Why I Don't Use Compressors Anymore

[Anderton]

Here's an excerpt from one of my recent blog posts:

 

This wasn’t a conscious decision, or something I planned. But when I looked through my last few songs while seeking candidates for a book’s screenshots, I was shocked to realize there were almost no channel inserts with compressors. Curious, I went back even further, and found that I’ve been weaning myself off compressors for the past several years without even knowing it. WTF? What happened?

 

First of All, It’s 2023

 

Compressors are at their very best when solving problems that no longer exist. I certainly don’t need to control the levels of the PA installations at the 1936 Berlin Olympics (one of the first compressor applications). In the pre-digital era, compression kept peaks from overloading tape, and lifted the quiet sections above the background hiss and hum. But it’s 2023. Compression isn’t needed to cover up for vinyl or tape’s flaws. Besides, we have 24-bit A/D converters and increasingly quiet gear. Today’s technology can accommodate dynamic range.

 

Even though compression does bring up average levels for more punch and sometimes more excitement, to my ears (flame suit on) the tradeoff is that compression sounds ugly. It has artifacts, and adds an unnatural coloration...

 

To read the rest of the post, including what I now use instead of compression, here's the link.

 

I'd be interested in your thoughts!

 

 

[KuruPrionz]

I'll read your link but I hardly ever use compressors either. 
I think of them as an effect rather than an essential, and it's not an effect that I am particularly drawn too. 

 

You're right about modern times, we have plenty of audio headroom to record beautiful sounding music without compressing anything. 

If I have a quiet part in a vocal I may use volume on the music to bring it down for those moments. 

Just not a fan of crushing the dynamic range. I've never used one for live music, I control my volume by changing picking velocity and proximity from the microphone, super easy. 

[Sundown]

The main applications I use compressors for are the mix buss or mastering. As a keyboard player who writes instrumental music, getting a mix to glue together can sometimes be challenging and I find that a dB or two of a good buss compressor can help. My favorites are the SSL G, the Vari Mu, or for less color, the UAD Precision Buss compressor. 

 

I will use a compressor on a part that isn’t poking through. I have a lot of the classic software emulations (1176, LA2A, LA3A, Fairchild, dbx160), but often times I’m aiming for transparency and I’ll use a Waves C1 or Renaissance Compressor. The 1176 can add some nice color on a big rock piano part or a guitar. I don’t record many vocalists but when I do, typically I’ll use at least one channel compressor.

 

But I’ll say this for sure: When I watch and listen to product demos on YouTube and I hear them squash the crap out of snare or drum buss, I’m scratching my head thinking “Why would I want to do that?”  That “splat” sound is terrible to my ears. It reminds me of this old late 80’s Eddie Money track.

 

I forget who said it (perhaps Chris Lord Alge), but he said “a compressor should always sound like I’m turning it up”. I think about that a lot, and if my settings or compressor choice are diminishing the sound, I’m probably doing something wrong and I re-evaluate.

 

Todd

 

 

[KuruPrionz]

It seems Craig and I have similar but different approaches to the bass. 

I'd like to share my way of going about it. I record bass going straight in to the interface and run it as loud as I can without overloading (-6db is more or less the limit). 

Peavey Fury made in USA with EMG P pickup and D'Addario Bright Flats, the "normal" set. 

When I get the bass track I want, I copy it and paste it in a new track directly below the first bass track. 

I put a high pass filter on the second one, centered around 200hz. Just rolling off the low end so it stays clean and full down there. 

Then I drop a tube amp plugin into the second track and roll that bass knob down (I've been using a Fuchs Overdrive Supreme 50 plugin) and get the distortion going nice and dirty. 

When I'm mixing I run the clean track as the main bass track and start turning up the distorted bass, it still sounds clear and full but it cuts through in the mix nicely and you can hear every note. 

That's it, simple but effective.

[jazzpiano88]

Don’t we still need it for mastering material for the wide variety of delivery vehicles?

I mentioned it a few months ago how they are able to get Steely Dan’s guitar solos to stand out in perfect clarity in studio, shopping mall, and car environments.   
 

One thing I hate nowadays is soundtracks where you can’t hear the dialog and then get blown out of the room with the sound effects.  For me, compression is a good fix for that but not everyone.  

[KuruPrionz]

42 minutes ago, jazzpiano88 said:

Don’t we still need it for mastering material for the wide variety of delivery vehicles?

I mentioned it a few months ago how they are able to get Steely Dan’s guitar solos to stand out in perfect clarity in studio, shopping mall, and car environments.   
 

One thing I hate nowadays is soundtracks where you can’t hear the dialog and then get blown out of the room with the sound effects.  For me, compression is a good fix for that but not everyone.  

As Craig mentions above in his link, while a compressor is one way of doing things, a limiter does a similar but different thing. You can limit how loud just the loudest parts are, evening them out with the rest of the mix, which has not been compressed. Then you can increase the volume of that part because the peaks will not cause distortion. With a DAW, automating these sorts of tracks for individual tracks is possible and may only be needed in some tracks and some places. I'm not saying that's what Steely Dan did but it certainly could be and technically it isn't really the same as compression, which squashes more of the signal.

[Anderton]

1 hour ago, jazzpiano88 said:

Don’t we still need it for mastering material for the wide variety of delivery vehicles?

 

Kuru anticipated my answer, I prefer multiband limiting. However, I give it an assist by using a Gain Envelope to bring down the nastiest peaks. Then, I can raise the average level without having to apply any dynamics processing. Once I've brought down as many peaks for which I have the patience , then I add limiting but the audio won't need much to achieve the same "hot" sound as if I hadn't brought down the peaks. You never hear the dynamics "working" under those circumstances.

 

Again, it's personal preference. Some prefer the sound of bus compression or multiband compression. One of the aspects I like about a multiband limiter (I use the Waves L3-16) is that because it's multband, I can also alter the gain for bands to add broad EQ tweaks. In my quest to use a minimal number of plugins, this means that for much of my music, the only plugin I need to use for mastering is the L3-16. But, I have the advantage of mixing what I master, so I can take the mix as far as I want before going to the mastering stage. When mastering other peoples' music, I usually need to use EQ too.

 

Perhaps something else of interest is that my goal with the mix is to not need mastering. Of course, mastering always gives it that "extra" push, but before something goes to mastering I have to feel like I wouldn't be embarassed if the mix itself got out into the world. I also never mix with anything in the master bus, because I feel it can obscure issues with a mix that need to be fixed before applying dynamics processing.

 

It's very old school approach, but I still separate the mixing process (achieve the best balance) and mastering (achieve the best sound). I know it's not the way most people work these days. I don't think it's because I'm resistant to change, I think it's more about having complete control over the process. I don't want anything exerting control over the mix's dynamics, but I definitely want something exerting control over the master's dynamics.

 

TMI, I know....

[ElmerJFudd]

I always use compression on vocal (exception being classical music).  The dynamic range of the voice is really wide.  We could of course use a look ahead fader riding type solution or draw in a lot of automation, but compression tends to be both attractive to the ear on voice and brings attention to the voice and lyric in a mix (especially if the other instruments are not using compression).   

[dmitch57]

I agree that general use of compression is less and less useful. One application I'll probably always use it for is certain high-gain lead guitar tones which need a lot of sustain. Compression before the amp is a super-useful way to get those sounds. 

[Greg Mein]

I remember when I was a kid, getting into the guitar a bit. I got a few dollars in my pocket somehow and headed to the music store to buy an effect pedal. I saw the compressor and thought, 'wow, that seems cool' so I bought it. I took it home, tried it out and thought it really didn't do anything so I took it back and got something that would actually do something like a delay, chorus, wah, etc..

[BMD]

29 minutes ago, Greg Mein said:

I remember when I was a kid, getting into the guitar a bit. I got a few dollars in my pocket somehow and headed to the music store to buy an effect pedal. I saw the compressor and thought, 'wow, that seems cool' so I bought it. I took it home, tried it out and thought it really didn't do anything so I took it back and got something that would actually do something like a delay, chorus, wah, etc..

 

I had exactly the same experience many moons ago!

[Anderton]

2 hours ago, dmitch57 said:

One application I'll probably always use it for is certain high-gain lead guitar tones which need a lot of sustain. Compression before the amp is a super-useful way to get those sounds. 

 

As I mentioned in the blog post, I still use compression as a tool. The two examples I gave were guitar compression for sustain, and to restrict dynamic range going into envelope followers. There are also some cool out of phase things you can do with compressors, but that's not really using compressors for their intended function.

[Anderton]

49 minutes ago, Greg Mein said:

I remember when I was a kid, getting into the guitar a bit. I got a few dollars in my pocket somehow and headed to the music store to buy an effect pedal. I saw the compressor and thought, 'wow, that seems cool' so I bought it. I took it home, tried it out and thought it really didn't do anything so I took it back and got something that would actually do something like a delay, chorus, wah, etc..

 

My first experience with compressors was before they were available as pedals. I was curious if compression would add more sustain, so I modified a voice leveling circuit for CB radio to accommodate guitar levels and impedances. It became part of "my sound" and made feedback more controllable. I pitched to Kaman (who had ust bought Ovation), and they thought it was really stupid to want more sustain out of a guitar

[Anderton]

3 hours ago, ElmerJFudd said:

I always use compression on vocal (exception being classical music).  The dynamic range of the voice is really wide.

 

Try the gain envelope and phrase-by-phrase normalization technique I mentioned, I wouldn't process voice any other way now. However, it is quite time-consuming to massage the vocal with DSP instead of just compressing or limiting it. If I was doing a song a week, I might have a less favorable opinion about spending that much time perfecting a vocal's dynamics.

[Sundown]

10 hours ago, Anderton said:

 

Perhaps something else of interest is that my goal with the mix is to not need mastering. Of course, mastering always gives it that "extra" push, but before something goes to mastering I have to feel like I wouldn't be embarassed if the mix itself got out into the world. I also never mix with anything in the master bus, because I feel it can obscure issues with a mix that need to be fixed before applying dynamics processing.

 

It's very old school approach, but I still separate the mixing process (achieve the best balance) and mastering (achieve the best sound). I know it's not the way most people work these days. I don't think it's because I'm resistant to change, I think it's more about having complete control over the process. I don't want anything exerting control over the mix's dynamics, but I definitely want something exerting control over the master's dynamics.

 

I’ve gone back and forth on this, and I’m still not sure if my approach is best. For a long time I mixed with a clean master buss (no plug-ins). But when I would ultimately apply a G buss compressor, or a mastering tape emulation, or some other finishing process, it would alter my channel mixing decisions.

 

Right now I’m doing more of a CLA approach with mixing into a gentle buss compressor (and sometimes even my buss EQ), because they are going to be there eventually. Then I’m not as surprised when the low end sounds different and I can make adjustments on the fly. I will also generally keep NFA’s Elevate Mastering bundle on the buss (in bypass or turned off), just so I can periodically hear what the final gain boost and limiting are going to do to the sound. 

 

I share the opinion that a mix should be good enough to distribute (absent level), as I shouldn’t be fixing things at the mastering stage. But I generally have to add 6-10 dB of gain in mastering to get to a 16 LUFS level, which makes me wonder if my mixes are too low to begin with. I generally aim for peaks to be no more than -3 to -6 dBFS on the mix, which gives me a bit of headroom for mastering processing. But I still find myself adding a lot of gain in Elevate to get to the -14 to -16 LUFS range for distribution.

 

Todd

[Anderton]

There's no problem with adding gain to reach a certain LUFS level, other than theoretical differences that don't really matter in a world where 24-bit audio is the norm. My "standard" for rock/pop songs is -12.5 LUFS. This reflects how I like to increase the average level (using the techniques mentioned above) so the sound has more "pop." Then the streaming service can turn it down to reach their target. Increasing the gain shouldn't need to alter your mixing decisions.

 

Remember that we're fighting the Fletcher-Munson curve in all this as well. Mixing into processing can compensate for the high- and low-end rolloff, but at this point that's in my DNA so I know what to expect when I add a little processing to the master.

 

 

 

[Notes_Norton]

My recording needs are different due to the fact that I make backing tracks for my duo to play over.

 

I never use a compressor on the tracks. Why? Dynamic response is one of the major ways music is expressed.

 

Reduce the dynamics, and you reduce the expressiveness. For that reason, I would never use compression unless I had to, and so far with well over 600 songs in our playlist, I've never needed to use one.

 

Insights and incites by Notes ♫

[Anderton]

1 hour ago, Notes_Norton said:

My recording needs are different due to the fact that I make backing tracks for my duo to play over.

 

Well, live performance is really different from recording. One of the main reason dynamics processors became ubiquitous is because recording and playback media didn't even come close to the dynamic range of live music, so the audio had to be shoehorned in there somehow. 

[GovernorSilver]

@Anderton which of your books at Sweetwater would you recommend as a first read for someone like me who has only used the mixer in a DAW to set levels and hasn't done much else?  I never tweak EQ, or use mixer plugins other than a limiter on the master channel.  

 

I can't decide between the Big Book of Dynamics FX, Secrets of Equalization, or Max Your Mix.

[Anderton]

Cool, thanks for your interest! I'd recommend Max Your Mix because it gives an overview of all the elements involved in mixing, which includes dynamics and EQ. If you scroll down far enough on the link, you can see the topics covered in each chapter. As one 5-star review said, "I highly recommend 'Max Your Mix' if you are anywhere below 'mastering engineer.' Craig's narrative is linear and comprehensive, yet it doesn't come at you like a tidal wave. Higher-end tech tips are highlighted, but the majority are in common terms you'll generally recognize."  I think that sounds like what you want.

 

The Big Book of Dynamics FX and Secrets of Equalization have some overlap with Max Your Mix, but they're deeper, more specialized dives. The 3 books are in the top 5 best-selling books Sweetwater has, so they're all about equally popular.

 

For what it's worth, there's currently a bundle on sale for $5 off with Max Your Mix and Secrets of Equalization. There are also some bundles that include specific plugins and Max Your Mix.

 

If you think you're going to get Secrets of Equalization someday, you might as well get it now and save the $ because the books have free updates. Whenever you're ready, just download it again and you'll get the latest version. Max Your Mix is on its 3rd update, and I'm working on the 4th one scheduled for release in Q2 2024. You'll be able to get that update for free when it comes out.

[Notes_Norton]

18 hours ago, Anderton said:

 

Well, live performance is really different from recording. One of the main reason dynamics processors became ubiquitous is because recording and playback media didn't even come close to the dynamic range of live music, so the audio had to be shoehorned in there somehow. 

That explains it. Many of our single and duo competitors use karaoke tracks.

 

By making my own, I can increase the dynamic range, and exaggerate the groove, plus bass and snare volumes to sound more 'live'.

 

I believe the audience knows the difference, even if they don't know "why".

 

Notes ♫

[The Real MC]

On 12/15/2023 at 3:51 PM, Anderton said:

I'd be interested in your thoughts!

 

In my experience, a day will come you will regret that decision.  Mark my words.

[Anderton]

14 hours ago, The Real MC said:

 

In my experience, a day will come you will regret that decision.  Mark my words.

 

Why? The music sounds fine as is. I can't picture myself saying someday "gee, I should have used a compressor on that track instead, I've changed my mind and now think compression sounds so much better than using gain envelopes."

[KenElevenShadows]

I've always gone easy on compression. I don't use it to excess.

 

In the analog days and when I had a board, I recorded vocals with a finger on the fader. I raised it and lowered it slightly. To do this, you better know the song, though.

 

Now, I record digitally. I use compression to "shave off" some of the transients, but otherwise have it so it's not very heavy-handed. I use automation or clip gain for a lot of stuff instead of compression. So I still go easy on compression.

 

Is it more work? Sure. Does it sounds better? I think so.

 

P.S. I am fully aware that if I had an extremely high-end hardware compressor, I may be singing a different tune. However, I do not, and neither do most people.

[Dave Bryce]

I love the gain envelope approach.  Been a fan for a while now - I’ve always seen it as the closest you can get to having someone intelligently moving your faders in real time, which is essentially what compressors do in many volume-related instances.

 

I do still use some compressors though…but almost entirely to get the sound they bring, so more as an effect than for taming dynamics.  I also tend to use hardware limiting on the way into the box to control peaks.  For years, my baby has been my Summit TLA-100 because of the ease of use, the smoothness of its optical circuit…and the lovely tube sound. 😎

 

dB

[Anderton]

1 hour ago, KenElevenShadows said:

P.S. I am fully aware that if I had an extremely high-end hardware compressor, I may be singing a different tune. However, I do not, and neither do most people.

 

However, that extremely high-end compressor won't be able to draw out an errant mouth click, alter the amount of breath fade in, smoothly fade out the sections between parts to zero, or not compress only the consonants at the beginning of several words in a rap song. Then again, if we have compressors with machine learning that can study how I do compression and make those techniques part of how they compress vocals, then I'd probably go back to using compressors.

 

1 hour ago, KenElevenShadows said:

Is it more work? Sure. Does it sounds better? I think so.

 

Absolutely it's more work, but then you never listen back to it later and go "gee, I shoulda..."

 

And I have no doubt that it sounds better, but that's based on the sound qualities that I want to hear from vocals. I know some people prefer the sound of compressed vocals.

[KenElevenShadows]

For sure. You might choose a high-end compressor for a number of reasons, sometimes like what Brother Dave is describing. But drawing out mouth clicks or altering the amount of breath fade-in isn't among those reasons.

[RABid]

I can sit down at a real snare drum, play a few measures that contains everything from soft ghost notes and ear popping rim shots and it makes me sad to think what someone with a compressor would do to those dynamics. I'm sure that a good acoustic guitarist would feel the same way.

[Notes_Norton]

5 hours ago, KenElevenShadows said:

In the analog days and when I had a board, I recorded vocals with a finger on the fader. I raised it and lowered it slightly. To do this, you better know the song, though.

While being on the musician side of the glass, I remember engineers who kept their finger on the faders and had lightning reflexes.

 

When doing the sound check, I would get close to the mic, and play my sax as loud as I could to give the engineer a reference point. Then I would step back a few inches, and play as dynamically as seemed appropriate.

 

I don't know what the engineers did with that reference point, but I know that I never had a ruined take due to distortion. Before he had a stroke, the guy in our local studio called me a “One Take Jake”. I got a lot of work because of that. I came prepared, and fortunately, what he put in front of me was never demanding beyond my capabilities. Mostly pop or country tunes.

 

When recording my backing tracks, I use the same model mixer and speaker as I use on stage in my studio. I'll mix the 6 or 8 sound modules with my ears, then record them all at once in mono leaving a bit of headroom. Then I normalize and make a few versions, normalized, -1db and -3db. Usually the -3db mixes best on stage with the rest, so I test that one first. The one that seems 'mastered' best is the survivor. In very rare instances, I may have to do a -2 or -4.

 

I mix in mono because in a live setting, the majority of people are not sitting in the sweet spot for stereo mix. I learned that sitting in a pizza parlor. There were speakers in the ceiling, and we were under either the left or right channel, and the popular songs they were playing sounded weird because we could hardly hear the other channel.

 

I don't know if what I'm doing is the right way or not, as I am totally self-taught. But it seems to work for me.

 

Notes ♫

[David Emm]

I lean towards a touch of compression on the master bus, because it does a nice glue thing on the final sound. I also know what kind of trap it can become, so I let things loop and pull back on it until it doesn't *quite* satisfy my ear. That's when I feel I've done the smart thing: add spice, but don't overdo it.

 

Limiting and proper EQ are wiser directions to take now, but compression is still useful. Its better for electronic music than string sections, but as it fades for mixdown duty, it rises a bit as an effect. I'm amused by that, because by the time I laid hands to a couple of them and started learning their merits, the technology pushed them into the background! 😜