Scrap
09-15-2003, 04:43 PM
Originally posted by synthguy:
I thot I'd do a quick effects tutorial, since some people have been asking about this again. My old one vanished in the sands of time, sigh. Of course by now if you know me, you know what my version of "quick" means... ;)
Reverb
Everyone knows what this is, or should. It's a recreation of a space a sound is being played in, either a room, chamber or hall, tho sometimes they add what's referred to as a "plate." A plate reverb is a model of a particular effect created years ago using a large, real metal sheet suspended in oil, and still used to this day. Basic reverb parameters are:
Type: room, hall, chamber, plate, and sometimes others such as "tunnel," and should be self explanatory. The different types have different character of sound, just like the real ones do.
Decay: how long it takes the effect to decay to silence, usually in seconds.
Pre delay: most effects have this, and it's a short delay before the reverb sound starts, usually in milliseconds. Better effects have a "bloom," which is a build up of the reverb sound like attack on a synth, and is slower based on longer pre delay settings. The really good ones have a bloom or attack parameter.
Early reflection: a series of echoes based on the type of reverb you choose, since there are echoes from the walls and objects in these chambers, like seats and ceiling of a concert hall. Usually you want the level set fairly low on this, like 30% or so, since this will add some clutter to the effect. If you have the reverb level set to zero and early reflect at 100%, it sounds like a series of spitty sizzly versions of what you send into it. Sometimes there is a predelay and decay parameter for it too.
Diffusion: this determines how smooth the reverb is, or how "diffuse." Low settings make the reverb grainier, like early reflections, and higher settings make it smoother and lusher. You should play around with this parameter to see how it sounds, tho usually you want it set around 70-100% for traditional lush sound. Ensoniq reverb seems happiest between 65-85%.
Low/high decay: the rate to which the low or high frequencies in the reverb decay to silence. Some reverbs have this, since most chambers loose some tone in their decay, usually the highs. To be authentic, these should probably be set somewhere in the 10-40% range, the low decay the least since the reverb sound will get "thin" as the lows decay.
Surface type: better reverbs have a parameter for what the walls of the chamber are made of, as this will affect how the chamber sounds to a great extent. Curtains will soak up sound, especially the highs, while brick or marble will be the most diffuse and even sounding, maybe with a little early reflection "slap."
EQ: you know what this is. This will equalize the overall tone of the reverb. Brighter reverbs will sound more distinct in a "busy mix," with a lot of jamming from a lot of instruments going on. Emphasizing the lows is a bad idea, as lots of low frequencies will muddy up the mix and make the song sound thick and cluttered. Lows belong with the instruments, not the reverb, but you should be careful with trimming the lows as the reverb will sound thin. In many cases the EQ should be flat or disabled.
Level: the overall volume of the reverb sound. Sometimes this is a ratio, or a separate parameter. The Ratio will be the level of the reverb in respect to the sound sent to it, so usually you want this less than 50%. 50% is a VERY verby sound. If there's a separate level parameter for the sound going to it, setting this depends on how the reverb is set up. If the reverb is set up in your instrument like an effects box you're plugging into, you want the sound level at 100%. See how it's programmed with your patches for a guide on how to set levels. Usually the effect is set up with an effect level or ratio. If you have a separate effect unit, how you set this up depends on how you use it. If you're plugging into it, then you want your straight or "dry" sound set to 100%. If you're plugging it into an effects send on a mixer, you want the dry sound set to zero. Otherwise you'll be mixing in a mono version of your original signal, or adding another level of your dry signal on top of itself, and either way it'll mess up your sound.
There are a LOT more parameters on the better reverbs, but that's what manuals are for. ;)
To check out how a reverb sounds, I like to send drum and percussion sounds into it, something short like a snare drum, tom and woodblock so that I hear clearly what the effect is doing. Then I can see if anything needs changing. Usually if the reverb sounds good with drums, it also sounds good with other instruments. Not always, but it's a good way to check how the verb will sound.
Chorus
Chorus is a swirling, thickening type of effect, usually stereo. It sounds similar to detuning, but richer. On early synths, this was the first and only effect, but it did a lot to add to the sound. It's accomplished by adding a short delay of 10-20 milliseconds, and using an LFO to sweep this time up and down slightly, causing the waveforms to shift against each other slightly. It's made thicker with a feedback amount. Generally speaking the best chorus is analog, but there are some very good digital chorus, and they all add a sweet character to the sound. The usual parameters are:
Delay: often in milliseconds
Depth (or Width): the sweep width of the LFO
Rate: the speed of the LFO
Ratio: the amount of "dry" or original sound to the delayed sound
Width (or Spread): sometimes this is the same thing as Depth, but usualy this is the stereo spread of the effect
Feedback: the amount of chorused signal being fed back into the circuit
As with all things digital, there are sometimes many more parameters on the better effects. Ensoniq is very good at giving many parameters to their excellent effects, but sometimes very bad at explaining some of them, and such is the case with their chorus. :p
As a rule of thumb, a good rate for a chorus is one second or slower. With faster LFO rates you want less depth, with slower ones you want more. Faster LFO rates with higher depth give you a kind of wobbly sound, which can be useful for a special effect. A very slow rate with a very high depth seems to suit many sounds. Generally you want the ratio of "dry to wet" at 50-55%, and usually not too much feedback, 30% to zero. Zero still gives you a nice sound, tho a little light, but sometimes less is more. If you have a choice between sine and triangle on your chorus, use triangle, as it gives the most satisfying sound. A sinewave gives it kind of a "rollercoaster" feel, as the chorus swoops up to a peak, hesitates, and then swoops back down, and hesitates. This also applies to the flanger, phaser or any other modulation effect.
Flanger
This is similar to chorus, and uses the same principle, but the delay is very short, 10 milliseconds or so to zero. The sound is a little like chorus but very distinct, sharper, sometimes metallic. This is because the delay is so short that the harmonics themselves are shifting against each other. It was originally discovered by Phil Spector in the 50s I believe, and he stumbled onto it when he tried to double two tracks on separate reel to reel recorders. He put his thumb on one machine's flange to slow it down and line the tracks up, and this amazing swirling sound came from the speakers. It's been called flanging ever since. Because this is such a distinct sound, chorus is used much more, but flanging is still a sweet sound, useful in many musical contexts. Many times it's a stereo effect but not always, and generally analog flangers sound a little sweeter. The parameters are essentially the same as chorus, tho the delay time is usually half that. There is usually a parameter added to the flager effect.
Center: this is used to position the sweep of the LFO in a range of delay time, and usually is an arbitrary number, such as 0-100. Generally you want this parameter set to the middle of its range, like 50 on my 0-100 scale, but playing around with it will give you a different character to the flange effect. And sometimes you want to narrow the sweep of the LFO and use the center parameter to isolate a range of the flange that works well with the sound you're flanging.
Phase shifter or phaser
Phasers on stun! ;) Not that kind of phaser. If you don't know what phase shifting is, you might have to go to the wayback machine, or to the ballad in Six Degrees of Inner Turbulence, where John Petrucci uses it on his guitar in the intro. It's a swirling sound like the others, but much different because it uses a series of notch filters, so it sounds a little like a wah effect with chorus. For a long time, not many musicians used it as it had been used too much to produce some cheezy music, like the wah pedal, and it fell out of favor. But over the past few years both of them have been heard a lot more and used creatively. The parameters are similar to the chorus, but because they're filter based they have a few others thrown in.
Stages: this can also be called type or bands or similar, and refers to the number of notch filters used. The sound will change considerably with different numbers of stages used.
Center: like the flanger, this sets the center of the frequency of the filters the LFO will sweep up and down from. Like the flanger, sometimes you want to narrow the sweep of the LFO and use the center parameter to place the sweep in a range that works well with the sound you're phasing.
Resonance: because this is a filter based effect, resonance adds some sharpness to the sound as those filters have the frequencies beside the notches emphasized.
Generally you want to set the phaser up like the chorus or flanger, with a fairly slow sweep rate, tho you can get a nice wobbly effect with fast rates, tho you may have to tweak it a bit to get something that doesn't sound seasick. They tend to sound smoother or creamier with resonance set low, between 0-30%. And as usual, the analog phasers sound a little better than the digital ones, but the digital variety sure sound good these days!
Rotary or Leslie
This is an effect that simulates the sound of an organ speaker called a Leslie. It has a rotating horn on the top part playing the high frequencies, and usually a downward facing bass speaker with a rotating horn. It's a very nice sound, very distinct and unique. The rotation adds some nice chorusing to the sound, especially when it speeds up and slows down, as the top horn accelerates and slows down much quicker than the low rotor. I'll give some general descriptions of the parameters because the number of parameters varies a lot between Leslie effects.
There will be some sort of speed setting for the slow speed of the rotation and one for the fast, since a Leslie operates at two speeds and toggles between them depending on whether you want a slow chorus or fast vibrato. Sometimes there are separate speeds for the horn and rotor. Usually there is a different rate of spin up for the horn and rotor, as the rotor spins up slower than the horn. There is often a ratio setting the level of the horn relative to the rotor, and thus the highs and lows. It's best to have this set somewhere near even. Many give you a drive or distorion parameter to grunge up the sound, as Leslie distortion is particularly sweet, and essential for rock organ. Some effects have a width parameter like with the chorus, determining the stereo imaging of the effect. Some even let you place virtual microphones! There's nothing like the real thing, but there are some good effects out now.
Delay, multitap delay
Delay is the same thing as echo, a repeating of a signal you send to the effect. The basic delay parameters are
Delay time: sometimes in seconds or milliseconds. This is the amount of time between the sound you send into the effect and the echo.
Delay level: the volume of the echo.
Delay feedback: the amount of echo fed back into itself. The more feedback, the more the echo will repeat. Be careful not to overload it, as it will increase in volume till it's a grungy roar!
There may be a pan and EQ parameters as well, or a "roll off" or similar named parameter. This will decrease the brightness of the echoes, much like old tape delays would. Some delays will let you sync the delay time to midi, or let you modulate parameters with midi controllers.
Multitap delay is a lot more powerful. Each "tap" is a separate delay, usually with its own delay time, feedback, and often its own pan position. Many multitap delays are very powerful, allowing you to select different algorithms or even determine which tap feeds into other taps! This effect allows you to set up very complex rhythmic echoes, especially if you can control which taps feed into the others, as you can have some repeat their echoes while others don't, and with panning, you can have the echoes bouncing around the stereo spread. Roland is particularly extravagant, sometimes giving you 20 taps on their cheaper multieffects!! Yowzers!
Watch this space for developments as they occur! Whee! :D
I thot I'd do a quick effects tutorial, since some people have been asking about this again. My old one vanished in the sands of time, sigh. Of course by now if you know me, you know what my version of "quick" means... ;)
Reverb
Everyone knows what this is, or should. It's a recreation of a space a sound is being played in, either a room, chamber or hall, tho sometimes they add what's referred to as a "plate." A plate reverb is a model of a particular effect created years ago using a large, real metal sheet suspended in oil, and still used to this day. Basic reverb parameters are:
Type: room, hall, chamber, plate, and sometimes others such as "tunnel," and should be self explanatory. The different types have different character of sound, just like the real ones do.
Decay: how long it takes the effect to decay to silence, usually in seconds.
Pre delay: most effects have this, and it's a short delay before the reverb sound starts, usually in milliseconds. Better effects have a "bloom," which is a build up of the reverb sound like attack on a synth, and is slower based on longer pre delay settings. The really good ones have a bloom or attack parameter.
Early reflection: a series of echoes based on the type of reverb you choose, since there are echoes from the walls and objects in these chambers, like seats and ceiling of a concert hall. Usually you want the level set fairly low on this, like 30% or so, since this will add some clutter to the effect. If you have the reverb level set to zero and early reflect at 100%, it sounds like a series of spitty sizzly versions of what you send into it. Sometimes there is a predelay and decay parameter for it too.
Diffusion: this determines how smooth the reverb is, or how "diffuse." Low settings make the reverb grainier, like early reflections, and higher settings make it smoother and lusher. You should play around with this parameter to see how it sounds, tho usually you want it set around 70-100% for traditional lush sound. Ensoniq reverb seems happiest between 65-85%.
Low/high decay: the rate to which the low or high frequencies in the reverb decay to silence. Some reverbs have this, since most chambers loose some tone in their decay, usually the highs. To be authentic, these should probably be set somewhere in the 10-40% range, the low decay the least since the reverb sound will get "thin" as the lows decay.
Surface type: better reverbs have a parameter for what the walls of the chamber are made of, as this will affect how the chamber sounds to a great extent. Curtains will soak up sound, especially the highs, while brick or marble will be the most diffuse and even sounding, maybe with a little early reflection "slap."
EQ: you know what this is. This will equalize the overall tone of the reverb. Brighter reverbs will sound more distinct in a "busy mix," with a lot of jamming from a lot of instruments going on. Emphasizing the lows is a bad idea, as lots of low frequencies will muddy up the mix and make the song sound thick and cluttered. Lows belong with the instruments, not the reverb, but you should be careful with trimming the lows as the reverb will sound thin. In many cases the EQ should be flat or disabled.
Level: the overall volume of the reverb sound. Sometimes this is a ratio, or a separate parameter. The Ratio will be the level of the reverb in respect to the sound sent to it, so usually you want this less than 50%. 50% is a VERY verby sound. If there's a separate level parameter for the sound going to it, setting this depends on how the reverb is set up. If the reverb is set up in your instrument like an effects box you're plugging into, you want the sound level at 100%. See how it's programmed with your patches for a guide on how to set levels. Usually the effect is set up with an effect level or ratio. If you have a separate effect unit, how you set this up depends on how you use it. If you're plugging into it, then you want your straight or "dry" sound set to 100%. If you're plugging it into an effects send on a mixer, you want the dry sound set to zero. Otherwise you'll be mixing in a mono version of your original signal, or adding another level of your dry signal on top of itself, and either way it'll mess up your sound.
There are a LOT more parameters on the better reverbs, but that's what manuals are for. ;)
To check out how a reverb sounds, I like to send drum and percussion sounds into it, something short like a snare drum, tom and woodblock so that I hear clearly what the effect is doing. Then I can see if anything needs changing. Usually if the reverb sounds good with drums, it also sounds good with other instruments. Not always, but it's a good way to check how the verb will sound.
Chorus
Chorus is a swirling, thickening type of effect, usually stereo. It sounds similar to detuning, but richer. On early synths, this was the first and only effect, but it did a lot to add to the sound. It's accomplished by adding a short delay of 10-20 milliseconds, and using an LFO to sweep this time up and down slightly, causing the waveforms to shift against each other slightly. It's made thicker with a feedback amount. Generally speaking the best chorus is analog, but there are some very good digital chorus, and they all add a sweet character to the sound. The usual parameters are:
Delay: often in milliseconds
Depth (or Width): the sweep width of the LFO
Rate: the speed of the LFO
Ratio: the amount of "dry" or original sound to the delayed sound
Width (or Spread): sometimes this is the same thing as Depth, but usualy this is the stereo spread of the effect
Feedback: the amount of chorused signal being fed back into the circuit
As with all things digital, there are sometimes many more parameters on the better effects. Ensoniq is very good at giving many parameters to their excellent effects, but sometimes very bad at explaining some of them, and such is the case with their chorus. :p
As a rule of thumb, a good rate for a chorus is one second or slower. With faster LFO rates you want less depth, with slower ones you want more. Faster LFO rates with higher depth give you a kind of wobbly sound, which can be useful for a special effect. A very slow rate with a very high depth seems to suit many sounds. Generally you want the ratio of "dry to wet" at 50-55%, and usually not too much feedback, 30% to zero. Zero still gives you a nice sound, tho a little light, but sometimes less is more. If you have a choice between sine and triangle on your chorus, use triangle, as it gives the most satisfying sound. A sinewave gives it kind of a "rollercoaster" feel, as the chorus swoops up to a peak, hesitates, and then swoops back down, and hesitates. This also applies to the flanger, phaser or any other modulation effect.
Flanger
This is similar to chorus, and uses the same principle, but the delay is very short, 10 milliseconds or so to zero. The sound is a little like chorus but very distinct, sharper, sometimes metallic. This is because the delay is so short that the harmonics themselves are shifting against each other. It was originally discovered by Phil Spector in the 50s I believe, and he stumbled onto it when he tried to double two tracks on separate reel to reel recorders. He put his thumb on one machine's flange to slow it down and line the tracks up, and this amazing swirling sound came from the speakers. It's been called flanging ever since. Because this is such a distinct sound, chorus is used much more, but flanging is still a sweet sound, useful in many musical contexts. Many times it's a stereo effect but not always, and generally analog flangers sound a little sweeter. The parameters are essentially the same as chorus, tho the delay time is usually half that. There is usually a parameter added to the flager effect.
Center: this is used to position the sweep of the LFO in a range of delay time, and usually is an arbitrary number, such as 0-100. Generally you want this parameter set to the middle of its range, like 50 on my 0-100 scale, but playing around with it will give you a different character to the flange effect. And sometimes you want to narrow the sweep of the LFO and use the center parameter to isolate a range of the flange that works well with the sound you're flanging.
Phase shifter or phaser
Phasers on stun! ;) Not that kind of phaser. If you don't know what phase shifting is, you might have to go to the wayback machine, or to the ballad in Six Degrees of Inner Turbulence, where John Petrucci uses it on his guitar in the intro. It's a swirling sound like the others, but much different because it uses a series of notch filters, so it sounds a little like a wah effect with chorus. For a long time, not many musicians used it as it had been used too much to produce some cheezy music, like the wah pedal, and it fell out of favor. But over the past few years both of them have been heard a lot more and used creatively. The parameters are similar to the chorus, but because they're filter based they have a few others thrown in.
Stages: this can also be called type or bands or similar, and refers to the number of notch filters used. The sound will change considerably with different numbers of stages used.
Center: like the flanger, this sets the center of the frequency of the filters the LFO will sweep up and down from. Like the flanger, sometimes you want to narrow the sweep of the LFO and use the center parameter to place the sweep in a range that works well with the sound you're phasing.
Resonance: because this is a filter based effect, resonance adds some sharpness to the sound as those filters have the frequencies beside the notches emphasized.
Generally you want to set the phaser up like the chorus or flanger, with a fairly slow sweep rate, tho you can get a nice wobbly effect with fast rates, tho you may have to tweak it a bit to get something that doesn't sound seasick. They tend to sound smoother or creamier with resonance set low, between 0-30%. And as usual, the analog phasers sound a little better than the digital ones, but the digital variety sure sound good these days!
Rotary or Leslie
This is an effect that simulates the sound of an organ speaker called a Leslie. It has a rotating horn on the top part playing the high frequencies, and usually a downward facing bass speaker with a rotating horn. It's a very nice sound, very distinct and unique. The rotation adds some nice chorusing to the sound, especially when it speeds up and slows down, as the top horn accelerates and slows down much quicker than the low rotor. I'll give some general descriptions of the parameters because the number of parameters varies a lot between Leslie effects.
There will be some sort of speed setting for the slow speed of the rotation and one for the fast, since a Leslie operates at two speeds and toggles between them depending on whether you want a slow chorus or fast vibrato. Sometimes there are separate speeds for the horn and rotor. Usually there is a different rate of spin up for the horn and rotor, as the rotor spins up slower than the horn. There is often a ratio setting the level of the horn relative to the rotor, and thus the highs and lows. It's best to have this set somewhere near even. Many give you a drive or distorion parameter to grunge up the sound, as Leslie distortion is particularly sweet, and essential for rock organ. Some effects have a width parameter like with the chorus, determining the stereo imaging of the effect. Some even let you place virtual microphones! There's nothing like the real thing, but there are some good effects out now.
Delay, multitap delay
Delay is the same thing as echo, a repeating of a signal you send to the effect. The basic delay parameters are
Delay time: sometimes in seconds or milliseconds. This is the amount of time between the sound you send into the effect and the echo.
Delay level: the volume of the echo.
Delay feedback: the amount of echo fed back into itself. The more feedback, the more the echo will repeat. Be careful not to overload it, as it will increase in volume till it's a grungy roar!
There may be a pan and EQ parameters as well, or a "roll off" or similar named parameter. This will decrease the brightness of the echoes, much like old tape delays would. Some delays will let you sync the delay time to midi, or let you modulate parameters with midi controllers.
Multitap delay is a lot more powerful. Each "tap" is a separate delay, usually with its own delay time, feedback, and often its own pan position. Many multitap delays are very powerful, allowing you to select different algorithms or even determine which tap feeds into other taps! This effect allows you to set up very complex rhythmic echoes, especially if you can control which taps feed into the others, as you can have some repeat their echoes while others don't, and with panning, you can have the echoes bouncing around the stereo spread. Roland is particularly extravagant, sometimes giving you 20 taps on their cheaper multieffects!! Yowzers!
Watch this space for developments as they occur! Whee! :D