Help me to understand Fractal compressors

slashy

slashy

Inspired
Hi everybody.
Since 2 days, i try to set the comp (pedal 1, pedal 2 ans studio) to have the more transparency signal but in order to have a more sustain, especially on clean and crunch tone.
For example in my pro tools , when i mix( with Api, Cl-2A, etc..), the more important thing to me is to keep the transient with the comp.
The attack must be generally set to a point where we hear clearly the attack and transient.
For example for a sound who contain 100 hz, we set 10 milliseconds minimum.
And in the Axe, comp are very difficult to set for a transparency signal.
I don't understand why, Cliff said that we need fast attack.... ?
The Studio comp seem to me very colored , even with low ratio.Is only me ?
Pedal 2 seem to be better, but i don't understand the standard settings, why put the attack at only 2 ms ?
We loose transient !
Why put the release at 100ms ?, too long for the guitar !
And i don't hear that Lookahead help a lot.....

If somebody have some advice, i take them ! ;)
Have you the same problem than me ?
 
I would love to hear more about people's experiences digging in deeply to the permutations of attack and release times, compression ratio, and threshold. I am poking around in semi-darkness as I play with these and listen to the differences.

One thing that I heard from my own past lazy experiments, which seems counter-intuitive to me, is that increasing the threshold (which defaults to around -30db or -40db?) to something like -25 or -20 or -15, in the process of doing that, the sound seems to get louder. I'm confused about how those components actually work.

I have a vague notion it is something like this:
1. When the signal is below the threshold, there is no processing. When the signal rises above the threshold, it starts a timer.
2. When the start timer exceeds the attack setting, the signal is compressed according to the compression ratio. I don't understand exactly what are the values in the ratio- for example, does a 2:1 ratio mean a -6dB sound level drop, and 4:1 ratio means a -12dB sound level drop?
3. There must be a ramp-up period from no compression to the target compression amount- otherwise there would be a discontinuity in the signal that sounds like a crack or a pop. So that must be what the "hard knee" or "soft knee" is. I would guess the most "transparent" type of sound would come from a soft knee to minimize the changes in the waveform. But does that transition from zero to target compression level start as soon as the minimum threshold is exceeded, or does that start after the attack timer duration is reached?
4. When the signal drops below the threshold level, I assume that is when the release timer starts. Or is the release timer measured from the start of the attack? It would seem to be more flexible for different musical styles and speed of notes if the release timer is measured from when the signal drops below the threshold.
5. To create a more "rich" and "in your face full" type of sound, the signal level can be boosted before the compression analyzer- the idea is to give more voice to the quiet passages of music, the fingerprints scraping the strings, etc- but to do so in a way that the higher level signals can be squashed to avoid clipping and hopefully in a way that doesn't make those loud signals get too sonically altered in the process of squashing it down to fit inside the clipping limits.


If I'm on the right base so far, here are a few generalizations or my guesses at the theory of how we should set the compressors:
1. Decide your priority for a given sound:
a) Maximize the nuance and articulation of quiet passages, string scrapes with skin texture, etc...
b) Make everything louder and in your face
c) Tame transients, loud spikes associated with wah pedals, very dynamic inputs, etc.

2) To maximize the quiet nuances, with possibility of distorting the color of louder passages, use a very high threshold (e.g. -10dB or even closer to 0dB). Dial the compression ratio to alter the coloration of the louder parts. This should make the quiet passages pop out. Turn up the pre-compression level (either with another block or within the compression block) to enhance the effect, but with more coloration of the louder sections. Attack and release times: probably short/fast? Hard or soft knee to taste?

3) Make everything louder and in your face: jack up the level 6dB or 9dB, and use a low or medium compression ratio (e.g. 2:1 or up to 4:1) and set the threshold much lower (e.g. -40dB) so it doesn't kick on and off as much, to hopefully color the sound less. Use longer release times with a soft knee. Not sure about appropriate attack times? Either very short to catch all of the transient, or long enough to kick in after the transient? Don't want to kick in during the rise and decay of the transient.

4) To tame the transients, (assuming that the overall signal level is good but just the transients clip a little too much) do not boost the level, but do set the threshold very high (e.g. -6dB or -3dB or even -1dB ) and with a very high compression ratio (e.g. 10:1 or more) and shortest attack and release times possible.


This is the first time I've attempted to be more disciplined in thinking about compression parameters, I haven't done experiments to validate these concepts yet, so I would appreciate any studio wizards to chime in confirming, refuting, or refining any bits of this. Understanding these things in more detail is important for purposeful tweaking of these parameters, especially in a post-processing phase when looking at a waveform where you can measure a signal level, attack duration, etc. and then apply the concept of those compression tweaks to your live sound.
 
6stringscott said:
I would love to hear more about people's experiences digging in deeply to the permutations of attack and release times, compression ratio, and threshold. I am poking around in semi-darkness as I play with these and listen to the differences.
Click to expand...

My suggestion: don't use a compressor block.

Try to obtain the sustain and tone you want in just the Amp + Cab blocks. Unless you have a very very very specific need, I try to stay away from the Compressor blocks since the Amp blocks are technically already "compressed". Just personal opinion/approach mate.
 
Have you considered using a gate/expander block in place of the compressors?
From my experience what I often use a racked compressor for seems to fall into the gate/expander block more so than the compressor.
 
6stringscott said:
One thing that I heard from my own past lazy experiments, which seems counter-intuitive to me, is that increasing the threshold (which defaults to around -30db or -40db?) to something like -25 or -20 or -15, in the process of doing that, the sound seems to get louder. I'm confused about how those components actually work.
Click to expand...
you probably have automatic gain makeup on, it adds volume dependent on how much compression is going on (more compression > less volume). everything else is more or less spot on though; threshold sets the point at which the comp acts on the signal, attack sets how many ms before the compression kicks in once the threshold is reach, release sets how long the compressor acts (assuming youre no longer past the threshold, if youre sawing away and the signal stays past threshold the compressor will stay on the whole time and the release will take effect once your drop below it), knee sets how hard the comp comes in, pretty sure thats the gist of it.

as far as getting the amount of compression you want with just an amp and cab sometimes that just isnt possible, its why there are even compressors at all. learning how it works and when to utilize it is far more useful to you in the long run than completely forgoing it y'know?
 
Keep in mind that using a compressor pedal in front of an amp is a whole different thing / result than using a studio compressor on a recorded guitar track. I generally consider pre comp as a color effect, so I use it when I feel it's necessary (i.e. slide guitar). On the other hand, post comp plays a major role within the mix, so I leave that to the sound engineer as he might want to dose it differently depending on the mix
 
6stringscott said:
I have a vague notion it is something like this:
1. When the signal is below the threshold, there is no processing. When the signal rises above the threshold, it starts a timer.
2. When the start timer exceeds the attack setting, the signal is compressed according to the compression ratio. I don't understand exactly what are the values in the ratio- for example, does a 2:1 ratio mean a -6dB sound level drop, and 4:1 ratio means a -12dB sound level drop?
3. There must be a ramp-up period from no compression to the target compression amount- otherwise there would be a discontinuity in the signal that sounds like a crack or a pop. So that must be what the "hard knee" or "soft knee" is. I would guess the most "transparent" type of sound would come from a soft knee to minimize the changes in the waveform. But does that transition from zero to target compression level start as soon as the minimum threshold is exceeded, or does that start after the attack timer duration is reached?
4. When the signal drops below the threshold level, I assume that is when the release timer starts. Or is the release timer measured from the start of the attack? It would seem to be more flexible for different musical styles and speed of notes if the release timer is measured from when the signal drops below the threshold.
5. To create a more "rich" and "in your face full" type of sound, the signal level can be boosted before the compression analyzer- the idea is to give more voice to the quiet passages of music, the fingerprints scraping the strings, etc- but to do so in a way that the higher level signals can be squashed to avoid clipping and hopefully in a way that doesn't make those loud signals get too sonically altered in the process of squashing it down to fit inside the clipping limits.
Click to expand...

Some quick answers for you:

That "timer" is your release setting. The ratio is reduction per db over the threshold, higher ratios mean more drastic effects, lower ratios are for light taming. The ramp up period from the moment you cross the threshold to the beginning of compression is your attack setting. The release timer starts as soon as you break the threshold

Basically I think of compressors as a rubber band. Higher ratios and faster attack settings mean its a stiffer response and the comp is soaking a lot of the energy, and applying it's color. I don't use it much for guitar, but chicken pickers love that stuff because of how it transforms the transients and the mids. On things like drums, it's great for keeping a couple of exceptionally loud hits leveled out, and is more transparent as the threshold is raised.

For guitar, its pretty pleasing to lower the attack setting, and lengthen the release. The original click of the pick attack goes through untouched, and then the ringing is sustained.

Comps are usually driven by low end energy too btw, you can have good results cutting in front of it, and then adding it back in down the chain.

For the OP, I think the "comp in front" approach is meant to be heard. Try your luck somewhere after the amp block if subtlety is your goal. Slamming the front of the amp results in compression too without manhandling the transients like a comp pedal might do
 
edo said:
Keep in mind that using a compressor pedal in front of an amp is a whole different thing / result than using a studio compressor on a recorded guitar track. I generally consider pre comp as a color effect, so I use it when I feel it's necessary (i.e. slide guitar). On the other hand, post comp plays a major role within the mix, so I leave that to the sound engineer as he might want to dose it differently depending on the mix
Click to expand...
JCBitB said:
Some quick answers for you:

That "timer" is your release setting. The ratio is reduction per db over the threshold, higher ratios mean more drastic effects, lower ratios are for light taming. The ramp up period from the moment you cross the threshold to the beginning of compression is your attack setting. The release timer starts as soon as you break the threshold

Basically I think of compressors as a rubber band. Higher ratios and faster attack settings mean its a stiffer response and the comp is soaking a lot of the energy, and applying it's color. I don't use it much for guitar, but chicken pickers love that stuff because of how it transforms the transients and the mids. On things like drums, it's great for keeping a couple of exceptionally loud hits leveled out, and is more transparent as the threshold is raised.

For guitar, its pretty pleasing to lower the attack setting, and lengthen the release. The original click of the pick attack goes through untouched, and then the ringing is sustained.

Comps are usually driven by low end energy too btw, you can have good results cutting in front of it, and then adding it back in down the chain.

For the OP, I think the "comp in front" approach is meant to be heard. Try your luck somewhere after the amp block if subtlety is your goal. Slamming the front of the amp results in compression too without manhandling the transients like a comp pedal might do
Click to expand...

You said "its pretty pleasing to lower the attack setting, and lengthen the release" to keep the pick attack.So, my question is why ? because we need higher setting to the Attack(in comp) to let the "Attack of the pick"untouched .
 
JCBitB said:
For the OP, I think the "comp in front" approach is meant to be heard. Try your luck somewhere after the amp block if subtlety is your goal. Slamming the front of the amp results in compression too without manhandling the transients like a comp pedal might do
Click to expand...
Sorry, but i'm french, could you precise your advice please.I'm not sure to understand.
Thank you in advance .
 
slashy said:
Sorry, but i'm french, could you precise your advice please.I'm not sure to understand.
Thank you in advance .
Click to expand...
Where you place the compressor block in the change will affect the impact it has on your sound. Before the amp block will be much more noticeable and helps to even out volume differences across the strings, after the cab block is more subtle and smooths the tone out if there's anything peaky going on
 
Brownmatthall said:
Where you place the compressor block in the change will affect the impact it has on your sound. Before the amp block will be much more noticeable and helps to even out volume differences across the strings, after the cab block is more subtle and smooths the tone out if there's anything peaky going on
Click to expand...
Thank you ;)
I mix a lot of song, i use compressor all the time.
But effectively, i use it always in insert on a track.So after the cab (for the guitar, of course).....
For the first time, i try to use it before, and i had never be happy with it ;)
So, this afternoon, i'll try to use it after the cab, but probably i'll prefer use my old method who consist into insert a comp like API 2500 directly on the track.
 
slashy said:
Thank you ;)
I mix a lot of song, i use compressor all the time.
But effectively, i use it always in insert on a track.So after the cab (for the guitar, of course).....
For the first time, i try to use it before, and i had never be happy with it ;)
So, this afternoon, i'll try to use it after the cab, but probably i'll prefer use my old method who consist into insert a comp like API 2500 directly on the track.
Click to expand...

You are describing two very different uses of compression. Guitarists using compressors before an amp are using it as an effect, they often see the coloring as desirable, so I'm not surprised you're hearing them more -- especially the pedals that are modeled in the Axe (as opposed to the studio comp, which can be set more transparent).

If you're expecting to put a compressor in front of the amp block, and get the same results as when you use a rack-mount on a track post-mic, you're going to be disappointed.

Furthermore, if what you're looking to do is compress as part of the mixing stage as you've described above, I wouldn't attempt to do that with the Axe-Fx. I would do that in my DAW so I have more control. Recording a track that has post-mic compression baked-in would limit your ability to get the desired result in context of the final mix.
 
My first idea was , having more sustained note on clean and crunch tone but with a transparency comp.
And eventually raise the lowest levels into rhythm parts or having less dynamic between the lowest and highest levels.Exactly that a comp is suppose to do for example.
But i see that the comp into Axe are too colored for me.
Even the Studio Comp is colored to my ears.
I find eventually the Pedal 2 more transparency with a mix of 75%.
 
But again, as others have said, you're talking about totally different types of compressors/compression. Stop comparing to a plug in a DAW, not even close to the same thing
 
It seems to me like you want to limit the dynamic range between the highest levels (the peaks) and the lower levels (the sustain), but you are setting your attack to allow the peaks to come through uncompressed. That will never work, a compressor lowers the dynamic range by lowering the peaks (and then you can raise your sustain by raising the make-up gain). If you do not compress the peaks, the compressor is basically not doing anything.

You may be able to do what you want with an expander (but that will increase noise as well.
 
You must log in or register to reply here.
Back
Top Bottom