Bus Compressor Shootout! - GSSL, Waves SSL, Cytomic 'The Glue'

Which do you generally prefer?

  • Gyraf SSL Compressor

    Votes: 3 13.0%
  • Waves SSL Compressor

    Votes: 12 52.2%
  • Cytomic 'The Glue' Compressor

    Votes: 8 34.8%

  • Total voters
    23

Ermz

¯\(°_o)/¯
Apr 5, 2002
20,370
32
38
37
Melbourne, Australia
www.myspace.com
In light of the big focus shift onto the quality of bus compressors in recent times, I figured it may be a good idea to do a shoot-out of some of today's big contenders. I focused the shoot-out mainly toward those aiming for the classic 'SSL' sound.

The units in the running today consist of two plug-ins and one hardware unit:

Gyraf SSL Compressor: A hardware SSL 4000 series bus compressor clone. I'm not exactly sure which model, whether it be E or G series. This particular one has the turbo kit board, which was engaged at all times during the testing. It essentially alters the side-chain so that the compressor behaves more like the classic SSL.

Waves SSL Compressor: This is a plug-in unit that has become many an engineer's go-to weapon of choice for software bus compression. I've used this one on just about every one of my mixes in the last few years. It is an emulation of the SSL G-series bus compressors.

Cytomic 'The Glue' Compressor: This is a relatively new plug-in, created by a lone developer. I'm unsure which model this one sets out to emulate, but it goes beyond the options of the hardware and offers more tweaking potential via a 'range' knob and side-chain filter.

Now it should go without saying that all these units respond and meter in different ways. As such I established a reference point and all the parameters that I could keep identical between them I did, but others had to be adjusted by ear in order to match the little operational nuances between them. For instance the Waves SSL would start pumping audibly at around 3dB GR, whereas 'The Glue' would be happy to go beyond 5dB and still remain relatively transparent. As such they were all adjusted to provide the same audible level of gain reduction, rather than metered. If you had to estimate, then 3 to 4dB of GR would be what's going on.

There are two track excerpts provided to give you a wider insight into how they respond to different programme material. One is a clean, ambient progressive rock track and the other is a fairly straight, punchy rock track. They are not 'mastered' very hot in order to retain most of the transient punch that the compressors emphasize.

Prog Track Settings:

Attack: 30ms
Release: 100ms
Ratio: 2:1
Gain Reduction: Adjusted to provide roughly 3 to 4dB, taking into account each compressor's likelihood to compress differently.

Rock Track Settings:

Attack: 10ms
Release: 100ms
Ratio: 2:1
Gain Reduction: Adjusted to provide roughly 3 to 4dB, taking into account each compressor's likelihood to compress differently.

Download links:

Prog track

Rock track
 
Interesting.
I think it's hard to make a judgement based on the prog track because it sounds like there's already some kind of odd compression on the snare on the raw track.
On the rock track, I prefered the Waves comp. More snap.
---Edit--- oops you said it first...
Btw, I tried the Glue and wasn't impressed. It sounded too clean for my taste.
 
The raw tracks are actually included in .wav format, so feel free to do that if you want!

here's Smart C2 in stereomode (I'm usually using dual-mono but since all the other comps are stereo this makes more sense).
there's one with and one without the external sidechain cables (HP at 150 I think to preserve the bass).
I also included the Digidesign Impact (plugin) that seems to be their take at the SSL buscomp.

I can't guarantee for EXACTLY the same gain-reduction, but the meter on the Smart was peaking around 4dB (perhaps in cases a cunthair more).
same for the Impact.

http://lsd-tonstudio.de/sonstiges/buscomp.zip

some more headroom on all the tracks would have been good though.I hope I didn't clip anything on the way in or out....so close to 0dBFs the converters' influences are bigger.

I think with the given material the differences are rather small but I think the "glue" loses for me. the digi makes it a bit too thick and dull for my taste, but that's also because I'd usually use it with less reduction than I'd use on the others.
the Waves SSL sounds pretty cool, perhaps a bit less "open" than the Gyraf and Smart.
all the differences are VERY subtle though.
 
I tend to agree. The Glue is a bit too transparent and subtle.. in fact I think it's the one least likely to 'glue' anything. It has a lot of transient punch, but that's it. The Waves SSL tends to pump a lot and sort of 'shrink' the mix, but it has a particular character to it, in particular the transient 'snap' it has. IMO the GSSL combines the best qualities there and ends up sounding more open to boot, so I'm quite glad I had it made. The meter is nigh useless though... really weird behaviour. I have to operate it almost entirely by ear.

I'll check out your C2 runs, Lasse.. thanks for doing that.
 
The difference is subtle indeed but once again the hardware unit sounds best. Less muffled sounding while still having the character.
 
I liked the Glue a lot when I demo'd it. I couldn't make it distort! I thought it sounded great.
On my last few mixes I've been using Stillwell The Rocket on the mixbus.

My preference (without listening) is The Glue, it's the only one I can afford!
 
on initial listen. There's not a huge difference between all of them. the only major discernable difference I can tell if between the glue and the gssl. The transients seem alittle more tamed in the gssl, in a good way. I think more extreme settings would show the differences better.

Honestly i think the raw didn't even sound that different than the processed ones. I'm, pretty familiar with real SSL bus comps and the wave's ssl both of which are more obivous to tell apart in the times i've used them.