When do you use Compression and Limiters and why?

[Mistro17]

Hi. I'm no professional but even doing my own home bedroom studio music, I want to have good results in my final audio file. In my music making journey, I'm at the point where I learned enough music theory to feel comfortable picking up any virtual instrument. I think I get the basic rules of mixing BUT, I also acknowledge I don't really understand why I reach for certain things that go beyond playing the instruments, panning and playing with the EQ. I understand some tools are for subjective reasons in shaping sounds but some are essential for a good output for listeners across many types situations when listening?

One of my biggest issues I want to overcome now is getting louder music files. For example, my music often sounds great coming out of my speakers but when played in a custom radio station in Grand Theft Auto for example, it's super low compared to professionally mastered songs I have in the same folder. Same when listening on cheap earplugs from my phone. I know it have something to do with using compression and/or a limiter (probably more things?). But before I just start grabbing those tools, I really want to know what I'm doing and why. And what should be done before reaching for those 2 particular tools?

[huggermugger]

I use compression in two ways...

1) to manage occasional peaks that get too close to maximum (aka Peak compression). Managing those peaks makes it easier to push your levels without redlining. Peak compression is easiest when you have good metering in the compressor, because you often can't hear the results.

2) to gently squeeze the overall level of a track or the master so that louder areas are held back a little bit (aka RMS. compression). This gentler approach helps keep all the sections of a track a little more consistent in volume, which can create the sense of a louder track, since the quieter parts are going to get louder with gain compensation. Metering helps but now it's also up to your ears. Sometimes I skip around in the track with the compressor turned on so i can quickly compare what it's doing to, e.g., the chorus vs the bridge.

1) Peak compression involves fast attack and often fast release times, a threshold that's close to your actual "too loud" level, and a ratio of maybe 4:1 or higher.

2) RMS compression involves slower attack and release times, a threshold that's a little lower than your actual "too loud" level, and a ratio of maybe 3:1 or lower, and probably a little Gain Compensation to bring the output level up a little (which brings up the softer areas).

[Mistro17]

I use compression in two ways...

1) to manage occasional peaks that get too close to maximum (aka Peak compression). Managing those peaks makes it easier to push your levels without redlining. Peak compression is easiest when you have good metering in the compressor, because you often can't hear the results.

2) to gently squeeze the overall level of a track or the master so that louder areas are held back a little bit (aka RMS. compression). This gentler approach helps keep all the sections of a track a little more consistent in volume, which can create the sense of a louder track, since the quieter parts are going to get louder with gain compensation. Metering helps but now it's also up to your ears. Sometimes I skip around in the track with the compressor turned on so i can quickly compare what it's doing to, e.g., the chorus vs the bridge.

1) Peak compression involves fast attack and often fast release times, a threshold that's close to your actual "too loud" level, and a ratio of maybe 4:1 or higher.

2) RMS compression involves slower attack and release times, a threshold that's a little lower than your actual "too loud" level, and a ratio of maybe 3:1 or lower, and probably a little Gain Compensation to bring the output level up a little (which brings up the softer areas).

Thanks! When do you suspect a track is peaking or is this something that should be checked on each track? How important can it be in some cases to use compression while mixing for drums for example and what should be considered while using compression during mixing thinking ahead for the final mastering work or when can compression not be needed? Do you put a compressor plugin in the mastering part of Reason even though for example you may have the "Default Mastering Suit" installed in the session?

Sorry if I'm asking too loaded questions. I watched too many YouTube videos on this subject and too often I see compression and all kinds of audio tools being used but sometimes have a hard time figuring out if it's for personal taste, or a matter of good service. I guess I'm looking for a broad picture of the whole process with the nuance.

[Jac459]

I am no expert neither but one thing you need to be aware of is compression is basically killing your dynamic.
For many expert, the loudness war is finished as major streaming services are now equalising 'sort of' (you can Google LUFS).
So it is not like 15 years ago when the goal was to make the listener 'feell' a louder sound by squashing the sound like crazy, killing all the dynamic...

Now what you should at least do is side chaining your important elements (for example your bass should react to your kick to slightly reduce to avoid a muddy peak) .
You should also look at the global volume of your song to see if there are some unwanted peaks. Suppressing these peaks will drastically increase your general volume as you will be able to normalize way higher.
Then if you are into dance, you can also see how to get pumping effect by over compressing. But there are more sound design than mastering.... (and you don't need a compressor or limiter for that, synchronous has some presets for it).

Finally, pay attention to your EQ, if you use too much a specific frequency (generally bass) you will loose in overall loudness feeling. (this was the most complex for me, I had to buy a tool for that (izotope)).

[Steedus]

If you want loudness your best bet is a Maximizer. Reason has one built-in as part of the M-Class effects, but I'd really recommend keeping your eye out for a sale (or a freebie offer) for Izotope Ozone Elements. It's what I started with and it really is great for a beginner (like me). I got it for $29 but it's definitely been free at times.

As for checking levels etc, there's a very good free VST called Youlean Loudness Meter which is handy - as long as you don't get overwhelmed with all the numbers and graphs - as a beginner I certainly fell into that trap.

https://youlean.co/youlean-loudness-meter/

[Jac459]

If you want loudness your best bet is a Maximizer. Reason has one built-in as part of the M-Class effects, but I'd really recommend keeping your eye out for a sale (or a freebie offer) for Izotope Ozone Elements. It's what I started with and it really is great for a beginner (like me). I got it for $29 but it's definitely been free at times.

As for checking levels etc, there's a very good free VST called Youlean Loudness Meter which is handy - as long as you don't get overwhelmed with all the numbers and graphs - as a beginner I certainly fell into that trap.

https://youlean.co/youlean-loudness-meter/

Yeah, izotope is awesome. I ended up buying the full suit...
I am not very good so having the assistant and visual feedback helps a lot. And results are quite awesome.

Pros can certainly do without though...

[Popey]

I think huggermugger has ready covered how I use compression but one thing I find handy is using a scope to see my waveforms and the effect of compression. If you are trying to get loudness this helps to see any spikes that would have your compressor or final limiter working too hard. There is a free vst called smexoscope I use which is handy.

[Chizmata]

Hi. I'm no professional but even doing my own home bedroom studio music, I want to have good results in my final audio file. In my music making journey, I'm at the point where I learned enough music theory to feel comfortable picking up any virtual instrument. I think I get the basic rules of mixing BUT, I also acknowledge I don't really understand why I reach for certain things that go beyond playing the instruments, panning and playing with the EQ. I understand some tools are for subjective reasons in shaping sounds but some are essential for a good output for listeners across many types situations when listening?

One of my biggest issues I want to overcome now is getting louder music files. For example, my music often sounds great coming out of my speakers but when played in a custom radio station in Grand Theft Auto for example, it's super low compared to professionally mastered songs I have in the same folder. Same when listening on cheap earplugs from my phone. I know it have something to do with using compression and/or a limiter (probably more things?). But before I just start grabbing those tools, I really want to know what I'm doing and why. And what should be done before reaching for those 2 particular tools?

huggermugger already explained two important things, but there is more to loudness than compression. you should already know that the human ear percieves different frequencies at different volumes. mid frequencies are percieved the loudest, followed by highs and then bass. thats why the bass usually has the highest decibel value in a mix. but that also means that you can raise mids and highs to a quite high volume and DB level before you hit the limiter - and therefore give the track a way louder expression. it has to be done carefully to keep a balanced sound though, but from my experience that is the main trick to achieve loudness (aside from more technical stuff like eliminating peaks, crackle, infrasound etc.). but be aware that constant loudness is not a must, but a style. you can handle the relation petween peak and bottom loudness in any way you want.

[Mistro17]

Thanks for all the advice so far. I do have some Izotope plugins I grabbed during giveaways but never really got into them (GAS). I tried them on a couple tracks but didn't really know what I was doing. I grabbed the demo versions of Sonible Smart:Comp 2 and Smart:Limit last night and I'm very impressed. Besides the capabilities of these plugins, what draws me to them is the visual feedback of what's going on which is great for beginners. But I don't want to just rely on the great sounding result I hear when I press that learn button.

After treating 2 songs last night, it sounds way better than before but I notice I have to go back and pump up the low end. There's only 3 instruments in one song, 4 in the other. Basically Scenic Wishful Harp, EZBass, and EZDrummer2. The one with 4 also have NOIRE piano by NI. One of the things I did before using copmpression or the limiter was very slightly bring up some mid frequencies in the instruments which brought out more harmonics (?) and made them sound brighter. I also put a high pass filter on the bass and drums to tame the low end a bit but I'm rethinking that after listening to the song.

I'm gonna look into those free level checker VSTs. I see that streaming services etc. have standard requirements. I also get from some engineers some of those guidelines are usually ignored. I'm not making music for any particular platform just sometimes I may want to make music videos to put on YouTube or share through Soundcloud. Is there a more universal approach that's a sweet spot when it comes to final product loudness? And speaking of loudness, I do aim to keep as much natural sound especially from acoustic instruments as possible. I do like dynamic range.

[selig]

IF loudness is your goal, you cannot overlook the power of clippers. Traditional compressors to me don’t get things ‘loud’, they get things smoother, punchier, more cohesive, or more ‘aggressive’.
I use compressors to align the crest factor of tracks for a more cohesive mix. Extreme example: if you have some tracks that are super transient heavy, and others that are super ‘sausage’ or pretty much at the far opposite end of the crest factor spectrum, it will be near impossible in my experience to get them to blend. The transient tracks compared to the sausage tracks will either be too soft to be heard or loud enough that their transients clip the mix. There is no way to find the sweet spot because it doesn’t exist - you can move faders around or even add EQ all day but you need a way to get them closer to each other in the micro-dynamic sense before you can even get basic a good solid starting point/basic balances.
Now to be fair, this is an extreme example, but to some degree it’s what I’m doing with compression across the entire mix process. Towards the end I will start to consider ‘glue’ type compression for cohesion, but you can’t glue things together if they don’t fit ‘tight’ to begin with IMO!
As for loudness: as it is the peaks/transients that keep things from getting loud, you really have little choice but to shave them down a bit if you want to retain any sense of macro-dynamics while still sounding loud in my experience. Compression/limiting does different things to clipping, and you’d be surprised how much clipping you can get away with even on acoustic instruments. The key is to keep the ‘clips’ short - you don’t want the waveform to be clipped for longer than a few milliseconds (maybe 5 ms max depending on the source) or have too many short clipping segments back to back. If you can keep your clipped bits to a small overall percentage of the time you can retain dynamics and still gain headroom. IF dynamics are super important, then add clipping across all transient tracks gradually and carefully. Even getting only 3 dB more overall loudness by using clipping can make a huge difference, but for more aggressive genres you can go much further on some tracks.
Like anything, the best way to learn what is “too much” clipping (or anything) is to go BEYOND the amount that sounds good so that you define the boundaries of what you’re willing to accept. Once you know what “too much” sounds like, you can recognize it sooner and not have to dial things back after going too far. But IMO you need to train your ears to recognize the effects of “too much” for any processing (too much compression, too much reverb, too much low/mid/high end, too much saturation/distortion, etc). And finally, remember that “too much” is defined by YOU.

[Mistro17]

IF loudness is your goal, you cannot overlook the power of clippers. Traditional compressors to me don’t get things ‘loud’, they get things smoother, punchier, more cohesive, or more ‘aggressive’.
I use compressors to align the crest factor of tracks for a more cohesive mix. Extreme example: if you have some tracks that are super transient heavy, and others that are super ‘sausage’ or pretty much at the far opposite end of the crest factor spectrum, it will be near impossible in my experience to get them to blend. The transient tracks compared to the sausage tracks will either be too soft to be heard or loud enough that their transients clip the mix. There is no way to find the sweet spot because it doesn’t exist - you can move faders around or even add EQ all day but you need a way to get them closer to each other in the micro-dynamic sense before you can even get basic a good solid starting point/basic balances.
Now to be fair, this is an extreme example, but to some degree it’s what I’m doing with compression across the entire mix process. Towards the end I will start to consider ‘glue’ type compression for cohesion, but you can’t glue things together if they don’t fit ‘tight’ to begin with IMO!
As for loudness: as it is the peaks/transients that keep things from getting loud, you really have little choice but to shave them down a bit if you want to retain any sense of macro-dynamics while still sounding loud in my experience. Compression/limiting does different things to clipping, and you’d be surprised how much clipping you can get away with even on acoustic instruments. The key is to keep the ‘clips’ short - you don’t want the waveform to be clipped for longer than a few milliseconds (maybe 5 ms max depending on the source) or have too many short clipping segments back to back. If you can keep your clipped bits to a small overall percentage of the time you can retain dynamics and still gain headroom. IF dynamics are super important, then add clipping across all transient tracks gradually and carefully. Even getting only 3 dB more overall loudness by using clipping can make a huge difference, but for more aggressive genres you can go much further on some tracks.
Like anything, the best way to learn what is “too much” clipping (or anything) is to go BEYOND the amount that sounds good so that you define the boundaries of what you’re willing to accept. Once you know what “too much” sounds like, you can recognize it sooner and not have to dial things back after going too far. But IMO you need to train your ears to recognize the effects of “too much” for any processing (too much compression, too much reverb, too much low/mid/high end, too much saturation/distortion, etc). And finally, remember that “too much” is defined by YOU.

Thanks. So basically, we are treating harmonic spikes to bring the quieter parts of the sounds closer to the gain levels without destroying the overall character (or destroy if desired)? I never considered using clippers. Thanks for bringing that up. I have a feeling many more audio tools are gonna make more sense to me going forward.

Are there good clipper tools stock inside Reason? If not, which one would you recommend?

[Popey]

Not sure about rack extensions but there are a number of free clipper vsts if you google it.

[Mistro17]

What does it mean when the instrument faders are not clipping but the master fader is slightly clipping in the mixer section? The master fader is bouncing a couple in the red. Nothing to worry about?

[crimsonwarlock]

Not sure about rack extensions but there are a number of free clipper vsts if you google it.

Gclip RE has been set free not too long ago.

[selig]

What does it mean when the instrument faders are not clipping but the master fader is slightly clipping in the mixer section? The master fader is bouncing a couple in the red. Nothing to worry about?

OK, backing up a bit with some terms. Clipping is indicated by clipping lights, only in three places in Reason (Master meter, transport meter, Big Meter). The “red” you mention is arbitrary, with no relation to clipping at all. In fact, you can turn the red LEDs off if you like by going to the Big Meter and turning the VU Offset all the way down/counter clockwise.
So, faders cannot clip, channels cannot clip - only the final output can clip.
Does that answer your question?

[selig]

Thanks. So basically, we are treating harmonic spikes to bring the quieter parts of the sounds closer to the gain levels without destroying the overall character (or destroy if desired)? I never considered using clippers. Thanks for bringing that up. I have a feeling many more audio tools are gonna make more sense to me going forward.

Are there good clipper tools stock inside Reason? If not, which one would you recommend?

All clippers basically do the same thing, but the interface can make one ‘better’ for you than others. Luckily there are free ones out there including REs so you can compare.

As far as ‘harmonic spikes’ I have no idea what that means, but it sounds like you have the general idea. It’s the difference between the peak and average levels (crest factor) that play a large role (along with frequency) to determine how we perceive loudness. While I don’t believe there is some ideal ‘loudness’ with regards to crest factor, as I said before I find it difficult to mix things of wildly differing crest factors.

[Steedus]

What does it mean when the instrument faders are not clipping but the master fader is slightly clipping in the mixer section? The master fader is bouncing a couple in the red. Nothing to worry about?

OK, backing up a bit with some terms. Clipping is indicated by clipping lights, only in three places in Reason (Master meter, transport meter, Big Meter). The “red” you mention is arbitrary, with no relation to clipping at all. In fact, you can turn the red LEDs off if you like by going to the Big Meter and turning the VU Offset all the way down/counter clockwise.
So, faders cannot clip, channels cannot clip - only the final output can clip.
Does that answer your question?

Might be good to add, that the reason why "clipping" the red LEDs on the channels etc isn't really "clipping" is because of the 32-bit floating point calculations in the softwa.... ehhh ... lets just say there's waaay more headroom that it doesn't actually "clip".

(amateur explanation)

But that helped me get past worrying about the channel meters and just focus on the Main Output meters.

[Mistro17]

OK, backing up a bit with some terms. Clipping is indicated by clipping lights, only in three places in Reason (Master meter, transport meter, Big Meter). The “red” you mention is arbitrary, with no relation to clipping at all. In fact, you can turn the red LEDs off if you like by going to the Big Meter and turning the VU Offset all the way down/counter clockwise.
So, faders cannot clip, channels cannot clip - only the final output can clip.
Does that answer your question?

Might be good to add, that the reason why "clipping" the red LEDs on the channels etc isn't really "clipping" is because of the 32-bit floating point calculations in the softwa.... ehhh ... lets just say there's waaay more headroom that it doesn't actually "clip".

(amateur explanation)

But that helped me get past worrying about the channel meters and just focus on the Main Output meters.

Ok, so I take that as nothing to worry about I have to unlearn a lot of terms I get from watching too many YouTube videos.

[Mistro17]

What does it mean when the instrument faders are not clipping but the master fader is slightly clipping in the mixer section? The master fader is bouncing a couple in the red. Nothing to worry about?

So, faders cannot clip, channels cannot clip - only the final output can clip.
Does that answer your question?

Just saying a formal yes. Thanks.

[Jac459]

Actually I find this video very clear, you should give it a go:

[RobC]

Actually I find this video very clear, you should give it a go:

Haven't watched it, because these clickbait videos are always a disappointment.
That said, regarding the thumbnail - the left waveform is a mix; the right one is a squashed and clipped loudness war master.

Now, to some discussion in this post. I understand if people want to make music, where sounds are kind of sustained, saturated and thus full of harmonics, aka "loud". But the thing is, you will be left with a ton of headroom, and I can't help but wonder, why not make use of that, and add dynamics to the harsh sounds? That edge will still be there and it will have life, too as a bonus.

[selig]

OK, backing up a bit with some terms. Clipping is indicated by clipping lights, only in three places in Reason (Master meter, transport meter, Big Meter). The “red” you mention is arbitrary, with no relation to clipping at all. In fact, you can turn the red LEDs off if you like by going to the Big Meter and turning the VU Offset all the way down/counter clockwise.
So, faders cannot clip, channels cannot clip - only the final output can clip.
Does that answer your question?

Might be good to add, that the reason why "clipping" the red LEDs on the channels etc isn't really "clipping" is because of the 32-bit floating point calculations in the softwa.... ehhh ... lets just say there's waaay more headroom that it doesn't actually "clip".

(amateur explanation)

But that helped me get past worrying about the channel meters and just focus on the Main Output meters.

Actually that’s not the reason. The “red” LEDS are WELL below clipping, starting at -12 dBFS if you don’t change this default.
I’ve asked several times for them to change the color to orange or yellow (like the sequencer audio input meters) because folks tend to associate red with things like “clipping”.
So the reason the red LEDs are not really clipping is because they are indicating a signal below 0 dBFS (not clipping).
BUT WAIT - it gets more confusing if you are talking about the channel meters, because they do not show peak levels. So while they may indiace the signal is below 0 dBFS, you don’t see peak level so you have no idea where the peaks are hitting.
BUT WAIT AGAIN - like you said, with floating point systems like Reason there is only one place you can clip, which is on the final outputs or when exporting from Reason. Hope I’ve not made things less clear…

[RobC]

[Jac459]

Actually I find this video very clear, you should give it a go:

Haven't watched it, because these clickbait videos are always a disappointment.
That said, regarding the thumbnail - the left waveform is a mix; the right one is a squashed and clipped loudness war master.

Now, to some discussion in this post. I understand if people want to make music, where sounds are kind of sustained, saturated and thus full of harmonics, aka "loud". But the thing is, you will be left with a ton of headroom, and I can't help but wonder, why not make use of that, and add dynamics to the harsh sounds? That edge will still be there and it will have life, too as a bonus.

Well you should still give it a try... This youtuber is part of the 2 excellent teachers you can find on the net. She already did video for Reason Studios Channel by the way and she is also an electro artists referenced in the raising talent Spotify playlist.
Not a fan but I guess it is worth give it a go...

[RobC]

Well you should still give it a try... This youtuber is part of the 2 excellent teachers you can find on the net. She already did video for Reason Studios Channel by the way and she is also an electro artists referenced in the raising talent Spotify playlist.
Not a fan but I guess it is worth give it a go...

I understand, but there are plenty useful tutorials already. I usually watch Dan Worrall's stuff to learn something I overlooked, and there's more than enough help just on this forum alone. And as a bonus there's discussion here - no tutorial can beat that.

Selig or Benedict for example are respectively excellent teachers, too.