Download Article As PDF
In this tutorial my aim is to fully explain the workings of just one compressor, Variety of Sound’s ThrillseekerLA. While much of what you learn will be applicable to other compressors, my goal is that you have all the information you need to get fantastic results with this one. This is not meant to be an extension of the manual, but a detailed explanation of the concepts behind the compressor, its controls and how they relate to the wider aims of compression.
This is an intermediate level tutorial. You should be familiar with the basic controls of a compressor and what they do. If you are a total beginner i suggest you read Dynamic Range Compression Pt 1 – What Is Dynamic Range?
Where Do I Download It?
ThrillseekerLA is available as a 32 bit VST plugin for PC’s from the download area of the VOS blog:
It was chosen because not only is it free, it also sounds amazing! It is just as good or better than commercial equivalents, so I am happy to recommend and promote its use.
It is not an attempt to recreate a hardware compressor in software, but its aim is to take the best elements of a type of hardware compressor know as an ‘opto-compressor’ (in particular the LA-2A) and update them.
Is ThrillseekerLA The Compressor For The Job?
What is needed to get clear from the start is that every compressor will sound different, the controls will act differently, and some will be better at doing certain types of compression than others. While the basic controls like ATTACK and RELEASE will have a similar function, the resulting change to the sound may be totally different.
Before you even think about compressing a track you should be thinking ‘what kind of compression do I want?’. This will effect your choice of compressor. This is the first problem for the beginner.
You may have read on forums that there are tube compressors that make things ‘fat’, that Daft Punk used an Alesis 3630 to make things ‘pump’, and maybe that software compressors ‘suck’ and you need to go and buy some hardware. Of course, there are grains of truth in these blanket statements, but it’s mainly out of context and worthless information. It serves to confuse, and is often linked to cynical marketing to get your money.
It takes a long time to learn the sound of compression, even longer to learn the subtle differences between different makes and models of compressor. If you don’t feel you are getting it yet, don’t worry, you will, just relax, Learn Digital Audio is going to help you
What Makes The Sound Of ThrillseekerLA?
What makes any compressor sound the way it does?
This is important to understand when you are choosing what compressor to use. There are 2 main things to consider, I will attempt to define them:
- ACTION. This is what makes it a compressor. By this I mean how the compressor reacts to the signal and changes the volume of the incoming audio. This depends on its design. For example, if you had 2 different designs of compressor with almost identical settings they would change the volume of the audio in different ways. Some designs are better for different sources, maybe you want a punchy design for drums and a smooth design for vocals? I would like you to imagine that the ‘action’ of the compressor as if it had no sound other than changing the volume of the audio, it is theoretically perfect with zero distortion or changes to the original sound other than raising and lowering its volume. This is impossible for any compressor, but just for the sake of this article please go along with me.
- TONE. In hardware this might be the fact it uses tubes, has a certain type of wiring, dulls the audio or a million other factors. In software, which has no tone of its own, this might be the fact it has emulation of hardware saturation included, this is addressed in the INTERSTAGE section of ThrillseekerLA.
ThrillseekerLA Is Based On An Opto, What Is So Special About Opto compressors?
What is good about opto compressors is their unique ‘action’. They compress the audio in a ‘program dependant’ way with a certain characteristic. Program dependant is another way of saying signal dependent, or ‘what you are processing’ dependent. Without going into too much detail, this means that the compression ATTACK/RELEASE times change with the gain reduction. Other compressors can be program dependent too, but they won’t react in the same way.
I could describe the way opto’s react as ’rounding’ or ‘larger than life’ or ‘natural’ or any other phrases you have heard 1000 times before. You need to hear it yourself to get it.
Here are 3 examples of different compressors attempting to do the same job, if you are new to compression you will probably hear very little difference. Listen out for the overall envelope shape of each word that is loud enough to trigger compression.
PSP MasterComp Compressed
It is the ‘action’ we are listening for, not the ‘tone’.
From the Thrillseeker manual it states:
“To get the most out of this device one should understand that its gain riding characteristics are highly dependent on the gain reduction amounts which the unit can be driven into”
and from the VOS blog:
“Depending on the load in the circuit the recovery from gain reduction slows down the more it comes to zero GR. The effect is similar to optoelectronic panel based (vintage) gain reduction stages. Recovery can get really slow by this effect and getting the most out of such units requires to find their specific sweet spot for any particular audio material.”
Basically however you set the ATTACK and RELEASE values, they will change depending on how much gain reduction you have. This helps to avoid unpleasant compression artifacts like pumping and bad sounding distortion. Opto compressors are famous for sounding great on vocals amongst other things.
What Do The Controls Do?
Mix-Level – select either 0dB or -18dB
This is to tell the plugin how loud your incoming audio is going to be. Many professional engineers will have all their audio tracks recorded at about -18dB RMS to prevent overloading the DAW, it is a part of ‘gain staging’. This is a VERY good idea that I suggest you start doing right away. When you record audio make the average input reading on your DAW RMS meters -18dB. Also output all your soft synths/samplers at this level too.
If you know most of your audio tracks are about -18db RMS in volume select -18dB on this switch. If your tracks are louder, maybe going up to nearly 0dB, then select 0dB
SC-Filter (Side Chain Filter) – select OUT or IN
The side chain is the name for the audio that the compressor reacts to, it is often a modified copy of the track you are compressing fed in externally, but in this case Thrillseeker has its own internal copy built in that you can modify with this switch.
It you were to cut bass from this side-chain signal and add some highs then it would make the compressor react less to bass and more to highs. This would effectively result in more bass and less highs in the compressed signal. This is exactly what this button will do if turned on! It is a preset EQ setting for the side chain. It may not result in exactly what I described happening for your audio depending on the ‘program material’, but please try flicking this switch on and off when you have gain reduction going on. You will hear for yourself.
SC Route (Side Chain Route) – Select INT or EXT
If you switch this to EXT (External) then you can use inputs 3 and 4 into the plugin to be the side-chain (the audio the compressor reacts to). This maybe impossible in your DAW, or easy (if you are using Reaper for example). You will need to read your DAW manual or get on the forums for help.
External side chain routing is an article in itself, it has many uses, the most popular today is probably the infamous dance music pumping sound. You would feed ThrillseekerLA’s external side-chain with your bass drum track so the source you are compressing will jump up and down in time. Alter ATTACK and RELEASE to fine tune the groove.
Input – Select -36 to +36
This type of compressor does not have a threshold control. The more you drive the compressor by increasing the input the more gain reduction you get. Turn the input control to the right to increase the input. You will see the change in the display on the VU meter.
Attack – Select 0 – 100ms
The value you enter here is strongly tied to the amount of gain reduction taking place. It is the amount of signal that is let through before compression begins. A typical use of longer attack is to let the first bit of a percussive sound through so it gains punch.
Release – Select 30 – 500 ms
Again, strongly tied to the amount of gain reduction. This is how long the compressor takes to recover after the attack phase.
SC Lowcut – Select 20 – 500 hz
This SC Lowcut control filters out the bass from internal side-chain so that the compressor thinks it is not there for processing gain reduction. REMEMBER a single band compressor is ‘stupid’ in that it reacts to overall volume level, not frequency.
The reason you may well want to cut the bass from the side-chain is 2 fold:
- Bass energy is stronger than higher frequency energy. This means the compressor may react more to bass and reduce it more which could cause a lack of balance in the compressed signal. This can be fixed with this control.
- You might not want the compressor to clamp down on the kick drum or bass line in a song that you are mastering. This will stop the compressor pumping in time with the bass content.
Side note: if you want the dance music pumping effect DON’T try to get it on the master bus. Far better to use side-chain compression controlled by the kick drum on the tracks that want it
This changes how much compression is possible. If you set it to 0 then you have no compression and you can use the plugin just for saturation.
This increases or decreases the volume of the final compressed audio. It is used to match the level of the compressed and uncompressed audio for the purpose of comparison.
This control is VITAL. You must match the perceived overall volume of compressed and uncompressed to be able to judge how well you have compressed the audio.
Interstage – select In or Out
This turns on and off the saturation effect that is controlled by the extra controls in the bottom part of the plugin
What Is The Interstage / Stateful Saturation All About?
(Remember we defined the ‘action’ as just the effect on the volume, ie the compression, and the ‘tone’ as all other changes in the sound, inc saturation.)
All hardware will add a little harmonic distortion (also known as saturation) to the audio passing through it. All the classic recordings before digital had varying degrees of saturation, and people love them. This harmonic distortion adds something to the sound. It is musical as it is related to the harmonics of the sound. When you heard tubes make things ‘fat’, the person was probably saying that the distortion that the tubes added was pleasing. There are many types of distortion that people do not find pleasing. Distortion can be good or bad depending on the context and your personal preference.
Software adds small amounts digital distortion too, and most people dislike this type. It often comes in the form of inter-modulation distortion and aliasing distortion. This is not musical distortion and no-one wants it (well, maybe some obscure genre of IDM is based on it ) Thrillseeker has very low levels of this type of distortion, and that’s one reason it sounds so good.
We have an even greater control of the saturation in Thrillseeker using the Spectrum control. This adds the harmonic distortion to different areas of the frequency spectrum. We can for example distort just the bass by selecting LF or just the highs selecting HF.
ThrillseekerLA’s Stateful Saturation Vs Hardware Saturation
Saturation in software has traditionally been very much inferior to using real hardware. This is because of the lack of complexity in the software’s algorithms and possibly the lack of CPU power to calculate the results needed. It has often sounded flat, fake and harsh. ThrillseekerLA uses a new approach called ‘stateful saturation’. I won’t go into the details here, but in my opinion it is a great improvement and can really enhance the sound of your audio. What is even better is, if you don’t like it you can simply turn it off.
Example Usage Of ThrillseekerLA
Ok, now you know what the controls do and hopefully understand the concepts behind them, lets go through an example of real life use. This is just one of many different ways you might go about compressing a track. Lets pretend this is a vocal we want to compress:
- Choose your MIX-LEVEL, 0dB or -18dB depending on the volume of the track being compressed.
- Turn INTERSTAGE out, SC-FILTER out and SC-ROUTE to INT, SC LOWCUT to 20, RANGE to 100. This is basically turning all the extras off and resetting to default.
- Set ATTACK and RELEASE to as fast as they will go, all the way to the left.
- Increase the INPUT until you start to see gain reduction on the meter. LISTEN. Make sure the meter is on GR mode (gain reduction).The aim in compressing a vocal is to stop words jumping out and to even out the level so it can all be heard clearly over the mix. Too much compression will make it sound flat and boring, too little and it will sound uneven and amateurish. Less compression often sounds more natural, but most modern music does not want to sound natural, it wants to sound in your face!In many compressors, when compressing vocals, its a good goal to see the GR meter fall back to 0 often and set the ratio to about 4:1. In ThrillseekerLA we have no control over the ratio, and also the gain reduction effects the internal ratio. We are best to take the advice of its creator:”2 – 6dB of gain reduction - the typical working range for almost all kind of duties. Pacing this range the compressor is able to achieve both: long term RMS level smoothing and fast short-term peak management w/o any obtrusive gain riding artifacts. This works on virtually any material and especially also on difficult to handle content: Whether it’s a vocal or brass take or even a slap bass with huge and fast dynamic changes – the program dependency is able to deal with all that without any intervention by hand.”
- Adjust ATTACK, now things get interesting. You aim is to make the very front transients of the vocals not be flat and boring, but to also not jump out too much. Often a good start on other compressors is about 5ms, here you will have to play around about 8 – 10 o’clock position until it sounds right. Keep an eye on INPUT and GR, remember this is effecting the ATTACK. You may need to play with them both at the same time carefully listening to what is going on.
- Adjust RELEASE. Again, now dial in your release. It interacts with ATTACK and INPUT levels. In some ways ThrillseekerLA is not a beginner friendly compressor as it has a lot of controls and setting the ATTACK/RELEASE is tricky as they are linked so strongly with the GR amount. In another way it is good for beginners as there is a lot of clever program dependent stuff going on which makes it harder to get bad settings.
- Turn on INTERSTAGE and try adding a little saturation. It is easier to hear what is going on by turning it all the way up so the red bars go all the way to the left. You can then flip through all the different SPECTRUM options hearing what they do. As an example, if your vocals sound thin, turning INTERSTAGE to 0.5 will add some warmth back to them. Adjusting the THD slider will decide how much warmth to add. If INTERSTAGE is used well it can add thickness, presence, warmth, brightness. If used badly it can add harshness, sibilance or just sound like its messed up your tracks EQ.
I hope this article has given you a better understanding of ThrillseekerLA and how to get the most out of it. If you would like any questions answering or have any suggestions, just leave them in the comments section below.
Please consider joining the Learn Digital Audio mailing list on the top right of the blog, also clicking ‘like’ on the Facebook plugin just below, it helps keep me motivated to write more in-depth articles like this!