| Author
|
Synth programming tips
|
jivamukti
IsraTrance Junior Member
Started Topics :
21
Posts :
342
Posted : Mar 18, 2006 19:30
|
I found quite a few synth programming tips in hardware synth manuals, especially those of Waldorf and Clavia. In addition I found a tips & tricks doc for Alesis Andromeda, but I'm sure they can be applied to other synths too, just like the other tips here. Check it out here: http://www.wohmart.com/a6/
So here's the result of reading over 3000 pages:
Alesis Andromeda:
----------
A popular trick in the 80’s was to record the reverb with the tape flipped over (ie. reverb sound playing backwards), so it would play backwards in the mix. The reverse reverb is a useful effect for percussive sounds, adding space without washing out the instrument.
Boosting reverb's bassrange can add warmth or at extreme settings rumble.
A bit of pre-delay in reverb can make certain instruments sound bigger.
Try using lower reverb density settings on sustained sounds, and higher density for percussive sounds.
Gating is the process which abruptly cuts off the reverb’s decay for a more “choppy”
sound. This effect was very popular on drums in the ‘80s because it makes them
sound HUGE. It is achieved by dropping the level of the signal very rapidly after the
initial attack making a short, sharp sound. Try also reversed gated reverb.
If you want to add sustain to a program without cluttering up the mix with reverb from the initial filter attack and initial transients, set the reverb LPF around 2 or 4kHz.
Korg MS-2000
-----
Use square osc for woodwinds.
Use tri osc for mellow bass.
Nord Lead
-----
Notch filter isn't very useful alone because allows too much frequencies to pass thru.
Instead use it in combination with other filters.
Notch filter with lowpass filter above it are suitable for example for soft & clear pads (with low Q probably).
Use triangle lfo for vibrato.
Use random lfo for effects.
Triangle lfo -> filter = autowahwah.
Lfo -> synced osc's pitch = pulsating change.
Triangle lfo -> amp = tremolo.
Pitch mod envelope -> synced osc's pitch is good for sweeps.
Pitch mod envelope -> frequency modified osc's pitch = fx sweeps.
Use (square) lfo to pwm squ osc = choruslike effect for strings & pads.
For resonant sounds use 12dB cutoff, for others 24dB.
Put high reso, then vary cutoff for example with envelope.
Use AD (attack-decay) envelope to modulate fm amount or synced osc pitch to create a different timbre for the sound's attack.
Nord Lead 2
-----
For classic fm sounds use sine on carrier osc and tri on modulator osc.
Nord Lead 2x
------
Suitable uses for osc pitch keyb track off:
1) sync - then the pitch is determined the by the master osc anyway;
2) with RM and FM to get varying timbre across keyboard;
3) for FX and percussion sounds.
Use ring mod for bell and metallic sounds.
Using keyb track on filter cutoff prevents sound from becoming muddy when played in higher octaves (without keyb track cutoff is constant while timbre changes the more the higher you play).
Nord Lead 3
------
Hold down SHIFT and turn the OSC MIX knob to replace the Oscillator 2 output signal in the mix with noise. Adjust the amount of noise with the OSC MIX knob. The more towards the Osc2 value you turn the OSC MIX knob, the more noise you’ll get in the mix. This may seem like a strange function but it’s
actually quite useful for adding “breath” noise in the attack phase of wind instrument simulations, for example. This feature is extremely useful in FM sounds where you mainly use the signal from the Oscil-
lator 1 section. To get noise only in the attack phase of a sound, you could assign the Osc Mix parameter to the Modulation Envelope, for example.
Roland SH-2
-----
PWM has more effect on low pitched sounds than high pitched ones.
Roland JP-8
--------
Use less steep filter slope for acoustic-modelled sounds such as strings because acoustic instruments tend to have richer timbre.
Portamento is especially effective for such sounds as whistle and trombone.
Roland SH 09
-------
When you have more than one oscillator, some oscillators at some octave intervals can be used to produce a sound that sounds like it only has one osc but with a thicker sound than with only one osc. An example of this is a square osc at -1 oct mixed with another waveform at 0 oct. The sub osc actually adopts much of the timbral character of the main osc in what is called "waveform staircasing". Many types of sound benefit from this. Experiment.
Inverted filter envelopes are useful for creating plucked or strong, punching synth sounds with lots of dynamic motion. Experiment with different waveforms, amounts of cutoff freq starting point and amounts of envelope modulation.
Set resonance to some other setting than "off" to emphasise filter movement for example when doing filter sweeps or
in patches that use filter envelope.
Try using a resonant filter that selfoscillates as an additional osc. (SH-09 uses this technique in a number
of patches such as an organ and a drum.) The sound of the selfoscillating filter changes slightly depending on the
pitch of the notes you play.
Bypassing volume envelope makes sense when creating wind, surf and sample & hold sounds.
Setting vol env's attack 0 will cause a click in the sound. Usually this is unwanted, but for some sounds such as vibraphone and guitar it is essential.
Short portamento is useful for melodies in which individual notes have to be articulated cleary, while longer portamento is useful for single notes and effects.
Roland SH-1000
-------
Sinewave tremolo is the usual tremolo whereas saw tremolo is mandolin tremolo.
Waldorf Microwave:
--------
About osc semitone:
Most organ sounds include a quint, therefore one oscillator's semitone parameter must be set to +7. There are also many lead sounds with an interval, e.g. a quart (+5 semitones). When making ring modulated sounds, try
to use +11 (or other disharmonic values) for the setting.
About finetune:
The audible result of detuning oscillators is a flanging. Use a positive setting for one oscillator and an equivalent negative setting for the other. A low value of ±1 results in a slow and soft flange effect. Mid-
ranged settings of ±5 are optimal for pads and other fat sounding programs. High values of ±12 or above will give a strong detune that can be used for accordeons or effect sounds.
About osc sync:
Try to use a random semitone setting while Octave is at +2.
Waldorf Pulse:
--------
• The lower the input signal, the greater the effect the filter has on the overall sound.
If you want a heavily filtered sound, set the oscillators' volume parameters to low
values in the mixer. On the other hand, high oscillator levels produce purer sounds.
• Try using an LFO to modulate the panorama position. This produces interesting
stereo effects, especially at high LFO frequencies.
• Change the relative pitches of the oscillators at musical intervals. Thirds (mediant),
fifths (dominant) and sevenths (subtonic) are suitable for this application.
• Self-oscillation of the filter at high resonance values produces sounds that are great for soloing.
• Modulate the oscillator's pitch drastically so that the upper frequency threshold is
exceeded. This will produce interesting results.
• Program a pitch modulation in musical intervals. The section entitled "Modulations"
contains an assignment table listing the Mod Amount parameters and the corresponding semitone intervals. For example, you can modulate from a major chord to a minor chord via the mod wheel.
• Assign a Keytrack modulation to the LFO speed. It should modulate the LFO in
proportion to the incoming note. You can thus achieve even frequency fluctuations
over a wide range of your keyboard.
• Modulate the pitch of an oscillator via Keytrack. Set the modulation "Amount" so
that this value is numerically equal to the "Tune" value of the oscillator, but make
one value positive and the other negative. This generates frequency fluctuations that
oscillate at the same speed throughout the range of the keyboard. Experiment with
different Amount values by alternating low and high notes.
The option of overdriving the signal vastly enhances the variety of sounds the Pulse
can produce. The Pulse is an analog device, so we can't give you a precise value
when a signal will be overdriven. As the volume increases, the signal becomes
slightly saturated and flows seamlessly into total disortion.
An overdriven signal has a richer sound, as overtones are added to the clean signal.
This is especially interesting in conjunction with sawtooth and triangular waveshapes,
as square waveshapes are inherently very similar in structure to other overdriven
waveshapes.
Distortion is most audible when you drastically detune several oscillators in relation
to each other, especially over a range of several octaves. This effect is even more
interesting when you tune the pitch of one oscillator a semitone or several semitones
above or below the true octave. |
|
|
jivamukti
IsraTrance Junior Member
Started Topics :
21
Posts :
342
Posted : Mar 18, 2006 19:32
|
Waldorf Rack Attack
-----------
(Note that these are mainly percussion programming tips.)
There are no natural musical instruments that generate a pure sine wave, although the instrument that would come closest is the tuning fork. Therefore, the sine wave will always sound slightly artificial to
the ear. However, the sine wave can be used to emphasize a certain harmonic while other oscillators are playing more complex waveforms, or as a source for frequency modulation. Furthermore, a sine
wave is perfect to give a bass drum its characteristic low frequency punch.
The reason why the triangle wave is so popular in classic synthesizers is because it is easier for analog circuitry to generate than a sine wave but it can still be used for the same purposes. It could also
behave as a sub-oscillator wave, used to emphasize certain frequencies or to frequency modulate other oscillators.
White Noise is the well-known noise form that contains all frequencies with the same amplitude. It can be
coloured by defeating or raising different frequencies.
S&H is perfect both for industrial-type FX sounds and as an FM source.
Lead and Solo sounds will sound interesting when you set one oscillator to a fourth (+5 semitones) relative to
the other.
• The triangle waveform consists of odd harmonics with very low magnitudes. It is
perfect for nearly all drum and percussion sounds.
• The sine waveform contains only the fundamental frequency, and has no harmonics at all. With a sine wave you can create clean bass drums and snares, and it is also perfect for FM.
• The square waveform produces a hollow/metallic sound and is perfect for bass drums and snare drums.
Triangle waves, sine waves and noise are especially suited for FM.
If you use noise as FM source, the sound will become more tonal when you use higher FM
settings.
To create a vibrato, set Oscillator 2 to a deep pitched triangle waveform and use very low FM settings. Playing this sound at low octaves creates a wobble effect.
Since ring modulation generates disharmonic components, it can be used to add metallic distorted sound characteristics. This is useful when generating crashes or cowbells. Please note that in a complex waveform all harmonic components behave like interacting sine waves, resulting in a wide spectral
range of the ring modulated sound.
Ring modulation can result in unwanted low frequencies when the pitches of Oscillators 1 and 2 don’t differ very much. This is logical because, for example, when you use one oscillator set to 100 Hz and the second set to 101 Hz, the resulting ring modulation is 201 Hz and 1 Hz, and 1 Hz is very low.
• Lowpass: With a low pass filter you can shape bass drums or snare drums.
• Highpass: You can cut the bass and mid range of high frequency drum sounds like hi-hats or crashes.
• Bandpass: Use these filter types when programming effect and percussion-like sounds.
• Notch: Use this filter type for programming effect sounds. Resonance has no great influence because it raises the
frequencies that the notch filter dampens. You can hear a little effect of phase changes, but not to any great degree.
Filter resonance
When the setting is raised to values above 113, the filter starts to self-oscillate, generating a pure sine wave. This feature can be used to create analog-style effects and percussion-like electronic toms,
kicks, zaps etc.
Programming Drum Sounds
To understand how to create drum sounds, you should know a little about how the classic drum machines worked. The following sections give some insights how particular sounds were built and information how to achieve similar results on the Waldorf Rack Attack.
Roland TR-808 Bass Drum
On the Roland TR-808, this sound was made by one filter with a high resonance setting, triggered by
a short impulse. Two controls were provided to adjust the bass drum: "Tone" was used to set the pitch
by changing the filter's cutoff frequency, and "Decay" was used to set its resonance, which in turn
controlled the decay rate.
On the RackAttack, you could use the filter´s self-oscillation by triggering it with a short noise impulse
made by the second oscillator, whose volume you can control by an envelope. But a better way is to
use Oscillator 1 playing a sine wave, and by changing the initial click impulse by means of the filter.
Roland TR-909 Bass Drum
The Roland TR-909 used an oscillator and a noisy click, controlled by three envelopes, to create a
bass drum sound. The oscillator played a sine wave whose pitch was controlled by an envelope and
the "Tune" control. The rate of the envelope’s decay was not adjustable.
This oscillator signal was routed to an amplifier with an envelope whose "Decay" parameter adjusted
the decay rate of the envelope. The second part of the bass drum sound was made with a short
impulse and a low pass filtered noise generator, both summed and routed into another envelope that
controlled their output volume. The "Attack" parameter controlled the overall pulse/noise level, and
the decay rate of the envelope was not adjustable.
On the RackAttack, you can make this sound as follows: Oscillator 1 plays a sine wave, and Envelope
2 is used to modulate its pitch. This means that the pitch of the oscillator becomes higher or lower
depending on the setting of the Decay parameter of Envelope 2, but this slight variation doesn't affect
the drum sound once it has been set up.
The noise of the impulse can be ignored, because it is low pass filtered anyway. But how do we
create an impulse with the RackAttack? The answer is simply to use a square wave with a very low
pitch setting for Oscillator 2, and to control its level with a very short envelope. Now we have an
impulse. This impulse is low pass filtered afterwards with a slightly resonating filter, preferrably set to
around 5000Hz with a resonance of around 18%.
With the Oscillator 1 Pitch and Pitch Env controls you can adjust the sound of the bass drum, while
Envelope 2 Decay controls its length.
Simmons SDS-5 Bass Drum
The Simmons SDS-5 bass drum consists of an oscillator and a noise generator, both routed into a low
pass filter and an amplifier. An envelope controls the oscillator pitch, the filter cutoff, and the
amplifier volume. The envelope has a decay shape that is in-between exponential and linear.
The oscillator plays a triangle wave whose pitch is controlled by a "Tune" control and a "Bend"
parameter that controls the influence of the amp envelope to oscillator pitch.
A "Noise-Tone" parameter controls the mix between the oscillator and the noise generator.
A "Noise" parameter controls the filter cutoff. (Very confusing, isn't it?)
A "Decay" parameter controls the envelope decay rate.
A "Click-Drum" parameter controls the most important aspect of the Simmons drums: the mix
between the original signal from the pad trigger microphone and the triggered drum sound.
On the RackAttack, you can make this sound as follows:
Oscillator 1 plays a triangle or sine wave pitched at around 30Hz, and Envelope 2 is used to
modulate its pitch. Use the "Vel" control to simulate the velocity-dependent pitch bend amount that
you would find on the SDS-5. You can simulate the click by setting FM Env to a medium value, with
Envelope 1 set to a very short decay. Oscillator 2 generates noise, and the pitch is set to center. The
Filter Cutoff can vary between 100Hz and 5000Hz, and Vel should be set at 25% or so. Filter
Resonance should be set to 10%. Envelope 2 should be set to an almost linear shape. Use Osc 1 and
Osc 2 Level to adjust the mix of tone and noise, and use Osc 1 FM Env to vary the click strength.
|
|
|
|
jivamukti
IsraTrance Junior Member
Started Topics :
21
Posts :
342
Posted : Mar 18, 2006 19:34
|
Roland TR-808 Snare Drum
On the Roland TR-808, the snare drum was made of two resonating filters and a noise generator with
high pass filtering. The "Tone" parameter controlled the output mix from the first and the second
filters, while "Snappy" controlled the volume of the noise generator. The noise generator was routed
through a separate envelope and a high pass filter.
On the RackAttack, you can make this sound as follows:
Oscillator 1 plays a sine wave at around 150Hz, and you can use a little FM to disturb the periodic
character of the sine wave. This trick makes the oscillator sound thicker, almost as if two oscillators
were running at once.
Oscillator 2 generates noise, and you should use Pitch to high pass filter it. In the Mixer, turn up Osc
1 to 50% and Osc 2 Env to 50%, set to Envelope 1. Set Envelope 1 to a shorter decay phase than
Envelope 2. Use the filter with a low pass setting and add a little resonance to emphasize the high
frequency range.
Roland TR-909 Snare Drum
The TR-909 Snare Drum was made with two oscillators and two filters for noise. The two oscillators
started in phase but were slightly detuned, and one of the oscillators was modulated a bit by a pitch
envelope. The "Tune" parameter controlled the basic pitch of the two oscillators. The noise was split
in two parts: there is always some low pass filtered noise during the whole snare drum sound, while a
high pass filtered sound is routed through another envelope whose level can be controlled by the
"Snappy" parameter.
One RackAttack sound doesn't feature as many different modules as the TR-909 snare drum had. One
solution can be to use two sounds, one emulating the first oscillator and the low pass filtered noise,
and the other emulating the second oscillator plus the high pass filtered noise. You will have to play
the two simultaneously in your track, but this shouldn't be a problem because the Attack has sample-
exact timing.
However, you can re-create the TR-909 snare drum with just one Attack sound instead, by doing the
following:
Set up Oscillator 1 to play a sine wave, modulate its pitch slightly with Envelope 2, and add a little
FM to it - around 0.1 to 0.5%. When you set the second oscillator to produce noise, you will hear that
the sine wave gets smeared, which means that you are not hearing an exact tone any more. This
already sounds very close to two slightly detuned oscillators and a low pass filtered noise. Now you
only need the "Snappy" part, which is added simply by using Envelope 1 to modulate Oscillator 2's
mix level. You can high pass filter the noise with the Pitch control, but in fact the result is already
quite similar without doing so. If you want a little more punch, use the Drive control carefully until
you can hear a slight distortion at the beginning of the sound.
Another variation can be heard in the sound library that comes with the RackAttack. This one uses a
very low noise signal level that is boosted greatly behind the high pass filter. The reason for this is that
Oscillator 2 plays the tone of the snare drum while the high pass filter dampens this tone heavily. To
raise it back up to a good volume, Drive boosts it to a normal level.
Simmons SDS-5 Snare Drum
The Simmons SDS-5 Snare Drum module was laid out identically to the Bass Drum module.
However, a number of parameters were set in a different way internally to create snare drum sounds.
When you want to create Simmons snare drum sounds on the RackAttack, just keep in mind that you
should use a very short envelope to frequency modulate the first oscillator, set the envelopes to
almost linear shapes, and use the "Vel" control for all envelope modulations.
TR-808 Side Stick
The TR-808 Side Stick (called RS on the 808, which stands for Rim Shot) sound is very tricky:
although it consists of only two oscillators running through an amplifier and a high pass filter, the
sound is very complex. This comes from the fact that one oscillator seems to “cut” the other oscillator
and that the VCA is used to add high harmonics. How Roland did it is something only they and
maybe a handful of people know. If you happen to be one of these people, let us know!
If you want a sound of this type from the RackAttack, use the representative sound from the library
instead of trying to simulate it on your own. Look at the parameters and try to find out why it sounds
quite close. A couple of hints: Crack is used with a very high frequency setting doing amplitude
modulation on the oscillators’ summed signal, and Drive is used to add further harmonics by
distorting the signal.
TR-909 Side Stick
The TR-909 Side Stick is made of 3 resonating band pass filters that are triggered by a short impulse.
Behind the band pass filter cluster there is a distortion unit, followed by a VCA with an envelope and
a high pass filter.
Its specific sound comes from the cutoff frequencies, the resonance, and the volumes of the trigger
impulses of the three band passes. These settings are:
* 500Hz, 20ms decay, full volume
* 222Hz, 45ms decay, half volume
* 1000Hz, 5ms decay, full volume
Now, the RackAttack doesn't have three band passes plus a high pass filter, but there's a way to
simulate the architecture with the Attack.
What produces a resonating band pass filter? Nothing more than a sine wave. So, why not just use
two oscillators producing two sine waves, plus a high pass filter that uses the lowest frequency setting
as the third sine wave generator. Thus the filter will include both oscillator signals and add its own
resonance to the sum.
The high pass filter is therefore set to a lower value, with a resonance of 127. Oscillator 1 produces
the 500Hz sine wave, while Oscillator 2 is set to a 1000Hz sine wave, but is controlled by a very
short Envelope 1 set to around 75% to produce the 5ms signal. The fact that oscillator 1 plays longer
than 20ms can be ignored, because it's not that noticeable. Don't add it with full volume, however;
set it only to a level of around 25%. This comes into play because there is an additional high pass
filter on the original TR-909 Side Stick that dampens lower frequencies.
Finally, add a good amount of Drive (around 30dB) to the signal, set Envelope 2 Decay to 45ms, and
you will have the sharp attack of the original sound.
TR-909 Hand Claps
TR-909 hand claps are made using the same signal routing as in the original TR-808. However, due to
the differing parts and internal parameter settings that were used in the TR-909, the TR-909's hand
clap sounded different. Essentially, the "Crack" (or as Roland called it "Sawtooth Envelope") was
clearer, and the reverb effect was longer.
|
|
|
|
jivamukti
IsraTrance Junior Member
Started Topics :
21
Posts :
342
Posted : Mar 18, 2006 19:35
|
Hi-hats
For hi-hats, we don't use references to classic drum machines, although there is a quite good
emulation in the TR-808 set included in the sound library. Hi-hats can be made in various ways:
The simplest method is to use the built-in samples of the Attack. However, those are provided just in
case you don't have time to "synthesize" a good-sounding hi-hat.
If you like a really "vintage" sound, use a high pass filtered noise. This gives the very archaic hi-hat
sound that was used by many drum machine companies for years. One of the last examples of this
sound was the good old Roland CR-78.
If you want more sophisticated results, use FM. Don't use noise as FM source, but instead use a sine
or triangle waveform with a very high pitch. The modulated oscillator can be set either to square or
sine. The FM of the RackAttack has a maximum amount of around 8 waveform cycles, which results
in heavy but tonal noise. When you don't use a static FM but change the amount by an envelope, the
sound gets really exciting. The noisy FM effect changes over time, resulting in a very lively hi-hat
sound. You will probably have to experiment with the settings of Oscillator 2 Pitch and FM Env, but
the results are very much worth the work.
A good rule of thumb is to start with the Open Hi-hat sound, and copy that sound to the location for
the Closed Hi-hat. Making a hi-hat sound shorter almost always succeeds, but making a short hi-hat
longer may result in an unwanted characteristics. Also, don't forget to set the sounds to the same XOR
Group so that they cut off one another.
Cymbals
With cymbals, the situation is similar to hi-hats. A sample is provided for an authentic crash cymbal,
though you can achieve more interesting and unique results using filtered noise or FM.
Ride cymbals are more difficult to create, and their sound is so special that you might wish to use a
good sampler or sample player to generate those sounds. If you want to create your own ride cymbal
sounds anyway, you might come up with interesting results using FM and ring modulation.
Toms
Tonal percussion instruments can be created easily. Just set one oscillator to produce a sine or a
triangle wave, modulate its pitch by an envelope, and set up the second oscillator to create either the
attack noise or the resonance skin. When you want to do the latter, just copy the settings of the first
oscillator and change the pitch or the envelope depth a little.
Also, it might be interesting to remove a little of the "tone" from Oscillator 1 by applying FM from the
second oscillator producing noise. Note that a short envelope used for FM creates astounding drum
stick hit sounds. Furthermore, you can high pass filter the result to get more punch and less tone into
the sound.
Congas
Congas can be made by using a sine wave oscillator, with a very short envelope controlling the FM
amount of Oscillator 2 producing noise. This, together with a medium fast attack on Envelope 2,
creates very authentic conga sounds.
Muted or slapped congas can be made by increasing the basic FM amount a little and using a high
pass filter to dampen the "tone".
Shakers and Maracas
Both are made with noise, either unfiltered or used to frequency modulate Oscillator 1 in order to
create strong colorization. A high pass filter can be applied to remove some low end.
The difference between shakers and maracas from the synthesist's viewpoint is that a shaker has a
longer attack and decay phase than maracas. Of course the sound depends a great deal on how you
play, so don't forget to set up velocity-based changes to the amplifier.
Claves and Woodblock
Claves and woodblock sounds are also very similar. They both consist of very short sine or triangle
waveforms. A woodblock is lower in frequency, and you can add the second oscillator to produce a
different frequency. Claves should be made with only one sine oscillator and a very short envelope.
TR-808 Cowbell
You are waiting for this one, aren't you? The TR-808 Cowbell is made of two square oscillators, one
oscillating at 540Hz, the other oscillating at 800Hz. The attack phase of the envelope is emphasized
heavily to create the strong click. Afterwards, the summed signal is sent through a band pass filter and
an envelope that stops abruptly.
A funny side note: on the TR-808, the square oscillators were the same that were used for the cymbal
and hi-hat sounds. However, those sounds used a cluster of six detuned square oscillators with
different band pass and high pass filter settings.
Tips & Tricks
• For a classic “Filter Trigger” sound, you don’t need to actually use a filter. You can also use
a sine oscillator that is pitch modulated by an envelope. This frees up the filter and you
can use it to perform other tasks, e.g. filtering noise for a snare drum sound or the like. |
|
|
|
jivamukti
IsraTrance Junior Member
Started Topics :
21
Posts :
342
Posted : Mar 18, 2006 19:35
|
Waldorf Q
-------
When playing multitimbrally, you can set up nice layer sounds with Transpose and Detune. Activate two Instruments.
Then just transpose one Instrument by one octave and you got a fat layer sound. Or set them both to the same octave
and set Detune of one Instrument to –05 and the other to +05.
Use legato for lead sounds.
Use triangle osc as sub osc or to emphasize frequencies.
Use ultralow-pitch osc with ring modulation to create rhythmic textures.
Very interesting is the use of Noise as FM Source on a high pitched oscillator playing a sine or
triangle waveform. With low FM amount, the oscillator starts to sound dirty or airy while
higher amounts create a coloured noise similar to a filter with high resonance. The nice thing is
that the filters are then still free for other purposes.
Use tri osc for soft pads and flutes.
Use sine osc for ring modulation.
About pwm osc:
A pulse width of 50% can be used for flute sounds or very hollow bass sounds.
A pulse width of around 30% can be used for e-piano or fat bass sounds.
A pulse width of around 10% is interesting for Clavinet sounds.
To create a thick oscillator sound, use a triangular LFO as PWM Source with full PWM and a
Pulsewidth of around 80. This basic setting is useful for very big string and lead sounds. When
you play different notes on the keyboard, you might notice that bass notes sound more detuned
than higher notes. To avoid this behaviour, set Keytrack of the used LFO to a positive setting
between 50% and 100%.
To create a Bass sound, use a decaying Envelope as PWM Source with negative PWM and a
Pulsewidth of around 80…127. This results in a fat attack phase, especially when only one
oscillator is used for the Bass.
When you want to create a sound with a wave sweep, you should roughly set the startpoint
(Pulsewidth) onto the desired wave, before you apply any modulation. This helps you to find
the basic waveform where all modulations start from. Note that you can apply unipolar and
bipolar modulation sources . For example, set Pulsewidth to 64, which is almost the middle of
the wavetable and apply a slow running LFO to PWM Source to sweep through the whole
wavetable.
About osc sync:
Use Sync for Lead or Solo sounds. Set Oscillator 2 to play one octave and 7 semitones higher, apply an envelope to its pitch with positive amount and you get a screaming sync sound.
Sync can also be very interesting on arpeggio sounds. Apply a slow clocked LFO to Oscillator 2 pitch and the arpeggio starts to move.
To create a sound whose pitch glides in, set osc PitchMod Source to a decaying Envelope with PitchMod set to around –25.
About glide:
Low values in glide speed will give a short glide time in the range of milliseconds that
gives a special character to the sound. High values will result in a long glide time up to several seconds which can be useful for solo and effect sounds.
Ring Modulation can result in unwanted low frequencies when the pitches of oscillator 1 and 2 don’t differ very much. This is logical because when you use i.e. one oscillator set to 100Hz and the second set to 101Hz, the resulting ring modulation is 201Hz and 1Hz. And 1Hz is very low.
Ring Modulation can be very interesting with a slow pitch modulation that is applied to one oscillator, i.e. a decaying Envelope. This creates spacey effect sounds.
For an E-Piano sound, you should use Ring Modulation where one high pitched oscillator’s Keytrack is lowered to i.e. 50%.
If you turn the pitch of one oscillator down very far, you can get a very similar effect to Amplitude modulation. Use this for sounds with a periodic element if you like.
Filter selfoscillation can be used to create analog-style effects and percussion-like electronic toms, kicks,
zaps etc.
About filter envelope amount and filter velocity amount:
You can create interesting effects by setting one parameter to a positive amount and the other to a negative.
Use bandpass filter for effects and percussion sounds.
Use notch filter for effect sounds.
How to use comb filter as chorus:
1. Select the Comb+ filter type for the desired filter.
2. Set Cutoff to a medium value.
3. Set Resonance, Drive, Keytrack and any other modulation to 0.
4. Select an LFO as Cutoff Mod Source.
5. Set Cutoff Mod to a medium positive or negative amount.
6. Adjust the LFO to generate a slow triangle oscillation.
7. Fine tune Cutoff, Cutoff Mod and LFO Speed until you get the desired Chorus effect.
You might ask why you should use a Comb filter type to produce the Chorus instead of using the FX section of the Q. Here are some reasons:
• The FX sections are still free for other purposes. Also, in Multi Mode, the Comb filter works no matter on which Instrument you use it.
• The Comb filter works per voice rather than per Instrument. This means that each voice has its own Chorus and therefore could generate a different Chorus effect. You can easily hear that when you change LFO Keytrack or Filter Keytrack. Then each voice sounds
with a different Chorus speed or frequency.
• You have full control over all parameters that make up the Chorus effect over MIDI by
using Control messages.
How to use comb filter as a flanger:
1. Adjust the filter as described in the Chorus paragraph above.
2. Turn up Resonance to create the typical Flanger sound.
3. Eventually lower LFO Speed to create slow Flanger sweeps.
4. Select either Comb+ or Comb- to change the character of the Flanger.
Set up Cutoff Mod Source to Modwheel to change the Flanger effect manually.
Use the Filter Envelope with positive or negative amount to create a “one time sweep” that is triggered at Note On.
Use Filter Keytrack with an amount of 100% to create really different flangings depending on the Note you play.
How to use comb filter for physical modelling:
A Comb filter is one of the most basic building blocks of the so-called “Physical Modeling”, which is an algorithmic description of an acoustic instrument. The Q is definitely not meant to replace a good physical modeling synthesizer so you shouldn’t expect to get a perfect emulation of any acoustic instrument like a trumpet or a violin out of the Q. However, the Comb filter enables you to create abstractions of those sounds that can be played expressively. To do so, you have to understand that the Comb filter is the part that actually sets the pitch of the resulting sound, the oscillators or the noise generator are only used to trigger the Comb filter somehow. When you play with the Comb filter, you can hear that it emphasizes or attenuates certain frequencies depending on the setting of Cutoff and Resonance.
When you set Resonance to a very high value, you will notice that the Comb filter oscillates heavily, and that’s the key to use it as a tone generator. When you set Filter Keytrack to +100%, you can play the Comb filter musically.
Now you need to know how to set up Cutoff. Filter Cutoff is scaled in semitone steps and you only have to find the Cutoff value where the filter oscillates with standard pitch. The following table gives
you these settings:
Oscillator Comb+ Comb-
64’ 11 23
32’ 23 35
16’ 35 47
8’ 47 59
4’ 59 71
2’ 71 83
If you need other tunings, just keep in mind that Cutoff is changed in semitones, so if you increment
Cutoff by 12, the filter oscillates one octave higher and vice versa.
To set up the Comb filter to produce a tone:
1. Select either Comb+ to produce a string-like tone or Comb- to produce a tube-like tone.
2. Turn up Resonance to around 114 to 127.
3. Turn up Keytrack to +100% to have Filter Cutoff change with well-tempered scaling.
4. Set Cutoff to 23, 35, 47 or 59 to have it oscillating in the same pitch as the oscillators are usually tuned. Please note that Comb+ oscillates one octave higher than Comb-.
5. Feed the Comb filter with a signal of any kind. The incoming signal’s pitch isn’t important for the resulting tuning, but both pitch and timbre of the input signal affect the timbre of the Comb filter output sound greatly.
Try to experiment with all kinds of input signals, i.e. different oscillator settings, noise, combinations of both and the like.
Also try to set up envelopes to control the levels of the input signal, i.e., use a short percussive signal that only triggers the comb filter shortly. The filter then oscillates to produce the final tone.
--------------
That's it.  More to follow later, if I find anything useful.
Manuals I read:
Alesis Andromeda
Korg MS-20
Korg PolySix
Nord Lead series
Nord Modular series
Oberheim OB8
Oberheim Matrix 1000
Roland Juno series
Roland JP series
Roland SH series
Roland V-Synth
+ some digital synth manuals
Sequential Circuits Pro-One
Sequential Circuits Prophet 5
Waldorf Q series
Waldorf Pulse
waldorf Rack Attack
Now go and make some good music. |
|
|
|
texmex
Started Topics :
5
Posts :
189
Posted : Mar 19, 2006 02:15
|
Hey thanks for the tips! It's great to learn new - especially if you happen to have one of those synths 
In particular I found those drum synthesis instructions very interesting.
My 0.02 cents:
You can use FM modulation to create filtersweep-like just by using following configuration:
osc 1(sin) moderately FM-modulated with osc2(tri)with lfo/modwheel controlling the amount of fm-modulation. This might save you a filter for other purposes.
Adding unison notes can be used to create fat sounds, but this works best for pads and other sounds with long body. Instead, use separate sound for the attack/transient part and for the body of the sound. This gives more control.
For example, you can create a sound with snappy sawtooth sound with fast amp/filter decay and on top of that add a triangle/pulse/whatever body. Both can use different filtering and modulations. Then you can make modwheel affect only the vibrato of the body and amp/filter env amount of the snappy part.
For example with MicroKorg (based on ms2000), you have dual timbral sound (mk has total 4 voices, so in layer mode you have only 2). But you have 4 lfo's, 4 envs (2x amp/filter) and 8x modulation (4x for each). Plus chorus/flanger/phaser/essemble and tempo sync'd stereo/cross/l+r delay.
Also remember to try out the digioscillator 26 in MicroKorg - instant house chord pleasure 
And vocoders can be used to create rythmic textures when drum loops are used as modulator. Just add tempo synced feedback delay between and you're set (i'm using er-1 -> microkorg and i just love it! great for live purposes )
|
|
|
|
Jeto
Jeto
Started Topics :
258
Posts :
3252
Posted : Mar 20, 2006 19:34
|
|
tribalizer
Started Topics :
5
Posts :
115
Posted : Mar 20, 2006 20:33
|
VERY cool!
 Felix
DJ Tribaldrummer
TribalQuest Sound Studio - Montreal
***
Time is nature's way of keeping everything from happening all at once |
|
|
|
punkah
Started Topics :
6
Posts :
210
Posted : Mar 20, 2006 21:40
|
|
ultra-nice |
|
|
|
|
