Setting up two-stage compression in Tracktion - tips?

[bipedal]

Definitely a technique I want (need!) to try out soon.

I followed the logic of two stage compression as presented in a recent Tape Op issue -- great article. Just haven't quite figured out how to actually implement it within Tracktion (I'm using v2.1).

Any tips from other Tracktion users?

Thanks in advance -

- Jay

[ashcat_lt]

Sorry (kind of) to make a post that doesn't actually help you accomplish what you're trying. I don't use Tracktion. But this thing has kind of been bugging me after everybody keeps coming asking "Two-stage compression in DAW-X?"

My question: Is it really necessary in the box? The idea is to have one compressor work on the quiet sections and another on the louder, no? Why not just automate the compressor settings?

[bipedal]

Hmmm.... Like, set up a compressor's ratio / attack /etc. so they adjust on the fly during the track? Good suggestion -- never occurred to me to automate that! That route might be a lot friendlier to my CPU...

Thanks -- I'll play with it.

Cheers,

- Jay

[;ivlunsdystf]

That's funny, I just did a two-stage sort of thing yesterday in Tracktion - it's a nice way to use two different compressors, of course - I just dragged them as VST's into the master out section (right before the master fader icon, lower right of the screen)

I resorted to this stunt because just using the Vintagewarmer alone was leading to way too much pumping in the bass, at least the way I was using it yesterday ...

I could have tweaked the controls of one of them, of course, and not used the other, but my ears told me that the two I used each set at high threshold and low ratio sounded much better (more subtle, in this case)

Of course your ears will eventually guide you to the best method for whatever you're trying to pull off ...

EDIT - for my application, a couple of compressors running as VSTs on the master bus seem to only take a negligible amount of CPU.

[Electricide]

you can set up a rack so that the signal goes to compressor A, then to the outs, AND ALSO from compressor A to Compressor B, then to a fader then to the outs. SO you could do parallel or series, and adjust the volume along any point(s) in the chain.


I think just setting two compressors is a lot less work than automating three setting of one compressor for a 3 1/2 minute song.

[;ivlunsdystf]

I think just setting two compressors is a lot less work than automating three setting of one compressor for a 3 1/2 minute song.

BINGO

[fossiltooth]

Please see the original thread on this subject. It's covered pretty comprehensively.

Don't overthink it. Just because Caffrey is trying to make "two stage compression" seem like God's gift to mixing doesn't mean that it's complicated. Here's the original thread. I recommend you read it. It's pretty epic:

http://messageboard.tapeop.com/viewtopi ... c&&start=0

I gave a comprehensive step by step answer on page 2. Here's the basic gist of it:

Do you understand Parallel Compression?

OK.

Do that.

Then, route your parallel compressed track and your uncompressed track to a bus instead of the main mix. Compress that bus and send it to the main mix.

If you have trouble understanding this, you have a lot to learn about signal flow and engineering in general, and there are probably other things you should focus on before approaching this a technique. Regardless, I think my posts in the original thread will help you understand signal flow a little better. My second post outlines a simple step-by-step process that should help anyone who uses a DAW make sense of this technique.

Sometimes I wonder why I don't write try to for the magazine. If you don't find my explanation on the original thread significantly clearer, simpler and less pretentious than Caffrey's, let me know, and I'll work on my communications skills!

Otherwise, please help me obtain the audacity and the balls to send Larry a resume.

Good luck!

[ashcat_lt]

Yeah, I've had that "Ass-u-me" thing happen before, but I thought the OP was talking "Two-Stage Compression" as per the TapeOp article, rather than Serial Compression, which can be another great approach, but nowhere near the same thing.

I guess I can dig that "it's easier..." thing, except that you still have to find the ideal settings for each compressor. Automating between two predetermined ideal settings is about as much work as setting up all the routing necessary for Two Stage to work.

A more compelling argument might be if you wanted to use two different compressors for their individual sonic fingerprints. Like if you wanted a nice clean "transparent" comp on the quiet parts and a funky old crunchy "color" comp for the loud parts.

[fossiltooth]

Yeah, I've had that "Ass-u-me" thing happen before, but I thought the OP was talking "Two-Stage Compression" as per the TapeOp article, rather than Serial Compression, which can be another great approach, but nowhere near the same thing.

I don't know ashcat. If you ask me, what Caffrey calls "Two Stage Compression" is extremely similar to regular old serial compression. It's basically a combination of serial compression and parallel compression.

the idea is to do parallel compression and put another compressor (in series) across the two parallel channels...

[ashcat_lt]

Sure. Based on the way I've seen the two techniques described and demonstrated in practice, though, they seem to be quite different.

In Two Stage, one comp effects the quieter section, and is basically non-existant (masked) when things get louder. The second comp effects the louder sections, and basically doesn't do anything in those quieter sections where the first comp is actually making a differnce.

An the other hand, Serial is usually set up to where both comps are working at the same time, in order to get more gain reduction with less artifacts.

Or something...

[vvv]

I am not sure how Tracktion works, and what I am about to say is basically said above, but I'll give a real-world example in Cool Edit Pro 2.1:

1. Copy your track.

2. Leave the original alone; process the copy through a compressor (easiest in Edit View) and save.

3. Mix down the unaltered original track and the compressed copy through the second compressor on the master buss (called "Master Mixer").

Or,

1. Copy your track.

2. Leave the original alone; put a compressor on the insert (called "FX" in the "Track mixer") of the copy track in Mixdown View.

3. Mix down the unaltered original track and the compressed copy through the second compressor on the master buss (called "Master Mixer").


I post this because CEP only allows one buss (plus one insert, although you can stack effects within either) at a time.

(Note that CEP auto-adjusts for latency.)

[mcaff]

Sorry (kind of) to make a post that doesn't actually help you accomplish what you're trying. I don't use Tracktion. But this thing has kind of been bugging me after everybody keeps coming asking "Two-stage compression in DAW-X?"

My question: Is it really necessary in the box? The idea is to have one compressor work on the quiet sections and another on the louder, no? Why not just automate the compressor settings?

Because you can't OTB. ITB it's a control thing. You can work with fader in your mix windero or on a cotnrol surface rather thna a mouse.


But to really take your question a stem further, why not just automate the level and not compress in the first place?

[mcaff]

you can set up a rack so that the signal goes to compressor A, then to the outs, AND ALSO from compressor A to Compressor B, then to a fader then to the outs. SO you could do parallel or series, and adjust the volume along any point(s) in the chain.


I think just setting two compressors is a lot less work than automating three setting of one compressor for a 3 1/2 minute song.

That's for sure!

[mcaff]

Yeah, I've had that "Ass-u-me" thing happen before, but I thought the OP was talking "Two-Stage Compression" as per the TapeOp article, rather than Serial Compression, which can be another great approach, but nowhere near the same thing.

I guess I can dig that "it's easier..." thing, except that you still have to find the ideal settings for each compressor. Automating between two predetermined ideal settings is about as much work as setting up all the routing necessary for Two Stage to work.

A more compelling argument might be if you wanted to use two different compressors for their individual sonic fingerprints. Like if you wanted a nice clean "transparent" comp on the quiet parts and a funky old crunchy "color" comp for the loud parts.

I think you see who you can get to the same result more then one way. Many people use mix templates that they import session data from.

Because that paralell compression can bring up the quiet parts so much, you can have nearly a universal setting from the second compressor where the dynamics of any performance sound really good no mater how extreme they are.


If time is a factor due to the nature of the work (post) or budget (everyone) starting further along can be a big help.

[mcaff]

Yeah, I've had that "Ass-u-me" thing happen before, but I thought the OP was talking "Two-Stage Compression" as per the TapeOp article, rather than Serial Compression, which can be another great approach, but nowhere near the same thing.

I don't know ashcat. If you ask me, what Caffrey calls "Two Stage Compression" is extremely similar to regular old serial compression. It's basically a combination of serial compression and parallel compression.

the idea is to do parallel compression and put another compressor (in series) across the two parallel channels...

"Extrmemely" is redundant.

What you've described is exactly right. The question is, is there a difference having that dry signal come through the first compression in the daisychain?

According to Greg at Pendulum, this is how the narrowed the dynamics of classical music to fit on to vinyl without making the need jump of the record.