NY/Parallel Compression - Questions

[Bezo]

I've been trying to implement this technique but I don't think I'm getting the results I'm looking for. I record instrumental rock, hip hop and funk, and I want my drums as aggressive as possible.

There may be variations of this technique, but I'll list what I do in case there's a problem with my set up:

• 2 stereo aux tracks armed with compression plugs with identical settings, with one compressor disabled.
  I'll call them Comp & Sub(disabled).
• Kick, snare and sometimes bass routed to both the comp & sub aux tracks
• All sends to the comp track are pre. All sends to sub are post
• No output from the original tracks. Comp and subs are routed to the master.
• Set parameters to taste. Threshold somewhere between -25db and -10db. Ratio somewhere between 4:1 and 8:1. Very quick attack and release times. Hard knee. And I increase the gain so that the loudest hits are back near 0db.

My first problem is that my drums are not as punchy as I would like. I've been experimenting with settings and I'm getting the best results with what I listed above, but still not the sound I want.

Secondly, any time I'm approaching the sound I'm looking for, I'm clipping the master. In order not to clip the master, I lower the sends or original track faders. My worry is that I'm not pushing the compressor. Although I'm able to make-up 5-7db of gain, should I worry that the output of the compressor/aux tracks are not near 0db?

If it makes a difference, I'm using either samples or drum machines for my drums.


Anyone have a better variation of parallel compression?

Anyone have an alternative to parallel compression?

[davetron5000]

I've done this with good results, but your setup seems a bit confusing and I've never had level problems. I do this:

• Create Two Stereo Aux Tracks with inputs Bus 9-10
• Route all drum tracks Outputs to Bus 9-10
• Put a compressor followed by an EQ on on stereo track
• Set compressor for 10:1 and lots of compression
• Use a hi-shelf at 10Khz and lo-shelf around 100Hz on the EQ and add about 3-6db of gain
• Set uncompressed stereo aux to unity and then bring up the compressed stereo aux as needed.
• Adjust EQ as needed (sometimes you end up with too much cymbal or kick drum)
• Link the two tracks and now you can easily adjust them for mixing

When I include the bass in the mix, it tends to reduce the low-end definition for me, but it probably depends on the song arrangement.

[Sevenlights]

The technique I've seen used most (usually on SSL boards) is all your tracks are routed eq'd and compressed to your taste to the master. Then a buss sends all the drums to a stereo comp with faster attacks and release to blend in with the original singal in master buss as well. Some people send the bass too and I've also seen sub mixes built in the same manner for everything else. I've tried or seen anyone do it they way you speak of, although I've heard of it, people floating the whole mix and only sending the subs to the master.

Or I've heard of mixing the way you normally would, then sending a pre-buss with unaffected signal to the master as well, thus blending in your unprocessed with your processed signal, I think this was a British origination.

[Bezo]

Thanks for the quick reaponse!

• Create Two Stereo Aux Tracks with inputs Bus 9-10
• Route all drum tracks Outputs to Bus 9-10
• Put a compressor followed by an EQ on on stereo track
• Set compressor for 10:1 and lots of compression
• Use a hi-shelf at 10Khz and lo-shelf around 100Hz on the EQ and add about 3-6db of gain
• Set uncompressed stereo aux to unity and then bring up the compressed stereo aux as needed.
• Adjust EQ as needed (sometimes you end up with too much cymbal or kick drum)
• Link the two tracks and now you can easily adjust them for mixing

A few questions about your set up:

Both your aux tracks are bussed to the same inputs?

You didn't say, but I assume you're disabling the compressor on one of the aux tracks since you refer to it as uncompressed.

Are your sends to both aux tracks post or pre?

I've seen it done that way, but what's the theory behind the low & hi shelf? I guess to control the extreme lows and highs after compression. I usually only send the kick and snare, but maybe I'll try sending the entire kit. Do you think that helps glue things together?

Ohhhhhhh! You're controllig the compressed signal with the aux fader. I've been doing it with the sends, and had the compressed aux track fader at unity. That could be problem #1. Are your sends to both aux tracks at unity?

Are you sending the original tracks, uncompressed aux and compressed aux all to the master?

What do you mean by "link the 2 tracks"? And how are you doing that?

[davetron5000]

Both your aux tracks are bussed to the same inputs?

Yup.

You didn't say, but I assume you're disabling the compressor on one of the aux tracks since you refer to it as uncompressed.

Well, one aux track just has no compression on it (there's no reason to put an insert on your track and then just disable it). Each drum track would be eq'ed or compressed as you saw fit. The process here is more like mastering the entire drum mix. So you have two seperate aux tracks, one with no inserts/fx and the other with this compression and stuff.

Are your sends to both aux tracks post or pre?

Pre. I would guess if they were post, it would be harder to control the
balance of the two signals.

I've seen it done that way, but what's the theory behind the low & hi shelf? I guess to control the extreme lows and highs after compression. I usually only send the kick and snare, but maybe I'll try sending the entire kit. Do you think that helps glue things together?

Honestly, I'm not sure, but it just ends up sounding a bit better. I read it somewhere when I first learned of this technique, but there are lots of variations. Depending on your drum sounds and arrangement, it might not sound good.

Ohhhhhhh! You're controllig the compressed signal with the aux fader. I've been doing it with the sends, and had the compressed aux track fader at unity. That seems to be problem #1.Are your sends to both aux tracks at unity?

Right. I'm not actually using any sends. I'm setting the output buss of my drum tracks to Bus 9-10 and then setting the input buss of the two different auxes to Bus 9-10, essentially creating two "master faders" for just the drums.

What do you mean by "link the 2 tracks"? And how are you doing that?

[/QUOTE]
Create a mix a group (Cmd-G I think). Then you can adjust the level of your "mastered" drum mix easily, while maintaining the balance between compressed and dry signals.

[Bezo]

The technique I've seen used most (usually on SSL boards) is all your tracks are routed eq'd and compressed to your taste to the master. Then a buss sends all the drums to a stereo comp with faster attacks and release to blend in with the original singal in master buss as well.

That eliminates the need for the uncompressed aux track. And doing it this way sends 2 compressed versions to the master. That sounds like it could add more punch, which is what I'm looking for. I'll experiment with this.

...and I've also seen sub mixes built in the same manner for everything else.

Everything else as in the other instruments? I thought about trying that, but there are not enough buses in LE. I use them up with reverbs, delays and other effects. I'd have to print as I went along.

I've tried or seen anyone do it they way you speak of, although I've heard of it, people floating the whole mix and only sending the subs to the master.

I need to brush up on my engineering vocab. Floating = no ouput?


Thanks for the response!

[Bezo]

I wonder if both my aux tracks use the same 2 buses. This is part of my templates, so I haven't paid much attention to that since I set it up. I'll have to double check that this evening.

Well, one aux track just has no compression on it (there's no reason to put an insert on your track and then just disable it).

Other than adjusting the compressor settings, I haven't given my set-up much though until recently. But when I learned how to do this, it was explained to me that this would delay the compressed and uncompressed signals equally. Maybe using the same buses for both aux tracks eliminates that problem.

Each drum track would be eq'ed or compressed as you saw fit. The process here is more like mastering the entire drum mix. So you have two seperate aux tracks, one with no inserts/fx and the other with this compression and stuff.

Makes sense.

Pre. I would guess if they were post, it would be harder to control the balance of the two signals.

That's my thinking behind sending pre to the compressed aux. I'll try sending pre to the uncompressed aux as well.

I'm not actually using any sends. I'm setting the output buss of my drum tracks to Bus 9-10 and then setting the input buss of the two different auxes to Bus 9-10, essentially creating two "master faders" for just the drums.

Ahhhhhh. I'll have to try that.

Create a mix a group (Cmd-G I think). Then you can adjust the level of your "mastered" drum mix easily, while maintaining the balance between compressed and dry signals.

OK. I'm familiar with groups, and I use them in this manner. "Link" threw me off. I need to get out more.

[sonikbliss]

I do it pretty much like davetron does.

First I get a good drum mix going, compressors & EQ's on individual drums if neccesary, reverb\fx sends on each track so I can have control over the amount of verb on each drum.

Once I have a good overall drum sound I assign the output of every drum track to Bus 1-2 (for example), which pulls them out of the master stereo bus, and setup two stereo aux tracks to receive input from Bus 1-2. This is effectively sending Post Fader signals to the two aux tracks since I believe it's the actual aux sends that can be set at Pre\Post Fader, not the buses (correct me if i'm wrong)(which is good because I want to preserve the mix that I already did).

Once I have this setup I use the two aux faders to control my blend. Usually I just leave one aux track with no plugins and the other has a compressor smashing the hell out of the signal. Blend to taste.

The only time I put a compressor on the clean aux and bypass it is if I am using a plugin (such as T-Racks) that causes latency and puts the two stereo aux's out of phase with each other. Sometimes I choose to eq the two stereo subgroups & bus them to the verb also.

This setup can work great with other sources too, try it out with vocals, bass, gtr. You might be pleasantly suprised.

[Bezo]

Once I have a good overall drum sound I assign the output of every drum track to Bus 1-2 (for example), which pulls them out of the master stereo bus, and setup two stereo aux tracks to receive input from Bus 1-2. This is effectively sending Post Fader signals to the two aux tracks since I believe it's the actual aux sends that can be set at Pre\Post Fader, not the buses (correct me if i'm wrong)(which is good because I want to preserve the mix that I already did).

The output is post fader. Sends can be pre or post. I'm definitely going to try this.

The only time I put a compressor on the clean aux and bypass it is if I am using a plugin (such as T-Racks) that causes latency and puts the two stereo aux's out of phase with each other.

This was my understanding when I first learned the technique. I'm using Digi's compressor. The explanation made sense, so I never tried it w/o the disabled compressor on the clean aux to see if latency was a problem.

Sometimes I choose to eq the two stereo subgroups & bus them to the verb also.

So, when you do this, your drums are not output to the master fader until after the reverb? Interesting. Or am I misuderstanding. I'm having a brain freeze. Sends and outputs route through buses, right? Are you setting the output of the sub groups to the verb, or are you sending to the verb?

Either way, this might be what I need to glue my drums together. I've been sending the individual tracks to a reverb armed aux in order to add different amounts to each drum instrument while still using just 1 plug.

But why the 2nd EQ on the sub mixes? Especially the non-compressed one?

This setup can work great with other sources too, try it out with vocals, bass, gtr. You might be pleasantly suprised.

I definitely plan on trying it on other sources if I can free up some buses. What effects do you tend to send to versus effects you put directly onto individual tracks? I bus to verbs, delays, hi & low pass filters and maybe one or 2 others that escape my memory at the moment. Every mix, I find myself wishing for more buses. But I've been wasting 4 buses with this technique. Using the output to send the drum tracks to the aux mixes will free 2 of those up.

And thanks for the response! I didn't expect to learn this much while talking about parallel compression.

[sonikbliss]

So, when you do this, your drums are not output to the master fader until after the reverb? Interesting. Or am I misuderstanding. I'm having a brain freeze. Sends and outputs route through buses, right? Are you setting the output of the sub groups to the verb, or are you sending to the verb?

Either way, this might be what I need to glue my drums together. I've been sending the individual tracks to a reverb armed aux in order to add different amounts to each drum instrument while still using just 1 plug.

But why the 2nd EQ on the sub mixes? Especially the non-compressed one?

I definitely plan on trying it on other sources if I can free up some buses. What effects do you tend to send to versus effects you put directly onto individual tracks? I bus to verbs, delays, hi & low pass filters and maybe one or 2 others that escape my memory at the moment. Every mix, I find myself wishing for more buses. But I've been wasting 4 buses with this technique. Using the output to send the drum tracks to the aux mixes will free 2 of those up.

Well, for #1 I use an aux send to bus the 2 stereo subgroups to the same verb (on it's own aux track) that I sent the individual drums to while keeping them assigned to the master outputs. This way you still only have to use 1 verb plugin. I don't always do this but sometimes it will glue the track together a little better and even when I do i usually add only a very little verb since the low end from the kick can muddy things up.

As far as the EQ, it's just a matter of taste. Sometimes I may feel like I like the overall balance of the drums but they could use just a little boost\minus here or there and it's easier to do that with one plugin on a stereo subgroup than to tweak each drum individually.


#2 The general rule of thumb is:

You send to time based fx since you usually want to be able to blend the dry\wet signals together (reverb, delays)

You insert dynamics fx since you usually want to completely replace the original sound with the effected one (eq, compression)

You can send to some dynamics fx if you wish to process all the sounds in the same way, which also cuts down on the amount of plugins you have to use (such is the case with NY Compression 1 compressor for multiple tracks).

You can also insert time based fx if you wish for a 100% wet sound


I would use inserts for hi/lo pass filters instead of buses


Also if you set the output of a track(s) to Bus 1-2 (or any bus for that matter) it's still taking up those buses. So unfortunately that's not going to free any up for you.

Hope this helps-Jason