Compression, along with EQ , is one of the two most important processing concepts to understand in audio engineering and music production. It has zillion different application reasons, but as one of the fundamental processors that deal with the dynamics of the audio signal, it is important to understand the basic. Lets simplify compression and it's parameters, in the most basic way. Compression is the process of attenuating audio signal. In its essence, it is "just" automated volume control. The purpose of the compressor is to reduce the dynamic range of the incoming singnal and attenuate the portion of the signal that is above the set treshold. Compressor attenuates audio signal when it rises above the set treshold. So, all the content above treshold gets attenuated. Treshold is a negative value and is expressed in dB. With the treshold of 0, compressor will not attenuate the signal. Ratio does NOT control WHEN the compressor "kicks in". Ratio is a parameter that sets the AMOUNT of compression applied to the signal above the treshold level. This control simply reflects the ratio between compressor's input and output. With the ratio of 1:1, no compression will occur, ever if the treshold is set to its minimum value. With the ratio of 2:1, for every 2 dB of audio above the treshold, the compressor will "let" only 1 dB of audio go through. High ratio values lead to extreme results and are very close to limiting (but in order not o mix terms, lets stop here and not go further into differences between compressors and limiters). On most compressors, ratio control has pre-defined values. Attack - this is a parameter that controls how long does it take for the compressor to reach full compression - gain reduction set by treshold and ratio. The value of attack is expressed in milliseconds. Release - this is a parameter that controls how long does it take for the compressor to stop compressing the signal. The value of attack is expressed in milliseconds. On some compressors there will be a Hold parameter, that controls how long full compression is applied to the signal, before gain reduction stops. However, attack and release can be set like so that on some compressors, hold function is not needed. Make up gain - as we said before, compressor attenuates the incoming signal, and the difference in input and output is our compression ratio. With the make up gain, you can compensate for this attenuation - the signal that comes out of the compressor will have reduced dynamic range, and make up gain allows you to bring the overall signal volume back to the initial value. As for the SEQ -side chain eq, it is important to understand what side chain is and also the concept of multiband compression. To understand side chain you need to know how compressor works. Signal enters the compressor, and (extremely quickly) compressor detects if the signal is above treshold. There are many different types of compressors and in technical terms, this "detection" and gain reduction is done in a different ways, but think of it like this - it is like you have your hand on a fader that controls the amplitude of the signal. If you hear that signal is too loud, you would pull the fader down, and then put it back up, when you feel signal is not loud anymore. Compressor literally does that instead of you - it analyses the incoming signal, and based on its settings, applies gain reduction. Side chain exists in any compressor - signal inputs the compressor, and in the compressor, it gets copied. One copy of the signal goes forward through processor, while another is sent to one of the analysing components. This creates the side chain. The copy is it is routed to a component, analysed and then the main signal is processed or not based on the dynamics characteristics of the copy that has been analysed and compared to the compressor settings . That means that input signal is the reference for the compressor to decide should it apply the gain reduction or not. Now, what happens if instead of the original signal copy, side chain input is fed with another signal? That will trigger the compressor to attenuate the incoming signal based on the dynamic behaviour of the signal coming into the side chain. This allows for what is commonly known as a side - chain compression - the original signal is compressed based on the characteristics and behaviour of another signal. for instance, if you apply side chain compression on the bass, and set your kick to be the side chain reference, each time the kick strikes, the compressor on the bass will attenuate the bass signal. It is important to know that ALL compressors form some sort of the side chain (remember, it is just a copy of the original signal), but not all of them can receive external input into the side chain. Most of the software and hardware compressors now days do have the side chain input function. AR compressor has four side chain modes: - OFF - side chain signal is the copy of the main signal, and therefore, compressor reacts by analysing the original signal, applying compression to the main audio evenly. - LPF - this means that the reference copy of the signal will first go through eq that will boost the low end of the signal. This means that when the signal is analysed, compressor will apply more gain reduction to the low end of the signal. -HPF - his means that the reference copy of the signal will first go through eq that will boost the high end of the signal. This means that when the signal is analysed, compressor will apply more gain reduction to the high end of the signal. - HIT - reference copy of the signal will be boosted both in high end and low end, with some low shelving and high shelving applied. This means signal will be compressed fairly evenly across the frequency spectrum. Mix - mixes compressed and dry signal. Basically a control that tells the compressor how much of the incoming signal to process, and how much to let unprocessed. Now, while AR architecture sends signal into compressor in series, with the mix button, you can blend two signals. That means that you can control how much processed and unprocessed signal will go through. this is extremely powerful, as you can apply parallel compression right there within the unit.
@@Ponchy not a problem:) I used to teach audio on an institute and it is extremely important topic. You know what kind of mess people make later on in their producer and engineer careers if they don’t grasp the concept now.
@@earlsfield that's awesome! now I know who to bounce ideas with! and yes I know exactly what you mean. There are no real presets for compression in every situation so you have to know how to set them manually.
Everything is nice, but not sure about your description of SC filters. Using them will not make any compressor multiband, it still will be one band compression, and it will be applied through the whole spectre.
Thank you Ponchy. Just a quick Elektron tip: when you compare eg wet and dry with a knob, if you hold function the value jumps faster. Anyway, great tutorials!
AR architecture places compressor as the last in the main signal chain, before main output. Any signal that goes through the main signal path will pass through distortion and the compressor - they are the two master output effects. Also, external input is also routed to the main signal path - meaning external instruments will also be compressed. However, in the signal path settings it is possible to exclude individual tracks from the main signal path, as well to re-route effects outputs. For this you can use either individual outputs (physical) or you can use Overbridge for multi track control. However, it is worth knowing that ANY signal that goes through the main signal path goes through master effects, distortion and compressor.
No. I believe the multi tracks, the vst plugin outputs are dry. But there is one post fx channel, which is the master. I do sometimes mute all the pads except one, to get a processed audio track. But it's not ideal
If my Octatrack was going into my Rytm I would set the SEQ to either OFF or HIT depending what exactly you were playing into the Rytm. I could also see it working well at LPF, if you have a lot of bass heavy stuff going on but I would start at HIT. And then probably just a low ratio like 2:1 or 4:1 and then set your threshold to where you start to hear a reduction in volume and then set the MUP by going back and forth with the MIX setting like I show in the video.
Thanks Ponchy, please do an entire tutorial for analog 4 and analog rytm.. like a true manual from basics to the advanced functions. I Will pay you for that. Thanks
@@oliviermissal8739 LOL my bad I was thinking about the A4. There is actually no way to control the levels of the audio coming in so that has to be done externally. Thank you for pointing that out
@@Ponchy it should not be so hard to implement… keeping finger crossed… every latest DT updates would also definitely put the AR at the top of the line regarding competitors if it’s not already the case… inputs management for instance, at least for gain staging and even better for Fx processing :)… well we’ll see…
Compression, along with EQ , is one of the two most important processing concepts to understand in audio engineering and music production.
It has zillion different application reasons, but as one of the fundamental processors that deal with the dynamics of the audio signal, it is important to understand the basic. Lets simplify compression and it's parameters, in the most basic way.
Compression is the process of attenuating audio signal. In its essence, it is "just" automated volume control. The purpose of the compressor is to reduce the dynamic range of the incoming singnal and attenuate the portion of the signal that is above the set treshold.
Compressor attenuates audio signal when it rises above the set treshold. So, all the content above treshold gets attenuated. Treshold is a negative value and is expressed in dB. With the treshold of 0, compressor will not attenuate the signal.
Ratio does NOT control WHEN the compressor "kicks in". Ratio is a parameter that sets the AMOUNT of compression applied to the signal above the treshold level. This control simply reflects the ratio between compressor's input and output. With the ratio of 1:1, no compression will occur, ever if the treshold is set to its minimum value. With the ratio of 2:1, for every 2 dB of audio above the treshold, the compressor will "let" only 1 dB of audio go through. High ratio values lead to extreme results and are very close to limiting (but in order not o mix terms, lets stop here and not go further into differences between compressors and limiters). On most compressors, ratio control has pre-defined values.
Attack - this is a parameter that controls how long does it take for the compressor to reach full compression - gain reduction set by treshold and ratio. The value of attack is expressed in milliseconds.
Release - this is a parameter that controls how long does it take for the compressor to stop compressing the signal. The value of attack is expressed in milliseconds.
On some compressors there will be a Hold parameter, that controls how long full compression is applied to the signal, before gain reduction stops. However, attack and release can be set like so that on some compressors, hold function is not needed.
Make up gain - as we said before, compressor attenuates the incoming signal, and the difference in input and output is our compression ratio.
With the make up gain, you can compensate for this attenuation - the signal that comes out of the compressor will have reduced dynamic range, and make up gain allows you to bring the overall signal volume back to the initial value.
As for the SEQ -side chain eq, it is important to understand what side chain is and also the concept of multiband compression. To understand side chain you need to know how compressor works. Signal enters the compressor, and (extremely quickly) compressor detects if the signal is above treshold. There are many different types of compressors and in technical terms, this "detection" and gain reduction is done in a different ways, but think of it like this - it is like you have your hand on a fader that controls the amplitude of the signal. If you hear that signal is too loud, you would pull the fader down, and then put it back up, when you feel signal is not loud anymore. Compressor literally does that instead of you - it analyses the incoming signal, and based on its settings, applies gain reduction. Side chain exists in any compressor - signal inputs the compressor, and in the compressor, it gets copied. One copy of the signal goes forward through processor, while another is sent to one of the analysing components. This creates the side chain. The copy is it is routed to a component, analysed and then the main signal is processed or not based on the dynamics characteristics of the copy that has been analysed and compared to the compressor settings . That means that input signal is the reference for the compressor to decide should it apply the gain reduction or not.
Now, what happens if instead of the original signal copy, side chain input is fed with another signal? That will trigger the compressor to attenuate the incoming signal based on the dynamic behaviour of the signal coming into the side chain. This allows for what is commonly known as a side - chain compression - the original signal is compressed based on the characteristics and behaviour of another signal. for instance, if you apply side chain compression on the bass, and set your kick to be the side chain reference, each time the kick strikes, the compressor on the bass will attenuate the bass signal. It is important to know that ALL compressors form some sort of the side chain (remember, it is just a copy of the original signal), but not all of them can receive external input into the side chain. Most of the software and hardware compressors now days do have the side chain input function.
AR compressor has four side chain modes:
- OFF - side chain signal is the copy of the main signal, and therefore, compressor reacts by analysing the original signal, applying compression to the main audio evenly.
- LPF - this means that the reference copy of the signal will first go through eq that will boost the low end of the signal. This means that when the signal is analysed, compressor will apply more gain reduction to the low end of the signal.
-HPF - his means that the reference copy of the signal will first go through eq that will boost the high end of the signal. This means that when the signal is analysed, compressor will apply more gain reduction to the high end of the signal.
- HIT - reference copy of the signal will be boosted both in high end and low end, with some low shelving and high shelving applied. This means signal will be compressed fairly evenly across the frequency spectrum.
Mix - mixes compressed and dry signal. Basically a control that tells the compressor how much of the incoming signal to process, and how much to let unprocessed.
Now, while AR architecture sends signal into compressor in series, with the mix button, you can blend two signals. That means that you can control how much processed and unprocessed signal will go through. this is extremely powerful, as you can apply parallel compression right there within the unit.
Thank you for taking the time to write out such a great description of compression! I know this will surely help a lot of people.
@@Ponchy not a problem:) I used to teach audio on an institute and it is extremely important topic. You know what kind of mess people make later on in their producer and engineer careers if they don’t grasp the concept now.
@@earlsfield that's awesome! now I know who to bounce ideas with! and yes I know exactly what you mean. There are no real presets for compression in every situation so you have to know how to set them manually.
love this, thanks a bunch!
Everything is nice, but not sure about your description of SC filters. Using them will not make any compressor multiband, it still will be one band compression, and it will be applied through the whole spectre.
this seriously unlocked so much for me. appreciate this video a ton!
Appreciate you saying that!
Merci 🙏 une très bonne explication de la compression parallèle 😊
Thank you Ponchy. Just a quick Elektron tip: when you compare eg wet and dry with a knob, if you hold function the value jumps faster. Anyway, great tutorials!
hey I appreciate that! In my defense, this vid was a long time ago before I had really started taking advantage of the FUNC scrolling haha.
hello Ponchy, thanks for this video and others on Analog RYTM, very informative, you must be liking the V1.70 firmware update?
Thanks! Super helpful!
Glad to help!
Amazing video, thank you!
No prob!
Hey buddy. Do you have a video on the performance knob and how to set it up?
Thanks
Hey man yes indeed! Let me know if this helps
th-cam.com/video/HHrsohQX5I0/w-d-xo.html
Thanks for the helpful video. Does the compression effect each kit individually or the whole project?
AR architecture places compressor as the last in the main signal chain, before main output. Any signal that goes through the main signal path will pass through distortion and the compressor - they are the two master output effects.
Also, external input is also routed to the main signal path - meaning external instruments will also be compressed.
However, in the signal path settings it is possible to exclude individual tracks from the main signal path, as well to re-route effects outputs. For this you can use either individual outputs (physical) or you can use Overbridge for multi track control. However, it is worth knowing that ANY signal that goes through the main signal path goes through master effects, distortion and compressor.
@@auralplex Correct, my bad.
Thank you!
Glad to help!
Can this be recorded via overbridge to individual tracks?
No. I believe the multi tracks, the vst plugin outputs are dry. But there is one post fx channel, which is the master.
I do sometimes mute all the pads except one, to get a processed audio track. But it's not ideal
How would you set compressor if you had Octatrack audio coming into it?
If my Octatrack was going into my Rytm I would set the SEQ to either OFF or HIT depending what exactly you were playing into the Rytm. I could also see it working well at LPF, if you have a lot of bass heavy stuff going on but I would start at HIT. And then probably just a low ratio like 2:1 or 4:1 and then set your threshold to where you start to hear a reduction in volume and then set the MUP by going back and forth with the MIX setting like I show in the video.
is there a SEQ setting so that the kick pops out more?
Yeah if you do HPF it will compress the high frequencies thereby leaving the low end to appear as if it’s coming out more.
Thanks Ponchy, please do an entire tutorial for analog 4 and analog rytm.. like a true manual from basics to the advanced functions. I Will pay you for that. Thanks
Will do! Ill be doing separate courses for those so which one would you rather see first?
@@Ponchy Analog Four MK2, and then Analog Rytm MK2. thank you so much!
How do I adjust incoming audio for the the two incoming audio lines? To get them louder?
Press FX then goto Osc1 and it’ll be there.
where the hell is osc1?
@@oliviermissal8739 LOL my bad I was thinking about the A4. There is actually no way to control the levels of the audio coming in so that has to be done externally. Thank you for pointing that out
Thanks!~
No prob!
AR misses the attenuation indicator the DT comp has to the right…
I know right?! I hope they add that in a future update because it makes it soooo much easier to judge how much gain reduction we are getting.
@@Ponchy it should not be so hard to implement… keeping finger crossed… every latest DT updates would also definitely put the AR at the top of the line regarding competitors if it’s not already the case… inputs management for instance, at least for gain staging and even better for Fx processing :)… well we’ll see…
yeah very weird, especially since the MK1 RYTM had this
@@tendingtropic7778 really?