7 Obscure Mixing Techniques Used by the Pros

➥ Mixing with EQ from Matthew Weiss is available now

Most of the time there’s an obvious choice. Need more midrange? Grab an EQ and boost the midrange. Need more control of the source? Volume automation or compression. Easy. But sometimes we face strange challenges — like how to get more bass in the kick without running out of headroom. Or how do we make something sound brighter that doesn’t have much harmonic content above 7 kHz except hiss. Well, where there’s a will there’s a way. Sometimes the way is just a little less predictable.

So with that said, here are seven counterintuitive mixing techniques you can use to solve unconventional problems.

1. Using a low-pass filter for brightness

What? How can using a low-pass filter make something brighter? Well, let’s say you have a distorted guitar. It’s power goes up to about 5-6 kHz, but after that it’s just noise. A treble boost will bring out that noise, clog up your mix, and make the guitar harsh.

Instead, use a low-pass filter with a very steep slope. This does two things. First, it cuts out noise and distortion. Second, it actually accentuates the tone at the corner frequency — so while you might be attenuating everything above say 6 kHz (for example), you’re actually boosting the 6 kHz region. This happens because the EQ generates resonance right at the corner of the pass band — and it’s actually pretty clean and clear!

2. Adding midrange for bigger bass

When we want to hear more bass in a bass guitar, kick drum, or other low-end element, the obvious solution is to boost the low end. However, sometimes what we really want to do is just draw more attention to that bass element.

We can do this by adding midrange: pulling up the thud of a kick or the gnarly overtones of a bass. This pulls our ear to that element, telling us there’s more of it there — even if it’s actually just more midrange.

This can be extremely valuable when you don’t have much headroom, or there’s something else competing for attention in the low end.

3. Using compression to increase dynamics

But wait, doesn’t a compressor restrict dynamic range? Not necessarily. It attenuates a signal that exceeds an amplitude threshold. In most cases that will restrict the dynamic range. However, if the attack is long enough, and the threshold is low enough, a compressor can actually exaggerate the attack.

This happens because the compressor allows the front of the signal to pass mostly unaltered, while still pulling down the sustain of the signal and making the attack more prominent relative to the sustain. This can be very useful when trying to bring an already over-compressed signal to life (over compressed … compress it some more!) — or for injecting some serious snap into a dull drum sound.

4. Sharpening transients before a limiter on the master buss

If you’re using a brickwall limiter on your master buss, chances are you’re doing so to make something loud. And to do that, you want the maximum amount of headroom available. So why on earth would you use a transient designer in front of a limiter? Wouldn’t exaggerating the attacks use up your headroom faster?

Well, yes and no. Technically yes, but remember that these things aren’t perfectly mathematical.

Sharpening the transients can do two things. First, you can legitimately get more transient through the limiter and still retain loudness because a transient designer is boosting in a different way than the limiter is cutting. Second, the limiter is pulling down everything in the mix. That means while your kick hits harder for that 10 ms, your bass gets attenuated for that 10ms as well. The attacks will poke out clearer in the mix, thus exaggerating the dynamic perception.

Warning: sometimes this sounds like crap, so use it when it works and don’t use it when it doesn’t.

5. Using distortion to make something sound cleaner

Now that really doesn’t make sense. In what way could distortion possibly make something sound “cleaner?”

If we define clean by clarity of tone rather than by purity of the original sound, we can use harmonic distortion to make something sound more “polished.”

Light amounts of harmonic distortion will exaggerate the overtones of a source. Our brain uses these harmonics to tell us what exactly we’re hearing. It’s kind of like saying we’re going to make this clarinet more “clarinet-y” by emphasizing its partials.

6. Using reverb for intimacy

Remember that reverb is used to create a sense of space. Without reverb, it’s hard to define the front-to-back relationship of elements in a mix.

Contrasting wet elements with room sounds to the elements that are almost entirely dry can actually create a more “in your face” effect than simply leaving a sound 100% dry.

The key to doing this is to keep your forward elements sent to a reverb that is a) primarily early reflections, and b) has a long pre-delay.

The other benefit to using this kind of “ambiance” reverb is that it reinforces the tone of the dry signal a bit, which often makes it pop forward as well.

7. Mixing quietly towards loudness

Not that I feel loudness is absolutely paramount to a successful mix, but in today’s climate of iPods, noise-ridden listening environments, and DJ controlled playlists, it’s important that the record lives within the same general vicinity of apparent loudness.

Or to say it another way: the record shouldn’t sound out of place amongst the other records being played shoulder to shoulder with it in the same genre.

Getting a mix to sound loud without losing tone, dimension, or punch can be tricky — especially when the references of today’s mixes are as loud as they are.

So I’ll say two things. First, trends are showing that the loudness wars are easing off in pretty much every genre except EDM — so aim to make your mix maybe a little quieter than your references. You’ll have a much easier time getting the mix to hang together.

Second, mix your record at low monitoring levels. The reason this works is because it forces you to create energy and excitement when loudness is not an option. This will force you to be more selective about EQ and compression settings, as well as general levels and imaging.

When all said is done, you’ll find that a record that creates the impression of a big sound at low levels will sound absolutely huge when it’s cranked.

Now it’s your turn! Drop some of your own counterintuitive techniques in the comments.

Matthew Weiss

Matthew Weiss

Matthew Weiss is a Grammy nominated and Spellemann Award winning audio engineer from Philadelphia. Matthew has mixed songs for Snoop, Sonny Digital, Gorilla Zoe, Uri Caine, Dizzee Rascal, Arrested Development, 9th Wonder, !llmind & more. Get in touch: Weiss-Sound.com.
  • Jordan Kell

    Great post. Most of these I have been using already consciously or not. However I have not tried the transient shaping tip. I will have to give it a shot. While my tip isn’t quite so counter intuitive I do get asked about it often. I create varing early reflections on my vocal reverb. I duplicate the track in Pro Tools and delete everything but the last syllable or two then slide them anywhere from an eighth note to a half note and pump that through my reverb plugin. Great end of phrase vocal technique.

  • http://weiss-sound.com matt weiss

    Nice tip! Sort of a flip on the concept of using a delay as a reverb. You’re using your reverb as a delay.

  • Tact Booige

    Very interesting techniques. They make sense.

  • John McNeill

    oh, that’s beautiful article! Nice and awesome!

  • Pingback: 7 Obscure Mixing Techniques Used by the Pros | Sound ProspectsSound Prospects()

  • http://www.resoundsound.com/ Ilpo Karkkainen

    Love these tips. Great observations and they are all very useful and practical techniques. My personal fav is #7. Great!!

  • Sander Smeekes

    Lovely tips, thanks!

    Sometimes when the baseline is a leading part in a track: I duplicate the baseline and remove the low-mid frequenties on the duplicated channel, then add some distortion make the baseline sound more dirty/raw/distorted.

    • Mister Guy

      Most good bass distortion pedals have a wet/dry mix so you can get this effect live as well. I wouldn’t buy a pedal without it.

  • Shorty_dammit
  • Brecht De Man

    “[…] while you might be attenuating everything above say 6 kHz (for example), you’re actually boosting the 6 kHz region. This happens because the EQ generates resonance right at the corner of the pass band […]”
    Definitely not (always) true. Some filters (e.g. second order with little damping) have a peak around the cutoff frequencies, but others do not. Keep in mind that cutoff frequency is not always defined in the same way, and especially with filters the signal will be attenuated significantly around 6 kHz already (and virtually unharmed around 4kHz and below).
    Some good points otherwise.

    • Matthew Weiss

      Steep filters. Not all filters. And I think there is one design (Buttersworth?) that can get steep without the bump, but I can’t think of any others. Maybe some weird FFT designs…

    • Brecht De Man

      Not convinced about that, I can think of more than a couple traditional filter designs (analogue and digital) that do not come with a bump. Steepness will depend more on factors like filter order (usually translated in e.g. 12dB/24dB/… per octave). Whether or not there is a peak will indeed depend on design but also parameters (most often translated into frequency and Q, or resonance/emphasis/contour dial on analogue synthesisers.
      Butterworth filters are actually very common in audio effect designs (similar for their analogue ‘equivalent’ as far as I know), and can show a bump depending on the parameters, contrary to what you say. In general a filter may well have some attenuation at the cutoff frequency (which can be defined in different ways, so beware of filter designs).
      All engineering considerations aside: use your ears! (And not eyes, general practice, habits… but you know that.)

      At any rate, I didn’t mean to discredit your point which was a very valid one, but felt a word of caution was in place for those not familiar with filter design. Happy to discuss further or agree to disagree…

    • Matthew Weiss

      Not taking it that way. My experience has been that on low pass filters, the more the poles or the steeper the curve the greater the resonance at/right below the cut off freq. But I can assume you are right. Filter design is something I know just enough about to be dangerous.

    • Brecht De Man

      Haha :) good.
      You’re undoubtedly right in a lot of cases, it just can’t be generalised.

  • Lucas

    Wow~ thanks for the tips~

  • The Catalyzt

    I underestimated the monitoring at low loudness thing so much for so many years, that when I tried it for the first time I was floored by how much EASIER it was to hear what was going on and mix properly, even in my jacked up room (I know, I know, I need room treatment like I need ears). I’m still learning about bass at low volumes though – it tricks me because if it sounds right loud it sounds kinda quiet soft. Sometimes I equate quietness in the bass with boring-ness, and its a little bit of work finding a happy medium between the loud and soft – but I tell you what, my mixes certainly are better.

  • Djsound Phaze

    What do you do if you have fl studio. 11 what type of eq or filter would you use? Since the options are insanely all over

    • Meru

      Any eq you want. I prefer parametric eq 2 because it has a nice grapical interface but that doesnt really matter. You could do this with any eq/compressor/distortion/etc if you know how to use them

    • Zach Drella

      I’d prefer to use Parametric EQ 2 if you have it, Fruity Limiter for the limiting, compression, and sidechaining on most of your mixer tracks, and Maximus for only your Master Mixer track and other mixer tracks only if you’re looking to get certain frequencies and elements out of your sound, or vise versa if you prefer it that way. For the reverb I’d prefer using Fruity Reverb 2 because it has a nice interface and easy to get around. And there are several different ways for using distortion to make your sounds clearer. The first and easiest way is to use Fruity Wave Shaper and tinker with the distortion shape. The second way is to use Maximus (or Fruity Limiter if you can) to distort the sound, but I don’t know how to do this, but I’m sure you can find tutorials on it. The third and most complex way to do this is to record the sound you want to distort and record it into Edison and import that recording into Harmor’s Image Resynthesis tool and distort it through the distortion section of the FX window. The advantage of this method is that you can also add chorus, reverb, delay, and Harmor’s built in compression section to add more clarity, emphasis, and depth to your sound. You can also find all sorts of different techniques to do this, but this is all that I know and I hope I was able to help you out.

    • Zach Drella

      I would also explain how to do all these if needed, but I’m not considering how long my reply email was.

    • Aldo Ferrusi-Xsry

      I use Maximus on all tracks to hone in the tone of each one individually. I use stereo enhance on mid to high patches alternate phase and stereo spread on each so they don’t conflict and sounds rich and full. You’ll need to judge this by ear. Kicks level maxed to -10 and bass maxed- 12 (minimal bass patch) or -14 (thick rich bass patch). All other levels to ear but no higher than -10. On the master I use the following in order from top to bottom:

      1. Parametric EQ 2: Cutoff at 30M @ 100% and cutoff 19K @ 100%. Basically should look like a half pipe ramp. lol. Removes subsonic rumble and high hiss.

      2. Fruity Compressor: Optomaster preset. Unchanged.

      3. Fruity Multiband Compressor: Mastering 2.4db preset. Raise each band gain to 5.5db (Approximately). All bands thresholds to 11.0 (should point right) and ratio to 6.0:1 (should point right).

      4. Parametric EQ 2: Use this EQ between 200M and 2K to boost the mid-range

      to your liking. I do this one by ear because different patches react differently but to achieve a solid sound song by song for an album, I find this part is most crucial.

      5. Fruity limiter: Default setting.

      Master level should not exceed -3db. I seem to get good dynamic response and heavy big sound from this.
      Example: https://xsry.bandcamp.com/album/dream-vehicles

      Any questions on FL Mastering techniques, I’d be glad to send you a project file preloaded with settings.

      Aldo of XSRY.

  • C Tan

    So would a a shallow curve (tending towards 0dB) or a steep curve
    (tending towards 100dB and higher) sound more reverberant?

  • Gary Count Kellam

    For the last step of a balanced mix, ill turn my monitors all the way off, and run it through a pair of headphones and just set them on the mixer. And then turn it down even more to where its barely audible. From the snare to the kick, vocal, guitar, keys, it will obvious If anything in your mix is too loud or too quiet.

    • Spencer Johansen

      I’ve used this trick multiple times, it works great as long as you know what freq ranges are already going to be hot because of the shape of the cone to begin with.

  • Binaire

    One simple technique is to slowly increase your volume from zero to check what comes first in your balance.

  • Anthony

    Running live sound, I find that if I have a stereo instrument, that doing a single channel delay helps the instrument pop and distinguishes it. (ex. I have an electric guitar that I delay they left signal by 16 ms. It brings it out more on the left side and distinguishes it from the keys player who is right side dominant, playing those higher ivories.) This also makes your instrument (in my case, the electric) more fat and full feeling.

  • https://www.mixcloud.com/reticuli/ Reticuli

    Ignore 7.

    • nikoli


    • https://www.mixcloud.com/reticuli/ Reticuli

      Mix for art and emotional intent, not loudness. The only exception I would suggest has to do with mono compatibility, and then I’d recommend accepting a lesser, inferior mono playback — just the basic main portions of the music translating in mono form when the stereo channels are combined. Don’t try to get every subtle thing to show through in mono. Allow stereo playback to reveal more of it to the listener. That goes for every frequency band.

      Crushed dynamic range that isn’t simply incidental to the art and emotional intent is detrimental to art and emotional intent and eventually just makes everything sound like shit as people up the ante. Don’t do it and don’t recommend anyone else do it. That is the major reason why people are flocking back to vinyl, because you can’t crush the dynamics and increase loudness too much without the stylus going crazy. Vinyl mastering engineers who know what they’re doing MUST use the more dynamic versions of the music and master with lower maximum peaks as a matter of the format.

      It is true that live acoustic recordings, such as of symphonies, sometimes require a little compression for the vinyl master in order to bring up the volume of the quietest stuff because of higher noise floors of the format that could burry some instruments at some points after just a few plays causing just a little groove wear, but that is so much more restrained and subtle than what is happening on modern, produced studio recordings and “remasters”.

      (By the way, tape degrades so much that sometimes, even if they DON’T ruin new transfers with compression, the tape has degraded enough after a few decades even a newer AD converter will sound inferior with it than the older converters did when the tape was new and just encoded to modest Red Book. Your ears can only really hear 20bit/48khz, anyway. Higher bit depth and sampling rates are only useful for more transparent post/production processing and to reduce the idiosyncrasies of DACs usually having to do with high frequency interpolation and low-pass filters in the output stage. HOWEVER, almost ALL portable digital music players, phones, tablets, laptops, desktops, DVD players, and BluRay Players can play back 48khz. There is no reason to not produce and distribute in in 16/48!)

      If at normal listening volumes there are things that you want people to hear down low that they can’t when listening in a fairly quiet environment, then either boost individual tracks in volume or use compression moderately on individual tracks to bring whatever you’re trying to hear out a little, but do it just enough to satisfy the sound and emotion you want, not for the sake of being heard by people too stupid to turn their volume up or use passively isolating ear/head-phones when in noisier situations. All that does is dumb everything down to the lowest common denominator.

      ANY suggestion to use compression or even just less artistically-appropriate separation in the mix anywhere in order to deal with auto or subway noise or to “stick out” on the radio or keep up with everything else out there is anathema to good recording, mixing , and mastering. It is bad advice that should be rejected outright and universally. It is both inherently unsound and a slippery slope no matter how you phrase it or provide caution. Don’t do it.

      Also, the traditional standard of normalizing linearly max peaks to 0dBFS on the final mix-down is now NOT recommended by those in charge of formalizing lossy codecs, like ACC and (the terrible) MP3 formats. To prevent inter-sample conversion digital clipping, tracks should be normalized to about -3dBFS to be safe. Sorry, but your attempts to be just as loud as the competition to fit in someone’s playlist just got thrown out!

      (A few peaks slightly over -3ddBFS is unlikely to be too bothersome after later conversion if the music was already mixed to around that or is a live recording pretty close to this standard that never underwent processing. Avoid normalization unless it’s necessary. Doubling or halving in the standard 6.02dB “perfect” increments is preferable.)

      In an ideal world, everyone would just switch to FLAC. It’s lossless, compresses based on encoding time used (longer time spent encoding results in smaller file), and it has a tagging system from its inception. In the case of lossless, there is no possibility of inter-sample conversion clipping, but in the flesh world the unfortunate reality is people ARE still using lossy codecs even with how cheap storage has become.

    • nikoli

      Thanks for the response. I’ll just say that I think you are misrepresenting the author’s point. And in your effort you are addressing things that aren’t within the scope of his point… like FLAC & specific dBFS levels.

      It’s almost like you read the first and second paragraphs of #7 and then just dismissed his entire point as hogwash.

      You say, “Crushed dynamic range that isn’t simply incidental to the art and emotional intent is detrimental to art and emotional intent and eventually just makes everything sound like shit as people up the ante.”

      However, the author says, “Getting a mix to sound loud ***without losing tone, dimension, or punch*** can be tricky” … which is direct in contrast to your interpretation of his point. He goes on to explain a method of getting a loud mix without “crushing the dynamic range” (as you say).

      Why would anyone have a problem with a mix that maintains tone, dimension and punch… and is loud? What is wrong with that? What is wrong with aiming to have your mix a little quieter than your reference? What is wrong with mixing at low monitoring levels?

      I’ll be honest, your post reads more like someone who is more into religious dogma than spiritual living. I’m sure that’s not your intention. Unfortunately that’s how it reads.

    • https://www.mixcloud.com/reticuli/ Reticuli

      Follow me to the promised land!

      Every tangent I took in the above was at least justified by some other tangent that was related to the prior point I was making connected to the article. I think…

      You cannot promote loudness in comparison to any other music and add sufficient provisos that will prevent an escalation of the compression epidemic. If dynamic range of a particular track in the multi-track is too low or it’s not as loud as is desired for intended emotional effect in the mix or can’t actually be heard at normal volumes in a fairly quiet room at all and is not something allowed by the artist to be a kind of hidden gem occasionally revealed, then you are promoting “loudness” in the pejorative sense and compressed total dynamic range (RMS) by definition. Volume is eventually a function of analog power amplifiers, transducers, and a control (pot knob or digital domain).

      If you’re mixing an album, then desired loudness of tracks in relation to each other can be controlled by maximum peak level of each track. It is not only unreasonable to create some kind of “fantasy album” of comparable tracks to match to (even if you are just slightly lower RMS), but is nowadays empirically detrimental to even worry about with the -3dBFS recommended approximate max peak.

    • nikoli

      “You cannot promote loudness in comparison to any other music and add sufficient provisos that will prevent an escalation of the compression epidemic.”

      I completely disagree with every sentiment in that statement. And it is a prime example of a straw man fallacy as it relates to the author’s point in #7.

      Even if you loath the notion of trying to prevent a record from sounding “out of place amongst the other records being played shoulder to shoulder with it in the same genre” … you still can’t deny that his two suggestions are good suggestions.

      1. aiming to have your mix a little quieter than your reference
      2. monitoring at low volume levels

      There’s nothing wrong with either of those suggestions. When you say “ignore #7″ you’re just throwing the baby out with the bathwater as a result of being overtly dogmatic in your approach to compression.

    • https://www.mixcloud.com/reticuli/ Reticuli

      1. Don’t use outside music references for RMS or max peaks. I already provided technical rationale for where to allow the latter based on the processing requirements of lossy codecs, and the former doesn’t need any reference. Let your music’s RMS speak for itself… derive from itself. Don’t let anyone else’s content dictate it.

      2. Monitor at normal, healthy listening levels. If that happens to be “low” in one’s opinion, then fine, but monitoring at very low levels (below 70dB; btw, the stapedius reflex begins in the mid 80s) starts producing results similar to the loudness wars’ and is just another way to get the same low RMS results, but through a euphemistic prescription.

    • nikoli

      Sooooo ignore the good advice of an engineer who understands and accepts the reality that engineers are going to use references… because using references leads to something you think is bad.

      This is like the religious fundamentalist who ignores safe sex education because they refuse to accept the reality that teenagers are going to have sex… because sex leads to something they think is bad.

    • https://www.mixcloud.com/reticuli/ Reticuli

      You’re comparing sex to compressed, low-RMS “loudness wars” shit sound. Good luck with that.

    • nikoli

      Another straw man.

    • Spencer Johansen

      This guy doesn’t even engineer music, I’d ignore him.

    • Spencer Johansen

      In other posts you complain about too many ‘advice gurus.’ I just listened to your newest mix on mixcloud and the leveling is pretty bad, I think you should take your own advice and let the pros give out the lessons.

    • https://www.mixcloud.com/reticuli/ Reticuli

      Leveling on a live DJ mix isn’t up to your standard? What does that have to do with compression? Clearly you’re still confused about the subject.

  • Vincent Perez

    I like to put Waves maxxbass on my hi hats and use a low pass filter, it makes them sound “meatier.” Then I put a guitar distortion pedal plugin on my 808 kicks, it makes them more sparkly and cutting edge. I mix everything at unity gain and render in Ape audio format.

  • Tim Britton

    nice list! A variation on #2: use a first order high pass filter on electric bass, set between 40 and 120 Hz depending, especially when tracked direct, to get rid of bottom end mud while accentuating the upper frequencies which define the pitch better, tax your amplifier/speakers less, make more room for your kick, and leave more headroom in the overall mix.