PULSAR COMPRESSORS

A space for learning and studying the Scope environment and music-making in general.

Moderators: valis, garyb

User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

I’m not the one to teach this, but since there is no one writing about anything at the moment, I just would like to push it a bit, starting a new tread about the use of Pulsar Compressors.

Please, if there is someone with enough knowledge to teach us about Compression, it would be great.

I would like to be answered the following questions to start:

· What is it a Compressor?
· What is does to the sound?
· What do we use it for?
· Which recommendations do you have about compression?
· Which instruments are best compressed and why?
· Which are the risks of compression when what you want is a natural sound?
· What kind of experiment can we do to actually and objectively learn to use the different Pulsar compressors?

Thanks for you help and contribution :wink:


_________________
Music is the most Powerful Language in the world! *INDEED*

<font size=-1>[ This Message was edited by: Nestor on 2002-05-20 10:22 ]</font>
User avatar
wayne
Posts: 2375
Joined: Sun Dec 23, 2001 4:00 pm
Location: Australia

Post by wayne »

Here's a link 4 now
King of Snake
Posts: 1544
Joined: Fri Apr 13, 2001 4:00 pm
Location: the Netherlands
Contact:

Post by King of Snake »

User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

Somebody said:

Whether rightly or wrongly, the way in which current pot music is produced involves strict control over the dynamics of both instruments and vocals. By dynamics, we mean the range between the loudest and the quietest sounds produced by an instrument, voice or complete piece of music.

Much of today’s music is rhythm based, but if you listen to the drum sounds on virtually any pop, rock or dance record, you’ll find that the level is very even: there’s little or no use of “light and shade”. This sometimes makes it difficult, for example, to distinguish between a recording of a drum machine and the real thing, or a synthesized bass sound and a bass guitar. And this, in turn, sets the pattern for the rest of the instrumentation. If light and shade is required, it is often created by dropping instruments in and out of the mix rather than by changing the balance of instrumentation to any significant degree. Similarly, the vocal level will be very even so as to ensure the vocals always sit comfortably with the backing music.

Moving slightly away from mainstream pop, you occasionally hear more “natural” sounding recordings from those artists who have realised the potential of compact discs for expressing wide signal dynamics. But even here, unless you are listening to a purist recording of an orchestra or a folk group, you are unlikely to be hearing the original dynamics of the performance.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

COMPRESSORS
Artistic considerations aside, even the latest digital recording systems are incapable of reproducing the full dynamic range encompassed by the human ear, so for both practical and aesthetic reasons, it’s often desirable to have some controllable means of reducing the dynamic range of a signal or recording. This is generally achieved using a device called a compressor, and the effect of dynamic range reduction brought about by the use of one of these devices is known as compression.

Compression ca be used on just about any sound source or mix of sound sources, but it is almost invariably used on pot vocals – and untrained inger-s voice is likely to fluctuate in level quite considerably, with the result that some words may be obscured by the music while others are too loud. Drum machines and synthesizers tend to have fairly well controlled dynamics, so further compression is seldom necessary in the context of controlling levels, but as I’ll explain later, compression may also be used as a special effect on these and other sounds.

The ability to compress the natural dynamics of a sound in a controlled manner is very important in contemporary music production, and it’s probably fair to say that after a reverb unit, a compressor is the most important signal-processing tool in the studio.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

GAIN CONTROL
To best understand the mechanism by which the compressor does its job, it’s useful to examine the way dynamic range control can be achieved manually. Before the compressor was widely available, manual “gain riding” was the commonly employed technique used to keep signal levels under control. Gain riding involves the mixing engineer sitting; hand on fader, trying to anticipate excessively low or excessively high signal levels so that he can make quick adjustments to keep the subjective signal level more or less constant. The ear follows the signal level from the loudspeaker enabling the brain to direct the hand, which moves the fader until the level is corrected.

If the session is being conducted by an experienced engineer who is thoroughly familiar with the dynamics of the music being recorded, then gain riding can be reasonably effective in levelling out untoward peaks and troughs in the dynamics of the piece. But of course, there are obvious problems with this approach. Gain riding requires the undivided attention of the engineer and this means that other aspects of the session may be neglected. In many situations the only way around this is to enlist the help of other members of staff, particularly when there are several different voices and instruments that have to be controlled separately.

There’s also the human response time to consider. It simply isn’t possible for an engineer to react quickly enough to the kind of momentary peaks in volume which characterise a vocal or instrumental performance unless he is already thoroughly familiar with the recording – inevitably, adjustments are made a fraction of a second too late.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

WHAT A COMPRESSOR DOES
All a compressor does is to reduce the level of signals that exceed a predecided level so as to reduce the overall signal dynamics. Most compressors operate on a threshold system; signals lower than a threshold level set by the user remain unaffected while those exceeding the threshold are turned down. The threshold is generally adjusted by a front panel control, and there is usually some kind of display to indicate how much gain reduction is taking place. How much gain reduction takes place when the signal level exceeds the threshold level depends on something called compression ratio.
*MUSIC* The most Powerful Language in the world! *INDEED*
King of Snake
Posts: 1544
Joined: Fri Apr 13, 2001 4:00 pm
Location: the Netherlands
Contact:

Post by King of Snake »

It is also important to realise that in most modern dance music, compression is not really used to overcome gain-fluctuations, but usually for more "artistic" reasons. It's used to sculpt the actual sounds, changing the envelopes and the way the sounds are perceived, usually in order to make them more powerful.

Most notable compressing drums gives can make them more "snappy" and "in your face".
The same goes for synth sounds, usually if played from a sequencer with even velocity values a synth's output volume will not be fluctuating very much so you really don't need a compressor for that. But a compressor can still be used to emphasize the attack of a lead or help to place basslines better in the mix. Please don't ask me exactly how, I'm still experimenting with it myself but the link I posted before might give some indications.


_________________


<font size=-1>[ This Message was edited by: King of Snake on 2002-05-28 09:17 ]</font>
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

This is very interesting. So you mean that you are actually using Compressors as FXs rather than as Compressors. But could you be a bit more specific about which kind of FXs can you pull out of it? I’ve never used a Compressor with the aim of effecting something, but rather to give it some punch, power and getting this flat result of a finer performance, but that’s it. I know you said: “Don’t ask me for”, but I will Snake… cos even if you know little, you can explain this little and will be useful too. Common King, give as a clue!
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

ELECTRONIC GAIN CONTROL
Fortunately, the development of early recording systems was closely followed by the invention of a practical electronic compressor design. A compressor works in much the same way as our gain riding engineer in that it constantly monitors the signal level then makes the necessary adjustments – but it does so much faster and far more precisely.

Obviously it isn’t practical to have a robot arm moving a fader, and even a motorised fader may not be fast enough, so an electronic gain control circuit is employed in its place. These may be based on VCAs (voltage controlled amplifiers), valves, or LEDs and photocells, though it is also possible to execute compression entirely in the digital domain. The optical gain control circuit using a photocell is interesting as instead of a VCA, the side-chain controls the brightness of a LED which shines on a photo-resistor. The resistor is placed in a gain control circuit, so as the LED changes brightness, the signal changes level. Photocells don’t always behave linearly and they tend to have a fairly slow response time, which is part of the reason this type of compressor has a characteristic sound. A number of optical compressors were designed in the 1960s, but renewed interest in vantage recording techniques has led to their rediscovery.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

COMPRESSION RATIO
Ratio is simply the change in output level that results from a given change in input level; for example: a compression ratio of 2:1 means that a 2dB change in input level will only give a 1dB change in output level. IN practice, a 2:1 ratio is adequate for a wide variety of applications but much higher ratios may be chosen to suit specific situations. This again is user-determined; this time by means of the ratio control. If the ratio is made very high, say 10:1 or greater, then we have a situation known as limiting where an input exceeding the threshold level is subjected to such a high level of gain reduction that the output is effectively prevented from rising above the threshold level by any significant degree. This assumes that the compressor/limiter can respond quickly enough to catch sudden signal peaks.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

COMPRESSION OR LIMITING?
Most compressors have a wide enough ration range that they can double as either compressors or limiters – hence the title compressor/limiter. Absolute limiting (where the output never exceeds the threshold value at all), requires a ratio of infinity: 1, but in practice, ratios greater than around 10:1 have much the same effect.

The reasons for employing limiting are rather different to those for compression, and intensive limiting almost always has an audible effect on the sound being treated unless the limiter is very carefully designed. More often than not, a limiter will be set so that it only operates on high signal peaks – most of the time it will do nothing. In cases where a signal must not be allowed under any circumstances to exceed a certain level, some form of limiting is essential. Typical examples would be in the case of the radio transmitter, where over modulation is not permitted, or the digital recorder, where there is absolutely no headroom above 0dBSF (0dB Full Scale). In live sound work (and in some active studio monitor systems), limiters are employed to make sure that power amps are not driven into clipping because the harmonics thus created can easily destroy expensive tweeters.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

ATTACK TIME
Just as our studio engineer takes a finite time to respond to peaks in the programme material, so does the compressor – but it’s still much faster than any human. Indeed, modern circuitry is so fast that the gain can be turned down before the input waveform has had a chance to overshoot far enough to cause problems. However, this degree of speed is not always necessary or even desirable, depending on the application.

The time taken for a compressor to respond to a signal that has exceeded the threshold is called the attack time, and, for reasons that we will investigate shortly, this is generally under user control, though some compressors also include an auto mode. Likewise, the release time, the time that the compressor takes to return to its normal gain once the signal has fallen below the threshold, is again commonly linked to a front panel control. Again, this may also be handled automatically by an auto mode. Auto modes are described towards the end of this chapter.

The attack and decay times are made variable because their optimum settings depends on the type of material being processed. For example, a powerful bass drum beat in a compete mix will cause the compressor to reduce the gain of everything in the mix, not just the bass drum. As a result, any high frequency, low level detail – such as a hi-hat occurring at the same time – will also be turned down, leading to a dulling of the overall sound. One way round this, apart from using as little compression as is practical, is to increase the attack time slightly so that the compressor doesn’t respond instantly. In this case, the leading edge of the beat, including the attack of the hi-hat, is allowed through at full level before the gain is pulled down, thus producing a more punchy sound.

Some designers also use a high-pass filter to, in effect, sneak a little of the high frequency end of the input material past the compressor without it undergoing gain reduction.

Setting a longer attack time is also extensively used when recording bass drums or bass guitars to allow the initial click or slap to come through strongly. It does mean that the signal may exceed what you consider to be a safe operating level for an instant, but any distortion caused by analogue tape overload is unlikely to be audible un such short duration sounds. In the case of digital recording or other more critical applications, it is common practice to insert a separate limiter into the signal path to catch any brief excess that might otherwise get past the compressor, and a number of commercial compressors also include a separate, fast acting limiter.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
kensuguro
Posts: 4434
Joined: Sun Jul 08, 2001 4:00 pm
Location: BPM 60 to somewhere around 150
Contact:

Post by kensuguro »

Sounds alot like the "correct" way to use comp/limiters to me. Well, it's true for real instruments that if you use too much of anything, the feel of the instrument and expressiveness is lost. With electronic instruments? Who cares! Compression, gate, whatever, is part of the expression! :grin:

Anyhow, I don't think I've found any specific tricks to do with comps, but I'll try to share what to do with situations as in: percussion, voice, percussive "real" instruments, and then finally, moog-ish synths.

So electro musicians have gotten into basing pretty much their lives on the kick. You wanna snap? wanna decimate? Want a quick one? Long one? Soft? Hard? There are as many kinds of kicks as there are shoes for women. :lol:

I usually set up my kick effects like
eq>comp>limit
or
eq>comp>saturate
or
eq>comp>saturate>comp>limit>eq>saturate>chili sause>garlic>stab it with a knife>etc. etc.

With percussion, you want to exaggerate a little with the comp. If you want a killer snap, max out the comp unit. Who cares, you can always limit it back to your perfered volume. That's how you gotta think. I'll use a kick drum as an example, but the same thing can be applied to anything else.

Specifically, you separate the kick into Snap, low attack, body, and then whatever that's left over. The perception of the kick depends on these elements. Start with max ratio, lowest threshold, lowest attack, and lowest release. This is always a good starting point for me. The compressor should be reacting like a mad man right now. If you like that sound? Use it.

Now, the start by fiddling with the attack. With the attack at minimum values, you're probably hearing too much Snap. Increase the attack a little to get some Low Attack. When you like it, start start increasing the threshold. You should start getting more and more body. When you like it, start lowering the ratio. Personally, I like to over do things so I pretty much do everything at max ratio. :grin:

You're probably going to get a very, VERY dynamic signal, so smash it some more with a limiter. If the limiter seems to be letting some of the Snap through, like the standard Pulsar limiter, then forget about the limiter all together and mash it with an overdrive. The standard Pulsar overdrive does a fairly good job at keeping the peaks down while adding some bad attitude to your kicks. Better yet, forget comping your kicks, but use a multicomp instead. But that's another story. :smile:

If you're going to do this sort of heavy dynamics processing though, you NEED separate tracks for every part. But, sometimes you can try putting the open hat channel and kick channel together to get a strange pumping effect. NEVER put the kick and snare together.

Finally, turn off solo and see how it sounds in the mix. Adjust as needed. NEVER finish fiddling with dynamics in solo.

So now your track sounds groovy and punchy. Now you want to work on them vocals. Vocals are tough to work with. It's much more expressive than a bunch of drums sounds. And a vocal track with bad dynamics is worse than anything else in the world.

With vocals, you want to be subtle. Dunno about everyone else, but I don't work too hard to make my vocals sound too loud. I work on mixing the track properly instead. Anyway, my compression scheme for vocals doesn't include maximizing the volume. Instead, I aim to make the lyrics clearer. If you want volume, stick it through a limiter. It's much faster than dealing with microscopic comp details.

So, start with threshold minimum, ratio max, attack and release minimum. First, play a loud portion of the vocal track, and set the threshold so that the comp reacts to 80-90% of the loud parts. Make sure that you can obviously hear the comp reacting. Then, go back to the softer section of the vocal, and see that the comp rarely reacts. Now set the ratio to zero and loop the loud portion of the song. Gradually increase the raio 'till you like the results. Then, adjust the attack so you're not getting too much of a punchy sound. Release can be pretty long, exaggerate a little. Longer releases keep the comp from over reacting.

Again, turn off solo and see how it sounds in the mix. Adjust as needed. I can't stress this enough... NEVER finish fiddling with dynamics in solo.

Working with real instruments is always a drag for me, though I do it alot recently. Plucked instruments in partucular, are very problematic in terms of dynamics. And unlike kick drums, you can't be too brute with the dynamics because you'll lose the "feel " of the instrument. Percussive instruments, like my conga, also give me headaches.

Both these instruments can be grouped together as DECAY instruments because all it has, is an attack, and decay element. So to overcome the dynamical problems, you'd want to keep these two elements under control. With decaying instruments, it's usual for most of the treble sounds to be towards the attack, and then the sound looses harmonics as it decays. So simply, the more attack, the brighter the sound, and the more it cuts through the mix. Hence the "in your face" effect.

The hardest part of these sounds is to find where EXACTLY the attack ends, and the decay begins. Start by setting ratio to max, and everything else to minimum. Slowly increase the attack. The first few milliseconds should sound like useless clicks. Then, increase it some more, and you'll start hearing more and more harmonics. Usually, I set it at "mostly click, but I'm generous so I'll let some harmonics in". Then comes the hard part... the release vs. decay setting. It totally depends if you want long tones, or just percussive sounds. Anyway, you gradually increase the release so as to get the right amount of tone (percieved as notes) you want. This is pretty much up to you. Then, set the threshold so that the comp reacts about 50% of the time. Then lower the ratio untill you like the sound.

OK, so that's enough of "real" sounds. Back to synth world. One of the synth sound that almost always benefits from compression, is the moog, or any variation of it. Compressing it definitely makes it sound more funkified and thus increases the general precieved funkalization of the mix.

Set ratio to max, everything else to minimum. Start by making the comp react ONLY to the attacks, by carefully tweaking the threshold. Be very careful. Then, adjust the attack time to compliment the attack time of the moog's filter envelope. Release can be kept at minimal values. Then, adjust the ratio while listening to the sound in the mix. The ratio will be the "back off" or, "come close" knob.

So there you have it. Doing comp in a nutshell. The methods written here can be a good starting point for beginners. When you get used to it though, you don't always have to start with ratio max, everything else minimum. But when things get out of hand, it's always a good place to start.

Happy compressing then! :grin:
King of Snake
Posts: 1544
Joined: Fri Apr 13, 2001 4:00 pm
Location: the Netherlands
Contact:

Post by King of Snake »

On 2002-05-28 22:39, Nestor wrote:
This is very interesting. So you mean that you are actually using Compressors as FXs rather than as Compressors. But could you be a bit more specific about which kind of FXs can you pull out of it? I’ve never used a Compressor with the aim of effecting something, but rather to give it some punch, power and getting this flat result of a finer performance, but that’s it. I know you said: “Don’t ask me for”, but I will Snake… cos even if you know little, you can explain this little and will be useful too. Common King, give as a clue!
Well actually that's what I meant really. Not using it to compensate for volume fluctuations, (because when using mainly electronic computer-controlled instruments the volumes will be relatively constant, as opposed to say a bass guitar or singer) but rather for giving things more punch, power, snappyness.
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

So we were actually talking about the same thing King.

Ken, great post. There is much on it, so I’ll keep reading it several times, slowly till I get the points. Thanks very much.
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

RELEASE
The release time setting is also important because if it is too short, the compressor gain recovers too quickly with the result that there is an audible “pumping” or “breathing” as the gain rapidly changes (listen to some of the early 60s singles, paying particular attention to the cymbals, for an example of this). Conversely, if the release time is too long, then the gain may not have recovered by the time the next quiet sound comes along, so this may be suppressed more than is necessary.

Another problem that can arise if the attack and release times are set to be very short is that low frequency sounds from bass instruments tend to suffer distortion as the gain of the compressor tries to change during each cycle of the signal rather than following its overall envelope. To combat this, some designs incorporate and adjustable hold time which prevents the compressor from entering its release phase until the hold time has expired. A hold time of 20ms or so would be enough to prevent distortion on the lowest audible sounds, but a wider range is sometimes available allowing the user to experiment. The majority of compressor designs, however, include an in build, fixed hold time with no front panel control over this parameter.
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

MAKE UP GAIN
Because a compressor reduces the level of any signals that exceed the threshold, the output level will tend to be lower than the input, so an extra stage of “make up” gain is normally introduced in order that the output level can be matched to any subsequent pieces of equipment. There is invariably some form of meter to show how much gain reduction is taking place at any given time and further meters might well monitor both the input and output signal levels too.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
Nestor
Posts: 6676
Joined: Tue Mar 27, 2001 4:00 pm
Location: Fourth Dimension Paradise, Cloud Nine!

Post by Nestor »

SIDE-CHAIN CHARACTERISTICS
The way a compressor performs is largely due to that part of the circuitry that monitors the level of the signal being processed – the side chain. This is often due to the way in which the side-chain circuitry responds linearly or logarithmically to changes in input level and whether or not any filtering is employed to make it respond more to some frequencies than others. This latter measure is sometimes utilised to compensate for the fact that the human ear perceives different pitches as being of different loudness. If the side-chain is not frequency conscious to compensate for this effect, then the compressor may appear to be responding more to some frequencies than to others. It also makes an audible difference whether the side-chain measures the signal peaks or uses some form of averaging.

An RMS or averaging side-chain responds to sounds in much the same way as a human engineer would – the human hearing system tends to average out peaks in levels so that short duration sounds appear to be less loud sounding than longer duration sounds of the same level. From this description, you may rightly assume that very short, high level peaks might slip by an RMS compressor unchecked, even if a very fast attack time is set.

Peak sensing side-chain circuitry isn’t fooled by brief transients, it responds to even the shortest signal peaks, which means the compressor is much less likely to overshoot on drums or other percussive sounds unless a long attack time is chosen. A number of modern compressors have switchable RMS or “peak sensing”, and as a very general rule, peak works better on percussive sounds while RMS works best on non-percussive sounds.
*MUSIC* The most Powerful Language in the world! *INDEED*
User avatar
kensuguro
Posts: 4434
Joined: Sun Jul 08, 2001 4:00 pm
Location: BPM 60 to somewhere around 150
Contact:

Post by kensuguro »

Oh yeah, a quick note about bad dynamics on live instruments.. If the dynamics is bad, (and I mean not bad as in being too dynamic but as in uncontrolled) it's probably correct to think that you're doing something wrong. Atleast this is true for vocals. It's very usual for an untrained voice to greatly fluctuate in volume, especially between different vowels. So, don't use comp to hide these problems. The volume may be leveled, but a bad performance is bad no matter what. Instead, PRACTICE!
Post Reply