New York Style parallel compression

[rasputin1963]

I'm investigating a VST compressor which reportedly allows what it calls "New York Style compression". Apparently this entails a signal being sent through parallel, presumably one signal compressed, the other, not. With surely a Mix knob allowing you to blend the two, to taste.

 

Are you familiar with this technique? What might be the sonic effect, or mixing advantage? What "New York" production styles first launched this sound, and for what musical genre? (My wild guess is that it probably has something to do with the 1970's Funk/Disco era...?)

 

Thanks, ras

 

[nat whilk II]

I"m sure I have no idea why New York gets the credit for this, but it's a very standard way to deal with compressed tracks - namely, you have one compressed track, and a second, un-compressed track, and as you ventured, you blend them to taste.

 

Quite the tried and tested technique, especially with compressors that color the sound significantly. You don't have to run 100% of the signal through the compressor - having the original, un-compressed track still blended in keeps you from committing the track entirely to the sound of the compressor. When you need to compress, but don't want it to have that compressed sound, this is the way to go. And it enables you to introduce only as much compression as is needed - which is usually less than people think.

 

The other way to gently but effectively compress without introducing too much of the particular sound of the compressor, is to compress in stages, a little at a time. Maybe compress just a tad during tracking, then a bit more during mixing, then a final touch during mastering. You can also then use different compressors at different stages of the process, and that opens up another bunch of creative options.

 

Other, more sophisticated studio types here will have more sophisticated reasoning and techniques, no doubt.

 

nat whilk ii

 

 

 

 

[rasputin1963]

Thanks, Milton. In this particular case, I am wondering how to tame sonorities with a harsh/loud attack. I'm thinking of a plucked string primarily. Sometimes the initial transient of a guitar pluck can be many times louder and denser than the decay/sustain/release portion. Yet most compressors you apply will, with a fast compressor attack, JAM that volume down very noticeably, then quickly recover, yielding the notorious "breathing/pumping".... that sounds nasty if you are going for a naturalistic effect.

 

(I do appreciate that, with any plucked string, you DO, in fact, need that sharp/hot transient to TELL you that you're listening to a plucked string.)

 

With a guitar, it's a bit more involved, because each of the six strings on the guitar has a different attack, a different POP to the first strike. A compression solution that tames the high-E string nicely, can have deleterious effect on the lowest E-string.

 

I suppose, push-comes-to-shove, you could manually tweak your note's envelope in your DAW just by drawing a Volume envelope over the waveform... both tedious and hit-and-miss, I'd think.

 

Thoughts? How do you get a guitar to sound very very gentle?

 

And P.S., just as a footnote: I am finding I most often do not like the sound of compression. That's probably my inexperience talking-- it's probably a necessary evil. I find I like a track with wide, even dramatic dynamics. Paul McCartney says that compression makes a recording sound more "record-y", haha.

[Iamthesky]

A very short look-ahead on the compressor (assuming the plug-in allows it) would let you run a milder attack while still controlling the transient peak. Depending on the context of the sound around the transient, that might give you another option.

 

I too use the wet/dry mix technique, and drawing envelopes to control unusual spikes. They're all just more tools in the bag.

[nat whilk II]

I am wondering how to tame sonorities with a harsh/loud attack. I'm thinking of a plucked string primarily. Sometimes the initial transient of a guitar pluck can be many times louder and denser than the decay/sustain/release portion. Yet most compressors you apply will, with a fast compressor attack, JAM that volume down very noticeably, then quickly recover, yielding the notorious "breathing/pumping".... that sounds nasty if you are going for a naturalistic effect.

 

(I do appreciate that, with any plucked string, you DO, in fact, need that sharp/hot transient to TELL you that you're listening to a plucked string.)

 

With a guitar, it's a bit more involved, because each of the six strings on the guitar has a different attack, a different POP to the first strike. A compression solution that tames the high-E string nicely, can have deleterious effect on the lowest E-string.

 

I suppose, push-comes-to-shove, you could manually tweak your note's envelope in your DAW just by drawing a Volume envelope over the waveform... both tedious and hit-and-miss, I'd think.

 

Thoughts? How do you get a guitar to sound very very gentle?

 

 

I would think just keeping the compressor on a low ratio would help with the pumping. Also, setting the threshold carefully can also solve that problem. If the volume level of the sustain portion of the plucked note is below the threshold, then the compressor should never touch the sustain.

 

You could tweak the attack transient by drawing a volume envelope...but really, that's what the compressor does anyway - imposes a volume envelope on the material, the shape being determined by the ratio, and the attack and release settings.

 

One thing you might try, if you have the time to experiment a bit, would be this:

 

Record a few guitar notes, then copy that to two more tracks.

 

Track 1 will remain as is, your original unmessed-with recording

 

Track 2, go ahead and draw the envelope, tweaking it to where it's just what you want.

 

Track 3, insert a compressor and methodically try different settings until it automatically matches (or close enough matches) what you're hearing in Track 2. This can be slow and tedious, so maybe starting with a preset would help. Once you have the settings that really work, save those settings as a preset (and even write it down somewhere in case you lose the preset) and that can be your go-to compressor setup for guitar.

 

You could also insert two or three compressors into Track 3, muting all but one at a time. Set them all to the same settings and see how differently they respond. Compressor behavior is very subtle and sensitive and can differ terrifically from one to the next compressor. And from one type of sound material to the next, also.

 

One way to get that gentle, breathier sort of guitar sound is to move the mic further away from the guitar, and/or use a room mic to pick up the ambient signal and blend that in with the more close-mic'd signal. Also using lighter gauge strings, and simply strumming more gently, can help. Lastly, for strummed guitar, double or triple track the part, and the attacks will not be all at the same time which will soften the transients (like a choir can soften sibilance.)

 

Best o'luck -

 

nat whilk ii

 

 

 

 

 

[Anderton]

Hey Ras...you use SONAR, right? If so, these are the droids you're looking for...they already do parallel compression due to having dry/wet controls.

 

However, I think what would soften the attack best for you isn't compression but transient shaping. SONAR has an excellent transient shaper, the TS-64. It has a lookahead time, though, so your overall latency will be longer...although that won't matter when you're mixing. The PX-64 Percussion processor also has transient shaping; it's not quite as sophisticated as the TS-64, but the lookahead time is much less so there's lower latency.

 

I sometimes think transient shapers "don't get no respect," but they can do mahvelous things in a mix.

 

 

[MikeRivers]

My concept of New York compression is what you need to do in order to get one of the pastrami sandwiches that they serve at the Celebrity Deli into your mouth.

 

I first heard about the "parallel compression" technique of mixing a compressed and uncompressed track 20 or so years ago, in San Francisco, though they called it "co-compression out there." This was back when recorders went "whirrrrr" and compressors were hardware. The idea was to maintain the existing dynamics of the track while increasing the density of low level content by gently adding a bit of the track compressed with a low threshold setting.

 

The DAW version probably makes most sense when it's a wet/dry plug-in so they can get the two streams accurately lined up in time. I suspect that some people who are copying a track and compressing it, then adding it to the DAW mix are getting something in the deal that they didn't plan on and may not recognize - some comb filtering due to the latency of the compressor that isn't perfectly compensated for (or just ignored).

 

I recall the NY-2A compressor from Electro-Harmonix, though I never found out what was NY about it other than that the company is located in New York.

 

 

Note the genuine tuning eye next to the VU meter for fast-responding peak indication. It used an electro-optical element for the gain control, and gave you a choice of a fluorescent, LED, or incandescent light source. I don't know if there were three different modules or one photocell with three different lights, but it was an interesting concept. I only saw one, once, at a NAMM show.

 

 

 

[Ernest Buckley]

Waves H-Comp does the "NY Compression" well with its dry/wet knob. I use it often on drums to give it a bit of a push and to fill out the sound. Not sure why NY is credited with it though… I remember reading about it years ago but honestly, don`t remember…

[Phil O'Keefe]

Mike, I always thought NY-2A was a play on LA-2A; they are both optical compressors, but EHX couldn't exactly use the same name without running into issues with Urei / Universal Audio.

[Phil O'Keefe]

Ras, Wiki actually has a pretty decent article on parallel compression. I also wrote an article on it a while back that you might find interesting.

 

One of the things I've always equated with "New York" style parallel compression is pretty substantial amounts (+6 to 10dB) of high and low shelving EQ boost (~100Hz / 10kHz) on the signal feeding the compressor, especially when the technique is used for drums.

[MikeRivers]

Mike' date=' I always thought NY-2A was a play on LA-2A; they are both optical compressors, but EHX couldn't exactly use the same name without running into issues with Urei / Universal Audio.[/quote']

 

I seem to remember thinking that myself, but it isn't really a much like an LA-2A other than the photocell gain element. But Electro Harmonix did a lot of odd stuff, some successful, some not so. And they're still in business, so they must have been doing enough of something right.