Two Stage Compression routing

Hello,

Does anybody have experience with Two Stage Compression in Cubase?

Two Stage Compression’s purpose is to have different compression for the quiet and loud passages of a drum performance in a song. Two Stage Compression routes all the drums to a stereo compressor, then routes the compressed return, along the original (unprocessed) drums, to a second stereo compressor to be compressed in parallel. The monitoring would be done where the output of the second compressor is.

The video explaining it is here: http://www.youtube.com/watch?v=NvHlx3XHpBA

Unfortunately, it’s explained in console language with hardware inserts and console returns. However, I’ve read at another site how it would be done within Cubase and wanted to check if the routing recommendations were verified to be correct here at the authorized Cubase site.

Would this routing setup in Cubase accomplish my goal?

  1. Set up a stereo Group Channel for all of the drums. Lets call this “All Drums Group”. Have all the outputs of each individual drum channel routed to this “All Drums Group”.
  2. Create two new stereo groups: “Bus Comp Quiet”, and “Bus Comp All/Loud”.
    • Bus Comp Quiet will have a compressor set for quiet sections
    • Bus Comp All/Loud will have the compressor set with a higher threshold so that it is only active during loud passages. This will also be the only group to be monitored in the mix.
  3. On the “All Drums Group” create one pre-fader send (use Send 1) to “Bus Comp Quiet” and set this to 0db (full).
  4. On the “All Drums Group” create another pre-fader send (use Send 2) to “Buss Comp All/Loud” and set this to 0db (full).
  5. Turn the fader of “All Drums Group” all the way down (this channel is now only sending signal via the pre-fader sends)
  6. Now send the output of “Bus Comp Quiet” into "All Drums Group.

Thanks for any information on this technique and your routing suggestions.

More or less right.

there are various ways to do this.

Send all drum channels (Drum Mix) to group channel, select group channel output to not connected.

Set up three stereo groups, Drums No Compression, Drums Light Compression, Drums Heavy Compression

From Drum Mix Group set up three sends (no need for pre fader) to the three groups @ 0dB

Then if you require a further Group and route the three previous groups to Kit All Level for easy drum volume control

That will give you Parallel compression as well as two types of compression to choose from

Never in a million years would I do it that way!
Scrap the pre-fade sends and use Post fade, as normal.
Do not drop the uncompressed fader, but leave it well alone.
The additional 2 groups are fine, simply automate the send on & off at the right points.

Advantage to using this method is that you get 2-stage parallel compression without reducing the dynamic range of the original submix, and the compressed groups are enhancing the original uncompressed group.

This is all very confusing! But I too would love to get my head round it…

From watching the 2nd part of the video here…
http://www.youtube.com/watch?v=MjrbarinF3k&feature=related
… Mike appears to be describing this:

  1. Split the signal from the drums into two.
  2. Send first half to a compressor that has a low threshold, in order to squash the signal flat (high ratio, fast attack, fast release)
  3. Combine the two signals back together. Now you have parallel compression that will let you hear the detail in the lower dynamic sections (you only need a small amount of the squashed signal, in order to hear a big difference)
  4. Now, take the whole lot and send it thru a second compressor. Set the threshold of this one higher, so that it only starts working during the higher dynamic sections. The output from this compressor is what goes to the main mix.

Is that it?

@Split: Not sure how you intend to recombine your 3 busses, but if you mix all 3 back together (in parallel), then how can you control the peaks? The “uncompressed” buss will always let peaks thru, even though the “heavy” buss and “light” buss are doing lots of compression.

Pass the aspirin!
FD :slight_smile:

I’m not Split, but there are many ways to defur a feline.
The approach of leaving the uncompressed group alone and sending to separate compressed groups works well for me (FWIW, this trick of multiple compression types also works very well in mastering - you can set one up to control the low end, another to control/enhance the mids etc).
I like dynamics, which is why I always have an uncompressed group in addition to the parallel one. Controlling the peaks is done by getting the gain structure right on the actual audio tracks, and there is no real need to have things ultra crushed - a light, subtle touch is often better than the sledgehammer, but it depends on the sources & the genre.

Another way to eliminate the peaks issue is to mix to R128. PPMulator 3 XL is great here, as it works on loudness, not peak levels - with this way of working, you get nowhere near 0dBFS so peaks are not really a problem as you have 18-23dB of headroom available.

Almost forgot.

Will recheck this to be sure it can be done in Cubase (I am certain it can) but another option here is to use your group tracks for all subgrouping & EFX, and leave the EFX tracks as “virtual VCA” type faders.
They are not real VCA of course, but the idea is that all the drum tracks (for example) can be routed to an EFX track instead of the 2-buss, giving you full control over the entire submix on one fader.

Yes, that’s it as I understand it.

@Split. Why 3 sends? I may misunderstand how consoles work, but my understanding is that there is no uncompressed drums during the “loud” passage.

@Neiwilkes. Appreciate your feedback. I think (not sure) that the original idea was to have the combination signals (compressor 1 + uncompressed signal) always being sent to the second compressor. I’m not sure how your method does this, but I am not that sophisticated in console routing or Cubase.

Thanks for your help and understanding guys. The thing I don’t understand about the original method I posted is how the return from the “quiet compressor” gets routed. By “sending” the output from the “Bus Comp Quiet” Group Channel back to the "All Drums “,Group Channel does he mean an aux send, or an output (at the top of the channel) send? Why not just aux send the “Bus Comp Quiet” to the “Bus Comp ALL/Loud”, along with the send going from the “All Drums Group” aux send going to the Bus Comp All/Loud”? That way, at all times the “Bus Comp All Loud” compressor is always getting a mixture of the compressed and uncompressed signals. It just does nothing during the quiet passage because the threshold is set too high.

This brings up a routing question regarding routing from one Group Channel to another Group Channel: When you have a Group Channel with a compressor insert and want to route it to another Group Channel with a compressor insert, is there any difference routing via a pre fader send at 0dB, compared to routing via an “output” from the Group Channel to the input of the other Group Channel?

Thanks again.

That’s okay.
As I pointed out, there are many ways to do this, none of them “righter” than any other - all that matters is getting the mix you want, and if you manage this then you did it the right way - simples!
You can manage this easily enough though - simply route from original uncompressed to the first and second compressors and switch the first one in & out as you need to by automating the send on/off.
Please feel free to send me your CPR file via email and I will set it up a few ways for you & send them all back as I would do it, and as I think you are looking for as well.

The method I described basically gives you one group of uncompressed drums, one group of compressed drums and one group of very compressed drums, hence the three sends. You could easily not bother with the uncompressed group send if you don’t bother to select not connected on the uncompressed drum group output. I merely suggest that you could combine these three groups to a further group to have easy control of the overall drum level.

It’s also quite good to include drum FX into either the overall level group or one or more of the other groups!

As Neilwilkes points out it’s probably better to have the uncompressed signal in the drum mix along with compressed drums so you don’t completely squash the hell out of it (unless that’s what you’re after) A well used technique.

Thanks for your reply Split.

I believe the goal of Two Stage Compression is to have the threshold of the second compressor set high enough so that it doesn’t squash the hell out of the two signals, the uncompressed send + the send from the Quiet Compressor. The video give a good representation of how open the drums sound using this technique. They don’t sound squashed at all.

Thanks for your reply Neil. I am looking to see if the way Two Stage Compression is described in the video can possibly be done in Cubase. Using your automation idea, my understanding is that you’d always have a Drum Group Channel send to the “soft compressor” Group Channel, but during the loud section there would be both a send from the uncompressed original signal to the 'loud compressor" Group Channel + a “soft compresser” Group Channel send to the loud compressor channel during the loud section as well. This would be done by (a) always having the send to the “soft compressor” Group Channel at 0dB, plus automating the send to the loud compressor from the Drum Group Channel during the loud portion, and also automating a send from the “soft compressor” Group Channel to the “loud Compressor” Group Channel during the loud segment. Correct?

If this is correct, would I reach the goal of having only the combined sound of the soft and loud compressors going to the master bus during the loud section? Do the Drum Group + “soft compressor” Group Channel sends prevent those channels from going to the master bus?

Thanks again for your help.

I don’t have a .CPR file yet. So, I am just in the theoretical stage of this.

Indeed, I think I was saying there’s nothing wrong with squashing the hell out of something if that’s what you want.

Of course the nice thing with having the three groups is you can choose to mix them any way you deem fit.

The technique of using multiple busses of various compression can be applied to other thing other than drums.

Check out this video

http://www.youtube.com/watch?v=6czwUasU9eU&list=SL

Thanks for the link to Michael Brauer’s appearance on Pensado’s Place.

I found this SOS article on Parallel Compression interesting as well: http://www.soundonsound.com/sos/apr09/articles/cubasetech_0409.htm

This is all great info! My vocals are sounding more expensive already - thanks for sharing!

Michael Brauer explains more about his “multi bus compression” here (see box at bottom of page)
http://www.soundonsound.com/sos/nov08/articles/itbrauer.htm

Care to share what compressors you are using, and if you’re using different post compression EQ’s ?