Logics stock compressor (no.3)
- Thread starter PeterN
- Start date
charlieclouser
Senior Member
I agree, mostly. I am a sucker for extreme or weird compressors (both hard and soft) in my search for the "one-knob-Portishead-drums" sound, and I have dozens. But my template has Logic's stock comp on all 976 instrument tracks, and I only go digging into the plugin folders in special circumstances. For most situations I only want transparent level control with auto-makeup-gain, so I leave the stock comp on the default ("Platinum Digital") mode, with 2:1 ratio, -30 threshold, minimum attack, auto release, hard knee, and auto-gain at -12. With these settings it's like "auto-mix". Everything is audible, quiet sounds get brought up to be audible, and loud stuff gets clamped down invisibly. No fun or mojo or anything, just "get it sounding healthy please".
For more creative uses I do go to the other modes before I go digging into the folders for more choices, but when I do go digging I like Waves Sheps Omni Channel, UAD or Softube opto and Fairchild emulations, Empirical Arousor and UAD Distressor. And of course for mastering and stem limiters I use Waves L3-LL MultiMaximizer and Ozone among others.
But dang that ancient Logic stock comp still holds up.
For more creative uses I do go to the other modes before I go digging into the folders for more choices, but when I do go digging I like Waves Sheps Omni Channel, UAD or Softube opto and Fairchild emulations, Empirical Arousor and UAD Distressor. And of course for mastering and stem limiters I use Waves L3-LL MultiMaximizer and Ozone among others.
But dang that ancient Logic stock comp still holds up.
Alex Fraser
Requires ☕️
<takes notes>
More gold.
Be still my beating heart.Claim
No plugin compressor can beat Logics stock compressor no.3 (Studio FET)
Hooray
Really quick compression, that one..
Nice move, but why on every instrument, not stem? Do you process the sound post-compression (pre fx)?
charlieclouser
Senior Member
I put compressors on EV-ER-Y-THING. On all of the individual tracks, plus "invisible" limiters like L3-LL or Ozone on stems. (But since I'm delivering stems I don't have anything on the final composite mix; that way the stems will all add up to exactly what the composite mix is.) But when I'm preparing mixes for stereo release on albums I use Ozone and automate the threshold of the Maximizer to bring up the quiet bits even more. This is basically like automating the level of the track, but in reverse I guess? So if you're hearing my mixes outside of the context of the film mix, that's what you're hearing. A LOT of compression, at no less than three points in the chain - channel, stem, stereo mastering.
Channel comp pushes around the relationship between quiet and loud samples within the sample maps and also affects tails and stuff. Stem comps let elements within a family fight it out and push each other out of the way within that family, adding glue and blurring the lines between stacks of layered spicattos or drums or whatever. Mastering comps further force all elements to fight it out and push each other aside on their way to the stereo out. All of this adds up to "mix mojo" I guess - stuff is moving all over the place and getting pushed aside - but with very precise digital control instead of just slamming the SSL mix bus and hoping for the best.
I usually prefer when I can't hear all the separate layers individually, when it's difficult to isolate everything by ear. I want things to merge and blur together... I guess that's "mix glue"? An analog console helps with this, but you gotta hit it just right or else it's crap. Much easier to do this on rock records where the elements are predictable. Anyway it's not really an option for me to use an analog console anymore, what with stems and recalls and all that.
On my individual tracks, eq comes first, then compressors. That way any boosts or cuts I make for tonal or resonance reasons won't result in such dramatic level changes - the compressor will help to even out the level changes when I eq for one note having a howling resonance or whatever. Then that nice clean signal hits the reverbs and delays. Of course there are special circumstances where I build weird channel strips with multiple eq and comps before and after delays or whatever, but that's usually for things like a tape-delay feedback that I want to make louder without bothering with automation. I'm not really a fan of parallel compression, to me it usually sounds like I'm just using less compression or increasing attack above minimum. But I do use it here and there, when I really like the tonality of an ultra-crushed sound but I realize it's just too squashed. Dialing back in a little (25-50%) of the dry feed can put back some of the attack and still retain that crushed tonality. But for me this is very rare.
I use so many quiet sounds that my mixes would sound like fleas dancing on tissue paper if I didn't. I love the timbre of ppp layers that are boosted until they're massively and unnaturally louder. Plus, on the projects I do there are rarely scenes where nobody's screaming and shooting guns and crashing cars, so the music's gotta have some strength, even when it's "quiet". I'm not doing quiet evolving films, but I do like quiet evolving sounds. So ya gotta do what ya gotta do.
Plus I come from a background of doing loud-ass industrial metal, and my skills and techniques were always about getting maximum force, and "normal" compressors like 1176 or LA-2a never had a fast enough attack; I could always hear them biting down too late. So when the first look-ahead limiters came out I was like, "FINALLY!". I got L-1 when it was a plugin for Sound Designer II and I'd take a Rob Zombie lead vocal and absolutely crush that sucker until it was a completely square brick of audio, no dynamics at all. That was the only way it would be audible over the tornado of sludge I was putting into the instrumental tracks. And the instrumental tracks were getting slammed as well. Hiphop drum loops turned into square bricks and mixed behind the live drums gets that turbo-sludge sound. Makes mixing easier too, no automation needed! I'm always more interested in the sludge and artifacts behind the sounds that gets brought up into audible ranges by compressors. I guess that's why I use so few synthesizers as opposed to samples of organic sounds - synths don't have as much sludge hiding in the background waiting to be discovered!
If something sounds too crushed or flattened I'll back off the threshold or lengthen the attack to allow for some "pip" or "bite" on the attack, but that's usually only on staccato type sounds, so it's almost like using the compressor as a transient designer (and sometimes I do use transient designers, but only rarely). Most of the time it's zero-attack look-ahead type limiting so the initial transients are clamped HARD and the tails get brought up to audible levels. To my ears, and for the type of sounds I like, that type of limiting doesn't change the timbre or impact of the attacks in a bad way, but it does improve the juice factor of the tails. For me, strong compression on short sounds is more about sculpting the tails than messing with the attack. That's why my default setting is zero attack. I usually use auto-release and only fiddle with it on low-end-heavy sounds when the release is so fast that I'm getting that "chattering" type of distortion as the comp releases and re-attacks with each cycle of a low fundamental, or when I want to sculpt how quickly the release is bringing up the tails on drums or other short+hard sounds.
On non-short sounds, mostly what I'm looking for is auto-makeup-gain, which I guess is upward compression. I want the quiet bits to be louder, while the loud bits stay the same. Just even things out. And I'm almost never dealing with a single track that includes a wide variety of tonalities, like a key-switched multi-articulation strings part or whatever. I put each articulation on it's own track with its own eq+compression, so I can tailor the processing to just the spiccato basses or the sul-pont-tremolo violins or whatever. Realism be damned! And my default is only 2:1 ratio, which is not all that strong in the scheme of things. I can preserve the tonal changes of the various dynamic layers in a mod-wheel dynamic-layer-crossfade but reduce the actual level changes. And I often edit the levels of the layers within the sampler so the quiet layers are louder than normal anyway, which is sort of like compression I guess.
If I'm dealing with a mod-wheel-dynamic-layer-crossfade sample map, and one note is too loud or too soft I adjust the note's velocity. Since in those cases velocity is only controlling loudness, while the dynamic-layer switching is on mod wheel, this is how I mix without using level automation very much. In fact, I only resort to level automation when I need to change the level of a long sample as it's sustaining. This happens fairly rarely. I use Logic Sampler / EXS24 for almost everything, and my default settings are that note velocity affects level by 24db, which is usually enough but I can adjust if needed. For sample sets like spiccato strings or drums, where note velocity DOES switch between the dynamic layers, there are often sample sets where the various layers have NOT been normalized, so the quiet samples really are quiet. In those cases I often have to reduce the velocity > loudness amount to zero in order for the quiet samples to be audible at all (this replicates the dynamics relationship as originally recorded), and if that's not enough I go under the hood and raise the levels of the quieter layers by 6 or 12 db and then adjust the velocity > loudness amount until it "plays" right. No biggie, it's super-quick in EXS. From there, the compressors do the rest of the work to keep the quiet layers audible.
In terms of stem compression, for me it's only limiting, and almost never "mojo" compression. Anywhere from 6-12 db of reduction is common on some stems during the loud bits. Some stems might not be hitting the threshold at all while others will be getting slammed. But I always have those limiters (and the per-channel compressors) in place while I'm doing sound design, composing, and mixing, so it's not like I put that stuff on only at the end and have to tweak stuff in order to get it to sound right. The mix is already pre-destroyed! When I'm picking sounds and building my templates I'm always listening through the channel comps and stem limiters, so there won't be any surprises later. It's like, "does this sample still sound good when it's getting stepped on like it inevitably will, or is it weak sauce no matter what I do to it?" In some circumstances I'll use saturators to get even more bite out of a stem if it's getting buried, like on spicatto strings stem in a heavy action cue with lots of loud war drums. At the moment Waves Abbey Road Saturator seems to be my favorite since it has that weird pre-empasis / de-emphasis eq thing going on. Can add some scratchy bite AND some compression all in one, so that's a "mojo" processor I guess. Still hits the stem limiter afterwards though, just to keep a lid on the levels.
But of course this is all done by ear, so if something is getting hit too hard I'll back it off, but I usually do this by backing off the fader and not adjusting the compressors or limiters. This just changes how hard the incoming signal is hitting the stem processors.
Last edited:
emasters
Senior Member
Interesting read - thx Charlie. I also use the stock Logic compressor as the default on all tracks, then add others as necessary/desired. Given the multitude of compression options available now, funny how the basic DAW compressor continues to have value and get a lot of use -- it does work well for what it does.
Right.
Sound workflow. Unfortunately it doesn’t work with the material I receive.
Ha. Loudness is your middle name.
Last edited:
PaulBrimstone
Far, Far Away
If it’s good enough for CC it’s good enough for me. From tomorrow I’m putting Logic compressor on my cereal and everything else.
proggermusic
Keyboards and Woodwinds, Jazz-Funk-Ambient?
@charlieclouser – Fantastic information, man, thank you for sharing all that! Fascinating to learn how various folks who are far along in the industry work. Cheers, you're quite generous with this thorough and excellent info.
I agree with the choir, Logic's stock comp is all I've ever needed for any real work a compressor does. If I ever use anything else, it'll be for color (usually something from Black Rooster or NEOLD). In addition to working great, the Logic comp seems to be virtually invisible to CPU use. How very handy!
I agree with the choir, Logic's stock comp is all I've ever needed for any real work a compressor does. If I ever use anything else, it'll be for color (usually something from Black Rooster or NEOLD). In addition to working great, the Logic comp seems to be virtually invisible to CPU use. How very handy!
@charlieclouser in terms of loudness/headroom, what are your thoughts on Dangerous Music Compressor?
charlieclouser
Senior Member
Never used it. The only hardware compressors I have these days are three Distressors (a stereo pair in my recording chain that is 2x Avalon U-5 > 2x AMS-Neve 1084 > 2x Distressors), plus a third dedicated on my bass guitar chain (SansAmp RBI > Distressor), a UA 6176 channel (with a Roland SN-700 in between the pre and the comp, dedicated for my lap steel guitars), and a UBK-modded Fatso (for when I want to absolutely punish something).
None of these are used on stems or mix bus; for that purpose it's only look-ahead plug-in compressors for me!
Love UBK hardware.
Gagarin Moore
(formerly Page Lyn Turner)
So, compressor on every instrument of an orchestral then? (piccolo, flute,... all the way down to Doublebass?) with the settings CC provided?
Default Platinum Digital mode
2:1 ratio
-30dB threshold,
minimum attack, auto release, hard knee
and auto-gain at -12.
Default Platinum Digital mode
2:1 ratio
-30dB threshold,
minimum attack, auto release, hard knee
and auto-gain at -12.
Gagarin Moore
(formerly Page Lyn Turner)
I'll call it Charlie as well : )Here:
charlieclouser
Senior Member
Well, maybe not if you're going for a realistic simulation of an orchestra with a natural balance.... but I'm never going for that, or even a halfway version of that. I stack things up so much that I might have a solo cello tremolo playing over the equivalent of a 800-piece string section, with a couple of other tutti slams, risers, and fx along for the ride. So natural sound is not my thing, which is why I'm free and easy with tons of compression just to get things to blend.
Having strong stem compression lets me do things like have three or four layers of spicattos stacked to get the grunt and bite that I want, but have one or two drop out here and there to get changes in density and intensity. Without the strong limiters on that stem, dropping a layer or two will cause the level to drop massively. And automating around all those spots would take ages, so I just let 'em slam the limiter hard enough that the apparent level doesn't change all that much, only the tonality. Seems to work okay. But, again... not natural.
charlieclouser
Senior Member
VIOLATION! The Attack knob is not quite at its minimum setting. Zero ms is required in order to say to every molecule of audio "you shall not..... paaaassss!"Here:
Haha. Same here. Instruments are all over the place in the stereo field. If it fits the cue and sounds amazing, who cares?
As for analog consoles, my ears truly hate the clinical aspect of the digital realm, so I pass all the stems through a character summing box and AD a 2-track sum back into the DAW. Audiences can never tell the difference…it’s mostly for my own comfort when monitoring the playback.
Last edited:
charlieclouser
Senior Member
Now that mixes are moving to Atmos, my delivery stems have changed from 5.1 to quad, and I may eventually switch to just stereo stems, which makes it easier for the folks on the dub stage to deal with in the Atmos array. So using an analog summing matrix or console becomes a possibility again. But it would have to be fairly big, with a bunch of stereo busses to route the stems to for printing. A BCM10 wouldn't do the trick, but a Duality might.... but I don't think I'll be going that way, it's just too much cabling and hassle. But it's funny to see things loop around all the way back to where that's an option (in theory).
A buddy who's had SSL consoles for decades finally dragged his last beat-up 4k console to the curb, and has mixed his last album in ProTools and it's his best sounding one yet. The bottom is big and clear like never before but the mix still sounds crunchy and thick as it should. And this is heavy industrial metal, the kind of material that would benefit most from mixing in analog. Go figure.
The last mix I did on analog was such a chore, with all 72 channels on the 4k used and a Mackie 8-bus leaning up on its side for more fx and sampler returns, and once I'd gotten the drums and bass to be hitting the mix bus just right and bending it just so, I unmuted the guitars and synths and then the whole mix started to fold up and I had to re-balance everything to get back into the mojo zone. Ugh. Stuff like that, plus automating channel mutes to lower the noise floor on quiet bits etc. is why I don't really miss mixing OTB.
Depends what a lot of cabling means to you and how much I/O you'd need to make this worthwhile. For example, with the 8816s, I was able to wire 8x DB25 to a Ferrofish A32, and that was pretty much it. Mind you that the 8816's L/R are inverted, so you'd have to put together your own DB25s or invert the channels in software.
I love the SSL for its DYN section but that's pretty much it. The rest doesn't really make much difference to my ears. The NEVEs, on the other hand, night and day. Pure joy. I'm not sure if it's because I spent way too much time in front of an 88RS or what....because I'm the only one who seems to care. When I show others different passes of a stem, they seem, at greatest, indifferent. haha.
I never really mix in the summing (as weird as it may sound to you)...they are just a feed in and spit out "character box" so I can get that aural vibe and stereo width I am "looking" for. I calibrated each individual channel to cap at -24dBFS (ARIB), and left it at that. Stems are sent to it and the sum returns as a 2-track sum track into PT. All for my aural pleasure because nobody seems to hear much difference unless they are mixers.
