DARK SYNTH 1.0 REFERENCE MANUAL

Transcription

1 DARK SYNTH 1.0 REFERENCE MANUAL

2 DARK SYNTH 1.0 REFERENCE MANUAL 2014 ConTempoNet

3 Dark Synth Reference Manual - INDEX Introduction 3 The Main Interface Spectrum Spectral Processing Oscillators Filter FX Amp Global 13 The Edit Window Spectrum Pitch Filter Amp Modulators Global 24 Sound Design and CPU: tips & tricks 27 Filter Licenses 30 5T

4 Introduction INTRODUCTION sound wave Fig. 5.1 An analog signal analog signal The Dark Synth is an additive synthesizer with some unique characteristics. You can define a waveform of up to 2048 partials (sine waves) for its main oscillator, the ADDITIVE oscillator, using presets waveforms, spectral processing algorithms and freehand drawing. Moreover you can add up to 128 copies of the Additive oscillator, and each copy (generator) can be detuned and phase-shifted with respect to the others. This creates a sort of chorus which is very natural and deep, because it is not realized with modulated delay lines, but using an actual "choir" of generators. This is probably the main distinctive characteristic of the DS. At its maximum you have 2048*128 = 262,144 "virtual" sine waves for a single voice. You also have a SUB oscillator, which can play classical band limited waveforms, and also noise tones. 3

5 Introduction The Dark Synth is particularly suited for creating monophonic lead synth sounds, but it is also capable of producing very interesting evolving timbres, whether mono or polyphonic. There are two UI areas: the Main Interface (the actual device interface) and an Edit Window (a floating window) for the more sophisticated parameters. 4

6 The Main Interface THE MAIN INTERFACE The Main Interface is divided in 7 Panels: 1) Spectrum With the upper menu you can set the number of partials (Overtones) for the waveform, from 8 to 2048 With the lower menu (Fill Spectrum) you can select some preset waveforms: Square, Sawtooth, Circular (similar to triangular in timbre, but with a more powerful sound), Smoothsaw (a sawtooth with more attenuation in the upper partials), Impulse (the narrowest impulse possible for the given number of partials), Sine, Random (a different random amplitude for each partial) and Rand(1/f): which has random amplitude for each partial like the Random waveform but the components are progressively rescaled (first partial not rescaled, second rescaled by 1/2, third rescaled by 1/3 and so on). The Smoothing parameter sets the transition time when you switch waveform, or when you edit it with the Spectral processing parameters or by freehand drawing. The Reset button, retrigs the waveform preset again: this can be useful to change the random waveforms (you get a different spectrum each time), or to return to a preset waveform after you have modified and "frozen" it with the Spectral processing parameters (more later) or with freehand drawing. 5

7 The Main Interface 2) Spectral Processing In the second panel you can see the actual waveform in the upper left side. Let's skip the Freeze button (upper right corner) for the moment. The Saturation parameter modifies the ratio between the partials' amplitude: with values greater than 0 the partials with low amplitude are emphasized so their amplitude increases, with values lower than 0 the partials with low amplitude are attenuated so their amplitude decrease. In the lower left corner there is a Brickwall Filter. It is not an actual filter but a spectrum processor which lets you "silence" the lower (X parameter) and upper (Y parameter) partials. You can modify it by using the bottom number boxes, or the XY graphic panel above them. The Comb parameter sets an interval between partials: 1 means all partials are on, 2 means every other partial is on (1, 3, 5, 7 and so on), 3 means 1 partial on and 2 off (so 1, 4, 7, 10 etc.), up to 32 (1 partial on and 31 off). 6

8 The Main Interface The Shift parameter shifts the Comb filter to the right by a certain number of partials (from 0 to 32). If Comb is not operative (i.e. its value is 1), the Shift parameter is similar to the X parameter of the Brickwall filter. Density sets the percentage of active partials. If it is 100% all partials are active, if it is 50% each partial has a 50% (random) probability of being active or silenced, and so on. The Dry/Wet parameters sets a ratio between the spectrum processed waveform and the dry waveform (as it is generated by the first panel) The R button resets all the Spectral parameters back to the default values: therefore the waveform will return to its initial (non processed) status. The Freeze button on the upper right, freezes the spectrum as it is modified by the Spectral parameters, and resets all the Spectral values. You can then apply them again to further modify the spectrum. If you press the Retrig button on the first panel, the modified spectrum will be deleted and substituted by the current preset selected in the "Fill Spectrum" menu. 7

9 The Main Interface 3)Oscillators This panel is divided in two sections: ADDITIVE oscillator (upper) and SUB oscillator (lower) - ADDITIVE oscillator: The Generators menu on the left sets the number of parallel generators all sharing the same waveform. It goes from 1 to 128 The Pitch Spread parameter below sets the detuning factor between generators. The Distribution button switches between random detuning distribution of the generators pitches and regular (even) distribution. When "Even" is set the pitch gap between generators is spaced evenly. The Rnd Phase parameter is used to give a different random phase to each generator. When it is 0% all generators are in phase; in this case, if the Distribution is "Even", it is possible to create interesting rhythmic patterns of "cascading harmonics" due to the beatings between the generators' partials. The OSC RATIO is the number box between the two panel sections. It goes from 0% to 100% and sets the volume ratio between the two oscillators. When it is 0% you hear only the Additive oscillator, at 50% both oscillators and at 100% only the Sub Oscillator. 8

10 The Main Interface - SUB Oscillator On the upper left you have an ON/OFF switch for the oscillator. When it is OFF the Osc Ratio is excluded (i.e. you always hear the Additive Oscillator at full volume). With the Envelope Menu you can choose which part of the amplitude envelope will be used by the Sub oscillator: from A (only the attack) to ADSR (the whole envelope) Below you have the Waveform Menu: here you can choose the waveform for the sub oscillator. You have the classic oscillators and three noise generators, white noise, pink noise and a variable frequency random noise generator. The Ratio parameter sets the ratio between the upper and the lower part of the waveform for Square and Triangular waveforms. (This parameter has no effect on other waveforms). The Tune dial sets the pitch intonation relative to the main pitch. The Fine parameter fine-tunes the pitch intonation with values lower tan a semitone. The Color parameter changes the energy of the upper partials of the Sub Oscillator (it is an LP filter). 9

11 The Main Interface 4) Filter With the Type menu you can choose among 14 different filters: 2 types for Low-Pass, Band-Pass and High-Pass both at 12 db and 24 db (so 4 LP, 4 BP and 4 HP), and two "analog" filters: LP Transistor and LP Diode. The Cutoff frequency is expressed in Octaves and it is relative to the note played. The Resonance sets the amount of emphasis you give to the cutoff frequency. The Depth parameters sets the amount of filter cutoff variation given by note velocity and/or envelope+lfo. The Dry/Wet parameters set the ratio between filtered and unfiltered signals for the two oscillators. 10

12 The Main Interface 5) FX This panel contains three parameters for distortion/noisy effects The Stretch parameter stretches/shrinks the partials of the ADDITIVE oscillator without changing the fundamental. So for instance if you have a sound with partials etc. Hz, you can stretch it to etc. Hz (i.e. the space between partials is 110 Hz even though the fundamental is 100 Hz). The parameter values represent the stretching factor as a power of two. So 0 means 2 0 (i.e. 1) and doesn't stretch the partials. 1 means 2 1 (i.e. 2) and doubles the spacing of the partials ( ). -1 means 2-1 (i.e. 0.5) and halves the spacing of the partials ( ) and so on. The SUB oscillator is not influenced by the stretch factor. The Overdrive parameter is a classical distortion effect, it goes form "warm" to totally noisy. The Ring Mod parameter sets a ring modulation between the SUB and ADDITIVE Oscillators. The values express the percentage of Sub Osc which ring modulates the Additive Osc. So, f.i. if this value is 60%, we have 60% of ring modulated sound and 40% of Additive Osc sound. The Sub Osc modulation can be set "pre" or "post" envelope (with the Pre/P.E. switch on the right of the Ring Mod dial) 11

13 The Main Interface 6) Amp In this panel we have the classic ADSR Envelope. The Curve parameter is used to set the slope of the ADSR segment, from linear (1) to exponential (8). The VelSens parameter sets the difference in db between a note of velocity 1 and a note of velocity

14 The Main Interface 7) Global The upper menu sets the number of voices available, from "mono" to 12 The EDIT button opens the Edit Window (see next pages) The Glide parameter sets the glide or portamento between notes. When the Legato switch on the left is on, you have a portamento only when two consecutive notes are played legato by the same voice (this is most effective in "mono" mode), when it is off, you have the portamento even if the two notes are not legato. The Transpose dial sets a global transposition for the synth (in semitones) The Gain dial sets the overall volume of the synth. 13

15 The Edit Window THE EDIT WINDOW You can open the Edit Window pressing the Edit button in the upper right corner of the Main Interface. The Edit window is divided in 6 tabs. 1) Spectrum In the graphic section of the Spectrum Window it is possible to modify the spectrum dragging the partials with the mouse. The number of partials shown corresponds to the Overtones parameter in the first panel of the main UI. The blue area represents the spectrum before the Spectral processing, the grey area represents the spectrum after the Spectral processing. When we press the Freeze button on panel 2, the blue area becomes identical to the grey one. With the two number boxes in the upper left corner of the window it is possible to change exactly the numeric value for any partial. 14

16 The Edit Window In the Spectral Modulation section you can assign one of the three global LFOs (see tab 5) to each of the Spectral parameter. In the Spectral Filters section you can activate a series of "harmonic filters" to select only a certain series of partials. This can give interesting variation of the timbre: f.i. an "Hammond like" or a "pipe organ like" sound. Fundamental: only the fundamental (1st partial) is selected. Octaves: all the partials which are one or more octaves above the fundamental are selected (2nd, 4th, 8th, 16th partial etc.). When the Hr switch is on, the harmonic series starting on 2nd partial is activated (2nd, 4th, 6th, 8th, 10th partial etc.) Fifth: the third partial (which is a fifth plus one octave above the fundamental) is selected, along with its octaves (so 3rd, 6th, 12th, 24th partial etc.). When the Hr switch is on, the harmonic series starting on 3rd partial is activated (3rd, 6th, 9th, 12th, 15th partial etc.). Third: the fifth partial (which is a major third plus two octaves above the fundamental) is selected, along with its octaves (so 5th, 10th, 20th, 40th partial etc.). When the Hr switch is on, the harmonic series starting on 5th partial is activated (5th, 10th, 15th, 20th, 25th partial etc.). Seventh: the seventh partial (which is a minor seventh plus two octaves above the fundamental) is selected, along with its octaves (so 7th, 14th, 28th, 56th partial etc.). When the Hr switch is on, the harmonic series starting on 7th partial is activated (7th, 14th, 21th, 28th, 35th partial etc.). 15

17 The Edit Window 2) Pitch In the upper left section it is possible to activate a pitch envelope: you can select the target (ADDITIVE and/ or SUB Oscillator), and the envelope mode (with or without a Sustain phase). The start and end of the envelope can be independently selected. The Retrigger switch activates the envelope retrigger for legato notes: if it is off a legato phrase uses only one envelope. The number boxes Key>Atk and Vel>Atk can be used to modify the attack time in relation to pitch and key velocity. 16

18 The Edit Window The number boxes Key>Rel and Vel>Rel can be used to modify the release time in relation to pitch and key velocity. The Attack and Release dials set the duration of the start and the end of the envelope. The Exp number boxes are used to give an exponential/logarithmic curve to the envelope segments. The Start and End dials set the relative pitch of the start and the end of the envelope (the sustain phase is always untransposed) The right vertical section contains parameters to set the Keyfollow for the two oscillators: 1 means the usual semitone step, 2 means whole tone steps, 0 means fixed frequency (C3 if it is not transposed in the Global panel). Negative values reverse the keyboard intervals (around C3). The other two dials (Rand Pitch) set a random factor for the pitch: from slight detuning to totally random. The lower section is for the LFO selection. You can assign one of the three LFO (see tab 5) to pitch modulation. It is possible to set a Fade In for each new phrase, and a Retrigger factor to restart it during each note of a legato phrase. There are actually 2 versions for each LFO, one local and one global. The local version is different for each voice and can be modulated by the Mod envelopes (see tab 5), the global version is common for all the voice (so they are all modulated in sync) and cannot be modulated by the Mod envelopes. The Global switch lets you select one of the two version. 17

19 The Edit Window 3)Filters The Filter window is similar to the Pitch window. It is used to modulate the cutoff frequency of the filter. In the upper left section you can choose between two envelopes: a specific Filter envelope a n d t h e Amplitude envelope. The envelope starts and ends at the Cutoff value set in the Filter panel, the sustain level depends on the Depth parameter and the Velocity>Cutoff parameter (right section of the Filter window). 18

20 The Edit Window In the right vertical section we have the previously-mentioned Velocity>Cutoff parameter, which sets the velocity sensitivity for the filter cutoff. If it is 0 the cutoff value is always Cutoff+Depth, for any velocity value. If it is 1, you only reach the Cutoff+Depth value at velocity 127. Intermediate values give intermediate results. The Keyfollow and Random parameters are like the similar parameters we saw in the Pitch window. The lower section is for the LFO selection. You can assign one of the three LFO (see tab 5) to filter modulation. It is possible to set a Fade In for each new phrase, and a Retrigger factor to restart it during each note of a legato phrase. There are actually 2 versions for each LFO, one local and one global. The local version is different for each voice and can be modulated by the Mod envelopes (see tab 5), the global version is common for all the voice (so they are all modulated in sync) and cannot be modulated by the Mod envelopes. The Global switch lets you select one of the two version. 19

21 The Edit Window 4) Amp In the upper section you have some parameter for the time scaling of Attack, Decay and Release of the Amplitude Envelope. You can scale the time of the envelope phases using velocity and pitch. The middle section contains parameters for the "wave normalization" of the ADDITIVE Oscillator. The amplitude of the Additive Oscillator can be very instable when you have lots of detuned parallel generators. The wave normalizer tries to m a i n t a i n a s t e a d y amplitude. As a side effect it can distort the waveform, 20

22 The Edit Window especially if the Norm Speed (second dial) is high. The lower section is for the LFO selection. You can assign one of the three LFO (see tab 5) to amplitude modulation. It is possible to set a Fade In for each new phrase, and a Retrigger factor to restart it during each note of a legato phrase. There are actually 2 versions for each LFO, one local and one global. The local version is different for each voice and can be modulated by the Mod envelopes (see tab 5), the global version is common for all the voice (so they are all modulated in sync) and cannot be modulated by the Mod envelopes. The Global switch lets you select one of the two version. 21

23 The Edit Window 5) MODULATORS In the upper section there are three LFO. For each LFO you can choose the Waveform (Sine, Triangle, Square, Sample&Hold, Random). The Ratio parameter sets the ratio between the upper and the lower part of the waveform for Square and Triangle waveforms. When it is 0, the Triangle waveform becomes a Sawtooth. The Timing can be Free (expressed in Hz) or Sync (expressed in note values). 22

24 The Edit Window When timing is in sync mode, the Mult parameter appears, which multiplies the note value by an integer factor between 1 and 16. F.i. if Rate is 1/4 and Mult is 3, the LFO rate is 3/4 beats. There are actually 2 versions for each LFO, one local and one global. The local version is different for each voice and can be modulated by the Mod envelopes (see below), the global version is common for all the voice (so they are all modulated in sync) and cannot be modulated by the Mod envelopes. With the two Mod menus you can use one of the two Mod Envelopes to modulate the local LFO frequency and/or amplitude. The Depth parameters set the influence the Mod Envelope has on the LFO frequency and/or amplitude: with positive values the frequency accelerates and the amplitude increases, with negative values the frequency decelerates and the amplitude decreases. In the lower sections you can set the two Mod Envelopes. The parameters are similar to the Pitch and Filter Envelopes parameters. 23

25 The Edit Window 6) Global The Granular Stereo Simulator (upper left section) is a stereo effect based on delay lines. The sound is delayed with different times in the left and right channel, and the delay time is randomly changed at a certain rate. The Width parameter sets the ratio between the dry mono sound and the stereo effect. The Separation p a r a m e t e r s e t s t h e maximum delay time: the higher the value the more left and right channels are separated in time. 24

26 The Edit Window With higher values you can clearly hear the two sound copies. The Granularity parameter sets the rate for the random delay time change: the higher values give a sort of "granular filtering" effect, which resembles a chorus. The Global Fine Tuning can be used to alter the synth pitch by very small values (less than a semitone) The Midi Controllers (lower left section) are used to assign Aftertouch and MIDI controllers to various parameters. You can set Aftertouch modulation for the parameters which affect the timbre of the sound: Filter Cutoff, Stretch, Overdrive and Ring Modulation. You can assign three MIDI Controllers to control the amplitude of Pitch, Amplitude and Filter LFOs. In this section you can also set the Pitchbend Range, which is 2 semitones by default. In the right vertical section there are three switches useful to lower the CPU and/or to get a noisy sound. They are all On (more CPU) by default. The Osc Wavefade switch turns on and off the waveform crossfade algorithm. When it is on the number of sound generator for the Additive Oscillator is actually doubled: when the waveform is modified there is a crossfade between the two copies of the generators to avoid clicks. The Add Bandlimit switch turns on and off the algorithm which continuously modifies the waveform of the Additive oscillator in relation to the note pitch, to avoid aliasing. The Mod Bandlimit switch turns on and off the algorithm which continuously modifies the waveform of the Additive oscillator during an Envelope or LFO pitch modulation. 25

27 The Edit Window The Band Limit Cut dial sets the bandlimit frequency in relation to the Nyquist frequency (half the sampling rate): when it is 100% it is exactly half the sampling rate, when it is 50% is 1/4 of the sampling rate, and so on. 26

28 Sound Design and CPU: tips & tricks SOUND DESIGN and CPU: tips & tricks As we said the real "vocation" of the Dark Synth is to produce great fat evolving monophonic synth sounds. But it can also be used to create other types of sounds, even polyphonic ones. You have probably noticed that the synth is not "cheap" in terms of CPU: its synthesis engine is doing a lot of work to deliver its unique sound. For instance the evolving spectrum is calculated with an inverse Fast Fourier transform roughly 50/100 times per second; some filters are oversampled to give a better sound quality; the pitch shifting algorithms operate on up to 128 generators per voice, each one with a slightly different pitch calculation; and so on. When you create your own sounds you should be aware of the hard work the synth is doing under the hood. While with monophonic sounds there are practically no limitations; the polyphonic ones should be carefully programmed to get the maximum out of the synth. First of all, keep in mind that virtually any parameter can influence the CPU: the LFOs, the Overdrive, the Stereo simulator, and so on. For the most part, however, the increase of CPU is negligible. Other than the number of voices used, there are a few parameters you should take care of: the number of "Generators", the Filter and some "Audio Processing" parameters. 27

29 Sound Design and CPU: tips & tricks The first parameter to consider is the number Generators in the Oscillator panel of the main UI. The more generators the more CPU, of course. But more generators doesn't necessarily mean "better sound", only "different sound". So try to change this parameter and lower the generator number: you'll be surprised how it can change your sound. Filters are expensive, especially the "svf" which are oversampled, and the 24 db filters, which are actually made with two stacked filters. If you don't need an evolving filter in your sound, but simply a color change (darker, or brighter), you should consider the spectrum "Saturation" parameter (Spectral Processing panel of the main UI) which has no additional CPU cost. Positive values give you generally a brighter sound, negative values give you a darker sound. In the same Spectral Processing panel the Brickwall filter can be used to change the sound color; and, last but not least, you can change the number of Overtones in the Spectrum panel (less overtones mean darker sound). In the right vertical section of the Global tab of the Edit window, there are three switches useful to lower the CPU and/or to get a noisy sound. They are all On (more CPU) by default. The Osc Wavefade switch turns on and off the waveform crossfade algorithm. When it is on the number of sound generator for the Additive Oscillator is actually doubled: when the waveform is modified there is a crossfade between the two copies of the generators to avoid clicks. If the spectrum doesn't change due to modulation, or real time editing, you can switch off this parameter and see if it doesn't add unwanted noises to your sound. 28

30 Sound Design and CPU: tips & tricks The Add Bandlimit switch turns on and off the algorithm which continuously modifies the waveform of the Additive oscillator in relation to the note pitch, to avoid aliasing. You can switch off this parameter if you use your sound in a limited range (a couple of octaves) The Mod Bandlimit switch turns on and off the algorithm which continuously modifies the waveform of the Additive oscillator during an Envelope or LFO pitch modulation. You can switch off this parameter and check if it doesn't add unwanted noises to your sound. 29

31 Filter licenses FILTER LICENSES Filters code ported to Max by Oli Larkin - LP_Diode License: // This code is released under the MIT license (see below). // // The MIT License // // Copyright (c) 2012 Dominique Wurtz ( // // Permission is hereby granted, free of charge, to any person obtaining a copy // of this software and associated documentation files (the "Software"), to deal // in the Software without restriction, including without limitation the rights // to use, copy, modify, merge, publish, distribute, sublicense, and/or sell // copies of the Software, and to permit persons to whom the Software is // furnished to do so, subject to the following conditions: // // The above copyright notice and this permission notice shall be included in // all copies or substantial portions of the Software. // // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR // IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 30

32 Filter licenses // FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE // AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER // LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, // OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN // THE SOFTWARE. Emulation of Diode ladder lowpass filter as found in Roland TB303 or EMS VCS3 source: LP_Transistor License //// LICENSE TERMS: Copyright 2012 Teemu Voipio // // You can use this however you like for pretty much any purpose, // as long as you don't claim you wrote it. There is no warranty. // // Distribution of substantial portions of this code in source form // must include this copyright notice and list of conditions. // 31

33 Filter licenses source by mystran Odf_svf filters (LP BP HP) Info // Example code for a TPT/ZDF-SVF // based on code written by Robin Schmidt, licensing/copyright: none (public domain) 32