Bus vs. channel compressor

[antic604]

So assuming that the compressor has just the typical controls (threshold, ratio, attack, release, makeup gain, maybe look-ahead) what differentiates bus compressor from channel compressor?

For example, what's the *functional* difference between Waves' G-Master Bus Compressor and the dynamics section in Waves' G-Channel. BTW, I'm not looking for you explaining to me that one has "analog" button and the other doesn't, that one has a dial for attack while the other has only fast/slow switch or that the scales or ranges on the knobs are different. I'm asking for actual differences in how they WORK, assuming they'd have the same controls.

[simon.a.billington]

There’s not a lot real functional difference aside from many or them being more “transparent” sounding, but that’s not a given and that they need to be capable of ratios under 2:1 and most likely have a built in sidechain. The SPL Iron and Elysia Alpha are good examples of this, as they were designed for mastering which is basically a bus compressor.

On a submix it’s always handy if they offer a parallel mix option, but you can always set up one if you need one. The same for a sidechain.

If we’re lookint at the differences between the G Master Buss Compressor and what you find in a G channel strip, they are very similar. In the real world, though, SSL would have had to shrink the design drastically to fit it into a channel strip, as there is much less space. So they would have found smaller components, as close to the original design as they can and they would have had to strip out some of th features. So in the end you have a less flexible, less refined compressor, but it works well and you have one for every channel. As opposed to th buss compressor where you have very few of those that you have to us selectively, but it’s much better designed.

It’s the reason why sometimes, traditionally, some engineers would patch in an external co press or and use it on sounds that needed slightly better treatment, like vocals for example.

In the end they are just compressors so as long as they do the job you need them to then you can use anything.

[jochicago]

Difference: Like with most anything else with compressors, the difference is their character.
Most of the time the controls are the same: threshold, attack, release, etc. But they can sound vastly different and there's plenty going on under the hood. A lot of the stock comps that come with DAWs flat out don't sound that great when compared to the typical paid plugs we all reach for.

The Waves SSL bus comp is called that because its character most naturally suits the bus. I don't have that plugin so I can't comment, but that's the general gist of compressors in general: forget the knobs, what does it do to the audio going through it? Then you end up with things like: 1176 is great for rock vocals, etc. Why? We can talk about detection mechanics, attack and release knees, shape of the release curve and release tail, any coloring they impart (particularly the ones that model the analog units), etc. But ultimately it can get pretty simple: does it sound good, do you like it?

Here's an experiment that I find pretty accurate. Do some research and read in the forums about people actually using these things. When you are looking at a good compressor (good as in it sounds good) you'll find plenty of people mentioning it in conversation "what's you master bus comp", "what's on your master bus" and the like. Typical names come up and not just because they are fashionable: MJUC, SSL Bus Comp or Duende, etc. I promise you, almost never someone says "reacomp" and means it (not to attack Reaper, I have issues with all the DAW comps I've tried).

[Tj Shredder]

Beside taste driven comments, ist a way to sell the same product twice...
Technically a compressor is a compressor is a compressor. If you build it for a hardware mixer, of course there will be a difference also for real estate reasons...
Now if you jump on the analog hype, you can model them differently, but the basic components would be the same...
The channel comp requires a side chain, on the bus you could live without. As its meant for mastering a mix... But why not incorporate a side chain if its there?
If a master comp has a side chain input you get probably more controls to tweak...
You could apply the same character to channels or mixes or choose a different one. Your taste should rule there anyway...

Think about the reason you need the compressor for. It could be a mixing aid, to reduce a signal if a kick kicks in, it could be a sound design reason to get some grit into the sound, it could be that you want to reduce dynamics...
These reasons are fundamentally different and a specific compressor can be aimed at a specific task. The underlying principle though remains the same...

[jupiter8]

This particular instance is fairly simple to explain: they're models of the SSL 4000 compressors. It had a compressor on each channel and one masterbus compressor. They were similar but not identical i think though i'm not sure about the difference.

[kmonkey]

Basically a lot of talking about nothing and even not being sure about that same nothing.

So in short - nothing to see here

[SoundPorn]

It has to do with buss compressors on hardware were 2 channel stereo and usually had a softer knee while the channel comps were mono.
In software there's still a difference between the stereo linkage of better bus compressors and also the HPF. A buss compressor is actually the one that should have the HPF because on you're 2 buss you don't want it overly reacting to the bass. And what alot o people don't know is that the HPF actually have to be good linear phase eqs or else it will all fall apart.

So not a compressor is not a compressor they are actually engineered extremely different for different purposes allthough the line is getting continually blurred as we just build on top of those whose built before us.

[jochicago]

Technically a compressor is a compressor is a compressor.

Not in music-making, friend. They are absolutely not the same.

Slap an IK Black 76 and a Tracktion ("premium") Bus Comp on a vocal. One is going to gently roll the spikes, add warmth, control the plosive (and possibly the sibilance) and add some mojo. The other is going to deliver mediocre uncomfortably-sounding compression that will need 3 plugins right after to get something out of that track.

They are 100% not the same. All comps listen differently, they grip differently, different knees, different interpretation of the GR depth, VERY different reduction curves, and almost every single one of them has its own signature interpretation of what "attack" and "release" times mean. Using testing tools I've seen a comp CLAIM 250ms attack and actually grip within 40ms. Even comps that I consider to be very good-sounding, will claim a 50ms attack and actually grip at 1ms.

And all of that is before even going into the "analog" modeling. Like a Fairchild ducking areas of the eq spectrum as you grip tighter. Or the fact that a lot of these analog units don't have attack and release knobs but instead they react to the sound and change their settings internally.

I'm sorry I don't have the Waves SSL Bus Comp to give you more details on that unit. But I promise you even without owning it there's just no way it doesn't have some unique characteristics that make it special. And even within Waves own code I don't think that SSL Bus Comp is the same as their SSL channel strip or any other version. I test with the H-Comp that has an SSL model on it, and I'm yet to find a situation in which I prefer it. I would most definitely not drop that on a master bus. Yet online I see mixers far better than me happily enjoying the SSL Bus Comp on the master. There's something there. It cannot possibly sound like older SSL channel models.

[simon.a.billington]

Technically a compressor is a compressor is a compressor.

Not in music-making, friend. They are absolutely not the same.

Slap an IK Black 76 and a Tracktion ("premium") Bus Comp on a vocal. One is going to gently roll the spikes, add warmth, control the plosive (and possibly the sibilance) and add some mojo. The other is going to deliver mediocre uncomfortably-sounding compression that will need 3 plugins right after to get something out of that track.

They are 100% not the same. All comps listen differently, they grip differently, different knees, different interpretation of the GR depth, VERY different reduction curves, and almost every single one of them has its own signature interpretation of what "attack" and "release" times mean. Using testing tools I've seen a comp CLAIM 250ms attack and actually grip within 40ms. Even comps that I consider to be very good-sounding, will claim a 50ms attack and actually grip at 1ms.

And all of that is before even going into the "analog" modeling. Like a Fairchild ducking areas of the eq spectrum as you grip tighter. Or the fact that a lot of these analog units don't have attack and release knobs but instead they react to the sound and change their settings internally.

I'm sorry I don't have the Waves SSL Bus Comp to give you more details on that unit. But I promise you even without owning it there's just no way it doesn't have some unique characteristics that make it special. And even within Waves own code I don't think that SSL Bus Comp is the same as their SSL channel strip or any other version. I test with the H-Comp that has an SSL model on it, and I'm yet to find a situation in which I prefer it. I would most definitely not drop that on a master bus. Yet online I see mixers far better than me happily enjoying the SSL Bus Comp on the master. There's something there. It cannot possibly sound like older SSL channel models.

If the H-Comp has a SSL model on it, I’d say it would be Analog setting 2. 1 seems very close to the API, 3 seems very close to the V-Como and 4 has too much harmonics for it to be an SSL. Possibly it’s an 1176 or a Fairchild.

The channel strips would indeed be different. I’m sure they are based on a similar design, but SSL had to squeeze a compressor into a much smaller space within the channel strip, so comprimimises would of had to be made. Such as a simpler interface, less options and smaller, albeit different components.

But hey, you had one on every track!! That was a big thing back in those days. It also would explain why it sounds different.

[Bouroki]

The channel comp was based on an RMS detector, dbx-style "over-easy" soft knee, and a program dependent attack. The bus comp is a peak compressor with fixed attack settings and a variable knee according to the ratio. Based on that I wouldn't say they shared a similar design at all.

[SoundPorn]

Yeah I love 4th setting it sounds like a Tube possibly a vari mu model, it adds amazing grit to vocals. Whatever 3 is modelled on makes my drums scream at me so possibly a Neve. 2 has a cleaner but distinct character so it may be the SSL. Whatever the H comp has become my favorite compressor I just love this thing it's so underrated.

And is it me or the punch knob, which i was told just delays the attack, actually seems to add gain and soft clip the signal before the attack. Because you think it would just add attack but it seems to be adding both attack and sustain to a signal before compression starts. Like the things this thing does to drums...just beautiful dark twisted things it does...

Technically a compressor is a compressor is a compressor.

Not in music-making, friend. They are absolutely not the same.

Slap an IK Black 76 and a Tracktion ("premium") Bus Comp on a vocal. One is going to gently roll the spikes, add warmth, control the plosive (and possibly the sibilance) and add some mojo. The other is going to deliver mediocre uncomfortably-sounding compression that will need 3 plugins right after to get something out of that track.

They are 100% not the same. All comps listen differently, they grip differently, different knees, different interpretation of the GR depth, VERY different reduction curves, and almost every single one of them has its own signature interpretation of what "attack" and "release" times mean. Using testing tools I've seen a comp CLAIM 250ms attack and actually grip within 40ms. Even comps that I consider to be very good-sounding, will claim a 50ms attack and actually grip at 1ms.

And all of that is before even going into the "analog" modeling. Like a Fairchild ducking areas of the eq spectrum as you grip tighter. Or the fact that a lot of these analog units don't have attack and release knobs but instead they react to the sound and change their settings internally.

I'm sorry I don't have the Waves SSL Bus Comp to give you more details on that unit. But I promise you even without owning it there's just no way it doesn't have some unique characteristics that make it special. And even within Waves own code I don't think that SSL Bus Comp is the same as their SSL channel strip or any other version. I test with the H-Comp that has an SSL model on it, and I'm yet to find a situation in which I prefer it. I would most definitely not drop that on a master bus. Yet online I see mixers far better than me happily enjoying the SSL Bus Comp on the master. There's something there. It cannot possibly sound like older SSL channel models.

If the H-Comp has a SSL model on it, I’d say it would be Analog setting 2. 1 seems very close to the API, 3 seems very close to the V-Como and 4 has too much harmonics for it to be an SSL. Possibly it’s an 1176 or a Fairchild.

The channel strips would indeed be different. I’m sure they are based on a similar design, but SSL had to squeeze a compressor into a much smaller space within the channel strip, so comprimimises would of had to be made. Such as a simpler interface, less options and smaller, albeit different components.

But hey, you had one on every track!! That was a big thing back in those days. It also would explain why it sounds different.

[simon.a.billington]

The channel comp was based on an RMS detector, dbx-style "over-easy" soft knee, and a program dependent attack. The bus comp is a peak compressor with fixed attack settings and a variable knee according to the ratio. Based on that I wouldn't say they shared a similar design at all.

Thanks for correcting me. I was guessing, postulating other reasons whe they may sound different.

Of course, knowing the facts would have helped!!

[simon.a.billington]

Yeah I love 4th setting it sounds like a Tube possibly a vari mu model, it adds amazing grit to vocals. Whatever 3 is modelled on makes my drums scream at me so possibly a Neve. 2 has a cleaner but distinct character so it may be the SSL. Whatever the H comp has become my favorite compressor I just love this thing it's so underrated.

And is it me or the punch knob, which i was told just delays the attack, actually seems to add gain and soft clip the signal before the attack. Because you think it would just add attack but it seems to be adding both attack and sustain to a signal before compression starts. Like the things this thing does to drums...just beautiful dark twisted things it does...

Technically a compressor is a compressor is a compressor.

Not in music-making, friend. They are absolutely not the same.

Slap an IK Black 76 and a Tracktion ("premium") Bus Comp on a vocal. One is going to gently roll the spikes, add warmth, control the plosive (and possibly the sibilance) and add some mojo. The other is going to deliver mediocre uncomfortably-sounding compression that will need 3 plugins right after to get something out of that track.

They are 100% not the same. All comps listen differently, they grip differently, different knees, different interpretation of the GR depth, VERY different reduction curves, and almost every single one of them has its own signature interpretation of what "attack" and "release" times mean. Using testing tools I've seen a comp CLAIM 250ms attack and actually grip within 40ms. Even comps that I consider to be very good-sounding, will claim a 50ms attack and actually grip at 1ms.

And all of that is before even going into the "analog" modeling. Like a Fairchild ducking areas of the eq spectrum as you grip tighter. Or the fact that a lot of these analog units don't have attack and release knobs but instead they react to the sound and change their settings internally.

I'm sorry I don't have the Waves SSL Bus Comp to give you more details on that unit. But I promise you even without owning it there's just no way it doesn't have some unique characteristics that make it special. And even within Waves own code I don't think that SSL Bus Comp is the same as their SSL channel strip or any other version. I test with the H-Comp that has an SSL model on it, and I'm yet to find a situation in which I prefer it. I would most definitely not drop that on a master bus. Yet online I see mixers far better than me happily enjoying the SSL Bus Comp on the master. There's something there. It cannot possibly sound like older SSL channel models.

If the H-Comp has a SSL model on it, I’d say it would be Analog setting 2. 1 seems very close to the API, 3 seems very close to the V-Como and 4 has too much harmonics for it to be an SSL. Possibly it’s an 1176 or a Fairchild.

The channel strips would indeed be different. I’m sure they are based on a similar design, but SSL had to squeeze a compressor into a much smaller space within the channel strip, so comprimimises would of had to be made. Such as a simpler interface, less options and smaller, albeit different components.

But hey, you had one on every track!! That was a big thing back in those days. It also would explain why it sounds different.

I agree. It’s a great sounding compressor.

[SoundPorn]

The easiest way to prove how drastically different compressors can act is to just download the MTurboComp trial and flip thru the presets that model different compressors. That will really make you understand and appreciate the differences while learning what they are. I did it and it gave me such a clearer picture of compression, it was like my DMT moment for compressors. Without the dmt.

[jochicago]

I believe the H-Comp #2 is inspired by SSL. I think the 3 and 4 are inspired by Neves (I have that thought in memory but can't confirm), which makes the #1 the API. I'm about 80% on that. 3 and 4 are pretty aggressive in their coloring.

It's an interesting compressor. Missed the boat a bit without a make-up gain, which is odd for the Hybrid line since the point was to give these more deep settings so they would be more flexible.

But the real problem is the compression behavior. All models behave roughly the same in terms of how they apply GR. They all have soft knee attack in roughly the same curve, and they all have a hard-knee release missing a gentle tail, in roughly the same shape. So when you change modes you are changing some analog coloring and even wacky behavior (looking at you #3), but not the gain-reduction behavior so much.

I think that's why people take sides with this compressor. If you like that design (soft-knee attack, hard release) then you are home with this thing. For people like me that prefer a soft-knee release with a long gentle tail (say an opto) you can never get it to "sound right" regardless of settings.

IMO definitely not an all-purpose bus-compressor in the sense that the release is too hard. I would say a bus compressor has to try to blend in. But the attack portion is very nicely done. This is in the minority of compressors that actually respects the attack knob speed, and grips with surprising elegance, so you can set it well to taste.