each of those independently from each other. Using this technique (or the tweakable sidechain alternative I mentioned), the entire frequency spectrum of the vocal is being compressed together. The compression is REACTING more to the upper midrange, in other words, the upper midrange is controlling the compression much more so than the low end, but ultimately, if the upper midrange is being compressed 5dB, so is the low end and every other part of the vocal since it's all being compressed together.
@@BetterMixes ah I see
I'd be interested to see how the effect changes if you put the compressor first in the chain.
Since the input signal is already quite even across the frequency range (the compressor reacts to any band that crosses the threshold, and an even signal say white noise would make no difference if we make the highs trigger the compressor), and the compressor isn't really 'boosting' the high mids/highs, perhaps it's the eq boost that's doing most of the brightening.
MIKE IT’S JACKSON HELLO
JACKSON!!! Hey! How's it going? It's been a while!
This would be cool to do with Smooth Operator. Would be pretty trivial
I don't know of Smooth Operator, I'll have to check it out!
Isn’t this equivalent to using a multi-band compressor?
Not exactly. You could definitely use multiband compression as another way to address the same issue! But it'd end up a bit different (maybe better, maybe worse!). This technique is more similar to using a compressor with a tweakable sidechain (something kinda like having both the sidechain highpass and peak buttons in on a Distressor or something) and then mixing that in in parallel.
I am confused. Fabfilter's compressor plugin can do exactly the samen right? You have control over the bands (multiband compression) and the sidechain is tweakable.. or am I wrong?
I would simply put a kramer master tape and it’ll do the trick