Compressors! why do they sound different? example! u-he etc.

[Caine123]

oh here was a test, i tested all 4 comps and i dunno why they sound different.... any ideas?

1. Fruity Compressor (the original setting and sound i like with its setting)



2. Nomad Factory Channel Strip (i had to set the gain to around -4 db ( 9.5 dB) compared to the fruity one to get nearly the same output!)



3. Renn Compressor (it sounds TOTALLY different....)



4. U-He Presswerk (also very different sounding !)



and here is my file !
https://copy.com/ogVxQIXT1LAFjmvk

0 - 8 sec NO Compression

8 - 16 sec Fruity Comp

16 - 24 sec Nomad Factory

24 - 32 sec RennComp

32 - 50 sec U-He Presswerl

i am confused

[jrides]

They look different too. Weird.

[valerian_777]

They look different too. Weird.

[bmanic]

There are so many reasons why they all sound different.. but some of the most obvious are:

1) Transfer function (the general "path" the input to output transformation follows)

2) time constants.. the shape of the attack and release (mainly the shape of the release) can have a huge impact on how the compressor behaves

3) side-chain processing / compression mode. Is it peak, RMS, average or some other weird combination of both or all 3? Is the frequency response of the side-chain really flat (in many vintage units this isn't the case at all)? This can have a huge effect on how the compressor is "voiced"

4) Topology. Feedback or Feed forward (or a mix of the two). Again, this can have a huge impact

.. this is just scratching the surface though. Compressors can be ridiculously complex when you start to really dig into them. I hope one of the TDR dudes chimes in so that we can perhaps a get more thorough and easier to read explanation for the various things that make up a complex compressor.

For now I suggest just testing a bunch of them (many good ones are available for free!) and each ones behaviour. Experiment and explore. You'll quickly notice that some are more "utilitarian" (good at doing very specific tasks) and some are more "tone boxes" (for more radical shaping and colouring).

[FabienTDR]

Seemingly trivial properties such as dynamic range, crest factor or loudness aren't well defined at all and represent rather crude ideas. Sadly, compressor's have the job of controlling these wishi washi things, so it's no wonder they all look, act and sound different. This is one of countless factors, but probably the most essential.

[Caine123]

thanks a lot for all the replies and i have to admit that i didnt understand everything you said here cause i still dont master a compressor . i just love the Renncompressor for example cause it shows so easily when it is compressing and trying settings our with it is GREAT!

BUT

with this example i heard that it doesnt seem to be a NEUTRAL compressor, especially not for the masterbus? where i have it
and i also wanted to test them each other cause i wanted to consider getting presswerk but my knowledge isnt well enough for it so i guess imma pass.

[hibidy]

That's just it. It's job specific and quite frankly it's tough keeping up with which one/go-to/etc.

Lately I'm in love with slate VMR. I find it the very best on a drum bus....by quite a bit. But that doesn't mean I'm throwing out vari comp (at least not yet )

For bass, I don't like the VMR that much. I find it too "smoochy" (either one) and much prefer DBX160 which in the NI/softube colab. For guitars I like VMR, but I probably will either use the NI/softube LA-2A on a bus which seems to provide what I'm looking for most of the time. BUT if I want a more pumpy "effect" of compression on a guitar track, I might go back to VMR or maybe some other 1176.



They have the same principle, but your ears have to be the final judge. You can read up on comps all you want (which is good) but most of them (other than stock DAW ones) behave with their own algos and such for varying results

[Synthetic Wav]

Many plugs comp can be made to sound very close, many are not all that much different in the end, it's like plugin eq to me, many can be made to sound close, juts go by ears and not settings with flexible plugins and you may be surprised ,specially with a flexible saturator like sdrr.

with plugin like tdr and dc8c i have matched compressors ( or even surpass these at what they are great thanks to more flexiblity ) even if these don't have much in common .. like tdr and api 2500 plug

[DaveGamble]

1) Transfer function (the general "path" the input to output transformation follows)

Sorry to be a pedant, but that's not really what that phrase means.

"Transfer function" as an expression, as bandied about liberally in audio refers to a mathematical description of a linear system (such as an EQ, filter, reverb, delay, etc).

You could talk about "signal path", which would be entirely meaningful, or "compression curve", which would again make sense. You /could/ talk about the transfer function of a compressor, but it would refer to any EQ that's an unavoidable part of the audio path, and excludes any dynamic description of the system. But since what's integral to the concept of a compressor is a dynamic gain-change stage, it's not really the same thing as your description.

Everything else you said was spot-on.

Cheers!

Dave.

[aMUSEd]

Is that Nomad Factory channel strip any good? It's dead cheap right now.

[bmanic]

1) Transfer function (the general "path" the input to output transformation follows)

Sorry to be a pedant, but that's not really what that phrase means.

"Transfer function" as an expression, as bandied about liberally in audio refers to a mathematical description of a linear system (such as an EQ, filter, reverb, delay, etc).

You could talk about "signal path", which would be entirely meaningful, or "compression curve", which would again make sense. You /could/ talk about the transfer function of a compressor, but it would refer to any EQ that's an unavoidable part of the audio path, and excludes any dynamic description of the system. But since what's integral to the concept of a compressor is a dynamic gain-change stage, it's not really the same thing as your description.

Everything else you said was spot-on.

Cheers!

Dave.

Thanks for correcting me Dave. I truly thought a basic waveshaper could be called a transfer function but of course that is quite literally completely wrong if it describes a linear system!

So what do I call the shape of a waveshaper then? I basically meant to describe this:



I find these "input/output transformation curves" very handy because even a novice can understand what is going on (well my image sucks as it doesn't say which axis is input and which one is output but it doesn't take a lot of brain processing ability to figure it out).

Baffled that it's not called a transfer curve. Just shows my ignorance when it comes to technical lingo.

Anybody got a better word for this?

EDIT: damn.. just read the wikipedia article on transfer function. Yeah, so that's out the window then. Heck, even transfer curve is out the window and I've seen that used in Sound on Sound and other publications. Will have to dig out my Mastering Audio (by Bob Katz) and see what he calls it.

EDIT2: Found it.. the correct word is Signal Transfer Function, no?

EDIT3: Hmm.. Wikipedia on Distortion. In this article they also use transfer function but mention it is just an approximation in complex non-linear systems. I guess it is okay to use transfer function as a general term for my image above.

[crashedthecar]

Thanks Dave and b manic that cleared up a few things for me as well

[werp]

Is that Nomad Factory channel strip any good? It's dead cheap right now.

I think it rated well in 2006. but what the hell, it's cheaper and better for your health than a pack of cigarettes..

[FabienTDR]

@Dave, are you sure? "Transfer function" is a very common term in control systems theory.

http://en.wikipedia.org/wiki/Closed-loo ... r_function

This extremely well developed area is surprisingly low valued by the audio community, though (now idea why).

Here's a very nice intro (definitely check out this channel, good stuff).

https://www.youtube.com/watch?v=O-OqgFE9SD4