In
Part 1 of “Keys To A Great Mix” I talked about various EQ techniques to
use when mixing. Now I’m going to follow up with tips on dynamic
processing (compression and expansion) and time-based effects (delay,
reverb, etc.). I’m going to give very specific settings &
examples, but just remember that these are my personal preference and
are not “correct” or “standard” by any means.
Why use Compression?
A lot of times, you want to control the dynamic range of an instrument (such as a lead vocal) so that it doesn’t disappear during softer phrases and blow out your ears on louder belted phrases. Also, compressing something like a vocal helps you place it in the mix and lift it up without overpowering everything else. For drums, compression helps you achieve a consistent sound/level for each hit (especially important with kick and snare).
Clean up your drums with Expansion
With drums, expansion helps to attenuate (get rid of) the unwanted noise from the rest of the kit when a particular piece (the snare, for example) is not being struck. With expansion, you’re basically singling out your hits so they sound almost like samples. This can help in mixing to clean up some of the extra noise and allow you to adjust the level of the kick, snare, or toms, without affecting the rest of the drum mix too much. You don’t always need to use expansion…it’s more of a problem solver (very useful in live situations) than a necessary key to mixing. You generally would only want to try using expansion on the kick, snare, and your toms. If it sounds good, use it. If not, don’t!
The expander will be “closed” until a sound hits the input high enough to cross a threshold (dB level) you set. For a snare, you want to set the threshold just above the noise level, so that only the hit from the snare is triggering it to open. To use an expander on a snare that’s consistently striking around -3dB, while the other noise from the kit is coming through at -15dB, you might want to try setting the threshold to -10dB. Set the attack time as fast as possible; the attack controls how quickly the expander opens up, and you want all of the attack from the snare. Listen closely to the snare and adjust the “hold” (how long the expander stays completely open) and “release” (how long it takes to close back up) so that you hear the full snare sound, and it closes smoothly and almost unnoticeably. I tend to keep my hold and release time longer than most to allow the snare to settle before attenuating it too much.
Compression Settings for Drums
Some of these settings may seem a bit extreme, and that’s because I’m using examples that would fit a pop rock sound with a drummer that hits fairly hard… In the end, the settings do not matter at all – it’s how it sounds. If it sounds great, go with it. Learning how to achieve certain sounds will help you know when and how to break the rules…and we like breaking rules in music
Kick – For a “tight” and “punchy” kick sound, try setting the attack at 30 ms (just enough to let the slap/attack through), the release at 60 ms, the ratio at 5:1, and the knee at 20 dB. Start the threshold at 0 dB (so that the compressor is inactive) and lower it until you’re consistently getting about -3 to -6 dB of gain reduction (GR). The threshold completely depends on your input level. Adjust the makeup gain when you have a nice sound to reach a useable mixing level.
For a “thumpier” kick sound, try adjusting the attack to .5 ms (you’re not letting the attack through with a fast attack), the release to 30 ms, & the knee to 10 dB.
Snare – I like to use moderate compression on my original snare track (to hear the attack/crack), then double it (send it by itself to an aux track at full volume) and compress the heck out of it to get that sweet pop…I like to think it sounds like a tennis ball being served.
Overheads - Use the same settings as the high-hat & toms. You can also experiment and try increasing the ratio to as much as 15:1 or 20:1 and see what happens. Since your snare drum is usually the loudest piece of the kit, you might see it peaking 3 – 10 db above everything else in your overhead mics. You usually want a nice smooth cymbal sound with your overheads, and you might want to try to compress them enough so that the snare is attenuated to be almost the same level as the cymbals. This gets a nice poppy sound from the snare and also makes your overheads easier to work with; you can bring your overheads up in your mix without bringing up your snare volume as a side effect.
Bass Guitar – You want to smooth out your bass guitar as much as possible to have a solid low end. Bass can easily get out of hand if not compressed enough. For bass, compression settings can vary widely based on the player/musician. You might need to compress slap bass more than someone using a pick, more than someone thumping softly with their fingers. Try using an attack of 8 ms, a release of 600 ms, a knee of 20 ms, and a ratio of 4:1. You almost always want to see at least 3 to 6 dB of gain reduction, and up to 20 dB of gain reduction. Don’t be afraid to crush it. Again, this all depends on the style – you may not need to compress the bass much at all if you want an open, airy sound.
Electric Guitar – I don’t usually compress my electrics (until I compress the stereo guitar buss with a strong limiter) because the amp is already compressing the signal so much. This also opens up the field of sound a little so that there’s more room for the guitars to bounce around enough room for a little magic to happen... I can always count on the limiter on my stereo guitar bus to keep them from peaking.
Vocals – I tend to crush vocals (mostly in pop rock) because I like to hear the lead vocal up close and personal (similar to most vocals on Top 40 radio). A popular vocal compressor that can achieve this is the Universal Audio 1176. The Pro Tools plug-in called Bomb Factory is modeled after this compressor. Here’s what one of my Bomb Factory plug-ins looks like on a rock vocal.
You want to use a fast attack (1 ms) and a fast release (200 ms) on a vocal. You want to hear the “punch” of the vocal, but set the attack quick enough to clamp down and smooth it out. You don’t want to hear a “pumping” sound while you’re adjusting the attack and release.
The compression ratio can range from 2:1 to 20:1…until it becomes limiting. I like to use a ratio of 20:1, push all the buttons in on Bomb Factory (and the 1176) and listen to the magic happen… Pushing all the buttons in technically still has a 20:1 ratio, but lets the compressor start to act a little funny. Just shift-click two of the ratio selector buttons to try it and you’ll see.
Subtle Reverb and Delay Tricks
When you’re sending your instruments to your reverbs, try creating a dedicated Verb for your drums and instruments, and a separate one for your vocals. You can use more than one reverb and mix them together if you want.
Try soloing the lead vocal and just the reverb aux track, starting with the reverb fader all the way down. Slowly bring the fader up until you can faintly hear the verb. Try to hear it creating a nice bed for the vocal without it sticking out. Focus on the lead vocal and see if it still sounds clear and out front. If the reverb is distracting or too noticeable, turn it down a little. If you keep your effects subtle enough so they sound good when they’re soloed, they’ll sound great in the whole mix.
Try sending your lead vocal to 4 separate stereo delay aux tracks. Create a mono slap delay with the first (150 ms to 230 ms). Use the same method of soloing the vocal and the delay track and slowly brining up the fader until you barely notice it. It should be enough to add body and depth to the vocal – and even a little bounce – without sticking out enough for your audience to notice its there.
Use the second delay track and set it at the quarter note length based on the song tempo (you might have to tap it out or actually listen for it), the third delay track at the half note, and the fourth delay track at the whole note. By doing this, you’re creating your own manual “echo” or “multi-delay” effect… You may want to use one of those automatic effects, but this is my favorite way of adding delays to the vocal. Add in these long delays ever so subtly based on your taste… You may want to automate them to be lower in the verses and more prominent in the choruses.
Now when you add everything else back into your mix, your audience shouldn’t even notice that you have reverb and delays on your vocal – unless you intend for them to – and your vocal will have more body and blend with your mix much smoother.
If you want to hear the song I referred to for most of my settings, you can go to http://www.jakehartsfield.com, click on the DP logo, and listen to the first track by Kristen Jaudon.
I know that was a long read, but I hope it was helpful. It should have at least inspired you to think “oh yeah, I can mix better than this pretentious fool!” ;)
Why use Compression?
A lot of times, you want to control the dynamic range of an instrument (such as a lead vocal) so that it doesn’t disappear during softer phrases and blow out your ears on louder belted phrases. Also, compressing something like a vocal helps you place it in the mix and lift it up without overpowering everything else. For drums, compression helps you achieve a consistent sound/level for each hit (especially important with kick and snare).
Clean up your drums with Expansion
With drums, expansion helps to attenuate (get rid of) the unwanted noise from the rest of the kit when a particular piece (the snare, for example) is not being struck. With expansion, you’re basically singling out your hits so they sound almost like samples. This can help in mixing to clean up some of the extra noise and allow you to adjust the level of the kick, snare, or toms, without affecting the rest of the drum mix too much. You don’t always need to use expansion…it’s more of a problem solver (very useful in live situations) than a necessary key to mixing. You generally would only want to try using expansion on the kick, snare, and your toms. If it sounds good, use it. If not, don’t!
The expander will be “closed” until a sound hits the input high enough to cross a threshold (dB level) you set. For a snare, you want to set the threshold just above the noise level, so that only the hit from the snare is triggering it to open. To use an expander on a snare that’s consistently striking around -3dB, while the other noise from the kit is coming through at -15dB, you might want to try setting the threshold to -10dB. Set the attack time as fast as possible; the attack controls how quickly the expander opens up, and you want all of the attack from the snare. Listen closely to the snare and adjust the “hold” (how long the expander stays completely open) and “release” (how long it takes to close back up) so that you hear the full snare sound, and it closes smoothly and almost unnoticeably. I tend to keep my hold and release time longer than most to allow the snare to settle before attenuating it too much.
Compression Settings for Drums
Some of these settings may seem a bit extreme, and that’s because I’m using examples that would fit a pop rock sound with a drummer that hits fairly hard… In the end, the settings do not matter at all – it’s how it sounds. If it sounds great, go with it. Learning how to achieve certain sounds will help you know when and how to break the rules…and we like breaking rules in music
Kick – For a “tight” and “punchy” kick sound, try setting the attack at 30 ms (just enough to let the slap/attack through), the release at 60 ms, the ratio at 5:1, and the knee at 20 dB. Start the threshold at 0 dB (so that the compressor is inactive) and lower it until you’re consistently getting about -3 to -6 dB of gain reduction (GR). The threshold completely depends on your input level. Adjust the makeup gain when you have a nice sound to reach a useable mixing level.
For a “thumpier” kick sound, try adjusting the attack to .5 ms (you’re not letting the attack through with a fast attack), the release to 30 ms, & the knee to 10 dB.
Snare – I like to use moderate compression on my original snare track (to hear the attack/crack), then double it (send it by itself to an aux track at full volume) and compress the heck out of it to get that sweet pop…I like to think it sounds like a tennis ball being served.
For the original snare, try using an attack of 15 ms, a release of 45 ms, a knee of 20 ms, and a ratio of 6:1. You should adjust the threshold to see between -3 and -12 dB of gain reduction. These settings also work well for your bottom snare mic.
For the super-compressed snare, try keeping the settings the same, but bringing the threshold down -20 db from where you had it. You should see a nasty amount of gain reduction (12 to 30 dB)
Toms – Try using an attack of 30 ms, a release of 10 ms, a ratio of 4.5:1, a soft knee at 30 dB, and adjusting the threshold to see about -6 dB of gain reduction. You want to let the attack/slap of the top come through and compress the rest of the tone so that it sounds smooth and full.
High-Hat – Use similar settings to the ones I described for the toms (attack: 30 ms, release: 10 ms, ratio of 4.5:1, soft knee of 30 ms). These compression settings work well for most drums & cymbals (overheads)Overheads - Use the same settings as the high-hat & toms. You can also experiment and try increasing the ratio to as much as 15:1 or 20:1 and see what happens. Since your snare drum is usually the loudest piece of the kit, you might see it peaking 3 – 10 db above everything else in your overhead mics. You usually want a nice smooth cymbal sound with your overheads, and you might want to try to compress them enough so that the snare is attenuated to be almost the same level as the cymbals. This gets a nice poppy sound from the snare and also makes your overheads easier to work with; you can bring your overheads up in your mix without bringing up your snare volume as a side effect.
Bass Guitar – You want to smooth out your bass guitar as much as possible to have a solid low end. Bass can easily get out of hand if not compressed enough. For bass, compression settings can vary widely based on the player/musician. You might need to compress slap bass more than someone using a pick, more than someone thumping softly with their fingers. Try using an attack of 8 ms, a release of 600 ms, a knee of 20 ms, and a ratio of 4:1. You almost always want to see at least 3 to 6 dB of gain reduction, and up to 20 dB of gain reduction. Don’t be afraid to crush it. Again, this all depends on the style – you may not need to compress the bass much at all if you want an open, airy sound.
Electric Guitar – I don’t usually compress my electrics (until I compress the stereo guitar buss with a strong limiter) because the amp is already compressing the signal so much. This also opens up the field of sound a little so that there’s more room for the guitars to bounce around enough room for a little magic to happen... I can always count on the limiter on my stereo guitar bus to keep them from peaking.
Vocals – I tend to crush vocals (mostly in pop rock) because I like to hear the lead vocal up close and personal (similar to most vocals on Top 40 radio). A popular vocal compressor that can achieve this is the Universal Audio 1176. The Pro Tools plug-in called Bomb Factory is modeled after this compressor. Here’s what one of my Bomb Factory plug-ins looks like on a rock vocal.
You want to use a fast attack (1 ms) and a fast release (200 ms) on a vocal. You want to hear the “punch” of the vocal, but set the attack quick enough to clamp down and smooth it out. You don’t want to hear a “pumping” sound while you’re adjusting the attack and release.
The compression ratio can range from 2:1 to 20:1…until it becomes limiting. I like to use a ratio of 20:1, push all the buttons in on Bomb Factory (and the 1176) and listen to the magic happen… Pushing all the buttons in technically still has a 20:1 ratio, but lets the compressor start to act a little funny. Just shift-click two of the ratio selector buttons to try it and you’ll see.
Subtle Reverb and Delay Tricks
When you’re sending your instruments to your reverbs, try creating a dedicated Verb for your drums and instruments, and a separate one for your vocals. You can use more than one reverb and mix them together if you want.
Try soloing the lead vocal and just the reverb aux track, starting with the reverb fader all the way down. Slowly bring the fader up until you can faintly hear the verb. Try to hear it creating a nice bed for the vocal without it sticking out. Focus on the lead vocal and see if it still sounds clear and out front. If the reverb is distracting or too noticeable, turn it down a little. If you keep your effects subtle enough so they sound good when they’re soloed, they’ll sound great in the whole mix.
Try sending your lead vocal to 4 separate stereo delay aux tracks. Create a mono slap delay with the first (150 ms to 230 ms). Use the same method of soloing the vocal and the delay track and slowly brining up the fader until you barely notice it. It should be enough to add body and depth to the vocal – and even a little bounce – without sticking out enough for your audience to notice its there.
Use the second delay track and set it at the quarter note length based on the song tempo (you might have to tap it out or actually listen for it), the third delay track at the half note, and the fourth delay track at the whole note. By doing this, you’re creating your own manual “echo” or “multi-delay” effect… You may want to use one of those automatic effects, but this is my favorite way of adding delays to the vocal. Add in these long delays ever so subtly based on your taste… You may want to automate them to be lower in the verses and more prominent in the choruses.
Now when you add everything else back into your mix, your audience shouldn’t even notice that you have reverb and delays on your vocal – unless you intend for them to – and your vocal will have more body and blend with your mix much smoother.
If you want to hear the song I referred to for most of my settings, you can go to http://www.jakehartsfield.com, click on the DP logo, and listen to the first track by Kristen Jaudon.
I know that was a long read, but I hope it was helpful. It should have at least inspired you to think “oh yeah, I can mix better than this pretentious fool!” ;)
Best of luck with your mixing!
Questions, comments? Share them here on the TuneCore Blog!
thanks so much for the free advice :)
Posted by: Poetry-kiddo | November 13, 2009 at 02:59 PM
Jake, thanks for this amazing article, and thanks to the many other experts who have commented here. I'm not a recording engineer,I'm more of a keyboardist, but trying to find my way with the many tools available through Logic Studio.
Do you know if Universal Audio 1176 or Bomb Factory are available for Logic?
Also, is there a best order that reverb and delay be added on to a track, one before the other, or is this not an issue?
Also, does anyone recommend any other software similar to Waves plug L-3 for easy ultramaximizing, for the purpose of getting a good gain in the final mix, yet retaining a good amount of depth and dynamics (sort of a compromise to address the issue that Bimbo Patrol was mentioning)?
Posted by: www.facebook.com/profile.php?id=504861479 | November 07, 2009 at 12:48 AM
good
Posted by: tavarus thomas | November 06, 2009 at 11:38 AM
Eric - Great advice, thanks!
Paul G - Those presets you're talking about using on the master mix are basically "mastering" presets that help shape your mix into a tighter, more compressed/cohesive sound, while bringing up the level as high as possible without peaking. These presets can be dangerous to use (they are certainly not going to fix a bad mix), but rewarding if you do not intend on having someone else master your mix.
You do not want to use those presets if you are sending your mix off for mastering. You may use very soft compression and a very soft limiter in this case. The mastering engineer will take your mix and polish it up using his/her own outboard gear and mastering software to master your mix (you'll find that a professionally mastered mix from an engineer with fresh, but seasoned ears, will always sound better than one you slapped those presets on).
If you are not having another engineer master your mix, and you want to try to get your mix to a commercial level, you may want to use compression on the master mix and a strong limiter. Be wary that "Saturation" is exactly what it is...and it saturates your mix with harmonics and extra noise that you usually do not want. You shouldn't need to use this if you have a great mix. That "Saturation" may sound like it's making your mix brighter and more full, but you're sacrificing clarity and the integrity of the original mix.
Posted by: Jake Hartsfield | November 04, 2009 at 12:31 PM
One of the ways I use an expander/gate on the toms is with EQ in a side-chain. This is, basically, where you're applying EQ to the 'threshold' of the gate ONLY and not the signal going through the gate, make it more or less sensitive to certain frequencies.
On the tom mics, for example, I'll place a High Freq Roll-Off at around 300Hz (listening to it helps) to eliminate any snare drum hits activating the tom gate. When the tom is hit, there's enough energy below the roll-off frequency to open the gate, letting the tom sound through. Because of this setting, the gate threshold can be set a bit higher than you'd typically have without the side chain roll-off.
Posted by: Eric Seaberg | October 29, 2009 at 12:47 PM
Thanks for your efforts by the way.
Posted by: Paul G | October 29, 2009 at 03:50 AM
With today's convenient presets - I tend to mix down with a blanket combinator (Compression, Saturator, EQ, limiter)already installed. It can sound very powerful sometimes with my ears. However, if I were to disable the blanket treatment, the track goes into the red way too easily. Yet, like I said, with the processing there are no audible effects opf such clipping, or so it seems. Thus, would you recommend mixing down the track free of blanket processing, adding it on after, or not?
Regards
Paul
Posted by: Paul G | October 29, 2009 at 03:49 AM
Thanks very much for these articles, Jake. They're really helpful!
DanW
Posted by: Dan | October 28, 2009 at 02:27 AM
Been recording at home using a VST "Addictive Drums" along with Sonar PE 8.5.. Cut out all of AD's own standalone effects and routed all the drums to there own tracks in Sonar and followed your guidelines for the compression and it has helped me get a much better and more realistic drum sound. In fact I'll be saving it as my template. Thank You
Posted by: www.facebook.com/profile.php?id=1244850191 | October 26, 2009 at 02:58 PM
20:1 on a vocal ??
and that's before mastering ?
lol.
you're kidding right ?
If not , you need to hire a real vocalist (or one who works the mic at the very least).
Sounds like you record karaoke in the local pub friday nights.
No offence meant
....but......
20:1 on a vocal ??!
lol
just had to comment on that one
Posted by: saywhat | October 26, 2009 at 07:29 AM
I thought it was a great article as you said. It's a starter not the bible. Thanks.
Posted by: Dancingmoondog | October 25, 2009 at 11:57 AM
James - Using compression (and limiting) on the master track is one of the stages that happens in mastering. If you are mastering your mix yourself, that is a completely necessary step - but you must be careful not to overcompress your mix, or it can sound squashed and lose the artistic depth you've spent so much time creating. A strong limiter can help you reach a commercially acceptable/viable level.
Mastering is an art in itself and I believe the most important thing is simply having someone with a fresh set of ears to listen and polish up your mix. A mastering engineer might also make slight EQ adjustments, bringing out a vocal or certain instruments, or simply smoothing out the frequency spectrum to sound as great as it possibly can, and also use special compression/limiting algorithms to achieve the optimum level for the master without crushing the dynamics... There's the bare bones of what a mastering engineer does. There are tons of other great articles out there specifically about mastering.
Posted by: Jake Hartsfield | October 24, 2009 at 01:41 AM
Very informative overview...I think it is understood that these are "personal preferences" and serve well as a primer. I am interested in knowing; What is your opinion on effects during the final mixdown? I like to use compression to get the levels up, but often wonder if I am doing something counterproductive...
Posted by: JAMES | October 23, 2009 at 02:34 PM
dave - Yes, I did forget to mention that the expanders I'm talking about can also be called gates. Most times, I won't use a "noise gate" that completely closes up below the threshold so there's still room for the track to breathe.
Also, I tried to make it clear that these settings are simply a starting point for people to use and experiment with. I don't expect any of these exact settings to work for anyone. I wanted to give people a very specific point at which to start from if they were unsure where to even begin. For different drum sounds, every compressor is going to need to be adjusted differently, sometimes even adjusted between takes during the same song (automated when using a DAW like Pro Tools).
Dave S-B - Thanks for filling the gaps I left out! Great comments.
Bimbo Patrol - Yes you are very right, a lot of things are overcompressed...You will probably listen to most of my recordings and think they're fairly overcompressed...
Compression is very often overused/misused, so i would caution most people to use it sparingly.
It's the "loudness war" that's gotten increasingly worse over the years - originally starting with radio compression. When radio signals were weaker, a song couldn't have parts that were too soft, because the signal from the station could get lost, or even interupted by a nearby radio station on a similar frequency with a stronger signal. When this would happen, the listener would usually switch the station, because the music was too soft and was boring, or because there was too much static. So radio stations started compressing their audio more to get the strongest, loudest signal - they still do it today in ways that would scare most audiophiles...
This "loudness war" carried over to CDs and mastering...Every artist & label wants their CD to sound bigger and badder than everyone else'...And they're willing to pay more money for something that has been crushed to death to make it sound bigger in people's stereos and on the radio.
It's rare to hear from an A&R person that they want less compression, so kudos to Bimbo Patrol :)
Charles - You should add compression later - but I'm sure you can record and use a compression plugin after you track your vocal...When you record digitally, you can listen to the compressor plugin while you record if you want, but you should be able to adjusting the settings after you record. I'm unfamiliar with Adobe Audition, but you should have the original sound file and the ability to even take the compression off after you record if its just a preset plugin.
If you're actually going through the compressor before the audio is being recorded to your computer...I would suggest not recording with compression like Dave S-B mentioned. Take his advice - start with a high quality vocal mic. Then find a high-quality pre-amp. THEN you should look for a nice compressor that you can use with mixing. If you ever compress anything while tracking, use light compression so you can tweak later. Some professionals know exactly what sound they want and exactly how to get it, and they'll track with a fairly agressive compressor, but that's risky because you can't tweak it as much when you mix.
Joshua - De-essers are certainly a necessary tool I forgot to mention for vocals to attenuate sibilance that sounds to harsh. Otherwise, I don't find myself using much side-chain compression, but there are certainly hundreds of scenarios where you might need it or use it as an effect.
On a similar note, it can help a kick drum cut through to use an expander/compressor on a bass guitar track that attenuates the bass every time the kick drum is struck - so the compressor is triggered by the kick drum track.
Posted by: Jake Hartsfield | October 23, 2009 at 12:28 PM
Thanks Jake! This is awesome stuff. Much like I get in those producer/ music magazines, but for free! I'm sure you'll probably cover this, maybe you already have before I subscribed - what about side chain compression?
Posted by: Joshua Pomeroy | October 23, 2009 at 11:16 AM