SWAT - TheTechnobear/SSP GitHub Wiki
swat - swiss army tool
swat is designed to be like the famous swiss army knife, a useful utility to bring out to solve your problems.
Documentation
Note: This is the primary page for documentation on this modules
- Wiki - https://github.com/TheTechnobear/SSP/wiki/SWAT
- Forum - https://forum.percussa.com/t/swat-swiss-army-tool-by-thetecnobear
Developer: TheTechnobear
I develop these plugins for free, please consider supporting my efforts with a donation. https://ko-fi.com/thetechnobear
Parameters
| Name | Desc |
|---|---|
| Various | depends on algo |
Inputs
| Name | Desc |
|---|---|
| X 1-4 | X input signals, usage depends on algo, see below |
| Y 1-4 | Y input signals, usage depends on algo, see below |
| Z 1-4 | Z input signals, usage depends on algo, see below |
Outputs
| Name | Desc |
|---|---|
| Out A 1-4 | A output signals, usage depends on algo, see below |
| Out B 1-4 | B output signals, usage depends on algo, see below |
Buttons
| Name | Desc |
|---|---|
| Left/Right | move between 4 instances |
| AG+/- | change algo for instance |
note: this is a generic module, so important to look at different algos... though there is help on the screen
General Usage
The goal of SWAT is pretty open ended, its designed to be a plugin that I can extend quickly and easily. It's inspired by Expert Sleepers Disting (but shares no code, or similar engines etc...)
The plugin can run a variety of algorithms, that can do a variety of (unrelated) functions.
The 'interface' into swat is the same for all algos - each has 3 inputs X, Y and Z, and has 2 outputs A and B. (This is fixed and cannot change!)
I aim to use thee as follows...
inputs
X - primary input signal Y - secondardy modifier signal Z - third input, often a gate or trigger
outputs
A - main output B - alternative output (often, will be inverse of A, depending on algo)
however, the specific meaning of each parameter is particular to each algo.
important notes
- most algos may be designed to work at audio rate.
- different algos will have different computational costs
- many inputs are optional, and have 'sensible' defaults (particularly y and x)
- check inputs... they aim to be consistent
- unconnected outputs will usually result in less cpu (dependent on algo)
- input gates are considered high > 0.5
- output gates are 0.0 low, 1.0 high (note: unipolar! )
note on signals levels
Theoretically, on the SSP , we have -5v to +5v currently -1.0 to 1.0f, where A4 = 440 hZ = 0v. where an octave = 0.2. however, CV inputs are currently NOT calibrated to this theoretical values. instead A4 = 0.02325f, and an octave is approximately 0.18837.
Usually, the user does not need to know/care about this, as all pitch voltages take this into account. however, when mapping or using external voltages this can become important. all SWAT algorithms that explicitly deal with notes/pitch take this into account.
Algorithms
Display
No Outputs
Displays X, Y, Z inputs as float, voltages and note values
Tips: I use mainly to be able to see note values (including cents)
Min Max
A = min(X,Y) B = max(X,Y) Z gate
Returns min and max signals
Tips: Z acts a kind of sample n' hold, also only using calculating min/max when high.
Switch
A = Z > 1 , X else Y B = Z > 1 , Y else X
outputs flip between X, Y depending n Z
Tips: Handy simple switch alternative to MSW8
Transpose
A = X + Note A B = Y + Note B
Transposes pitch CV. A0 = zero transpose, A#4 = +1 semi , A1 = +1 Octave etc.
Tips: if X is disconnected, its defaults to 0v , so A = Note A as v/oct
Comparator
A = gate X > Y B = ! A Z Hysterisis
Returns a gate where X > Y Hysterisis is useful X might be fluctuating slightly.
Windowed Comparator
A = gate X > (L-Y) & X < (H+Y) B = ! A Z Hysterisis
Returns gate high when X is between low and high. Y acts as a 'spread' to low/high Hysterisis is useful X might be fluctuating slightly.
Tips: if you set low/high to the same value (e.g zero), you can use Y to modulate around this. low and high set to same value plus hysterisis allow detection of a particular value (with voltage/float imprecision accounted for)
Note Comparator
A = gate (X > L & X < H ) && Y B = ! A Z Hysterisis
Returns gate high when X is between low and high notes Hysterisis is useful X might be fluctuating slightly.
Tips: Z is used to allow a bit of quantization around note values. Low and High the same can detect a particular note e.g. to trig a particular sample on SAM from a given note!
Logic - And
A = X AND Y B = ! (X AND Y) Z is gate
result is X AND Y , where high = 0.5 Z acts as a sample and hold
Logic - Or
A = X OR Y B = ! (X OR Y) Z is gate
result is X OR Y , where high = 0.5 Z acts as a sample and hold
Logic - Xor
A = X XOR Y B = ! (X XOR Y) Z is gate
result is X exclusive OR Y where high = 0.5 Z acts as a sample and hold
tips: XOR = X or Y NOT both !
Map Value to Value
A = X : in range to out range B = Y : in range to out range
Map an input range to an output range. output values are clipped to min/max out.
Tips: many usages e.g. converting unipolar to bipolar. by making output min > max, you can invert ranges!
note: often you can do the same/similar by using the SSP scale and offset features.
Map Note to Value
A = X : in range to out range B = Y : in range to out range
Map an input range of pitch CV to an output range. Similar to Map Value to Value, however, this algo takes ito account how pitch is represented and convert to a numeric range.
Tip: this is useful when we want to convert a pitch cv into a modulation signal - mapping a midi keyboard octave to a range e.g C4 - C5 -> -1.0 to +1.0 for start position of a sample or granular buffer
Map Note to Note
A = X : in range to out range B = Y : in range to out range Similar to Map Value to Value, however, this algo takes ito account how pitch is represented and convert to a new pitch. This could be seen as a kind of transposition, however, it will not only offset the pitch, it will also scale or invert.
Counter
A = A + -0.5>X<0.5 B = B + -0.5>Y<0.5 Z = reset
A simple counter that increase by step amount when X/Y exceeds 0.5 however, X/Y are treated as bipolar so X > 0.5 steps forward , X < -0.5 steps backwards !
Z high will reset to the minimum value on true, and stay there until taken low.
Tips: This is audio rate, so this can for a kind of stepped phasor. Useful with MSW8 to act as a sequential switch
Delay Time
X = Signal Y = Delay (% buffer) A = Delayed X ( Delay + Y) B = Delayed X ( Delay + Y) + X
This is intended as a 'microtime delay' down to sub millisecond, hence the small values by default. (it will go to 5 seconds, but Delay parameter is more scaled to small size values ;)
Y is an offset to Delay time. B is wet + dry signal.
Important notes: Changing the SIZE parameter resets the delays, if you want to modulate delay time then use Y with a larger SIZE parameters modulation of Y is not really intended for audio effects, for effeciency it is read at control rate ( 128 samples@48k), so low audio rate, and there is no interpolation across sample reads, and changes of Y are immediate i.e. its not really the same as changing read head speed.
I've no intention of changing this ;) ... I'd like to make a more elaborate audio delay FX with speed modulation and multi head but this will be a separate VST.
Delay Sample
same as Delay Time, but using sample as unit for delay
Precision Adder
A = X + Y + Z B = X - Y - Z
Sums or Subtracts X/Y/Z mostly a development tool, given SSP has native summing of inputs.
Constant
A = Constant A + X B = Constant B + Y
mostly a development tool, given DCG can output constant values and SSP has native summing of inputs.
tips: if X/Y are disconnected then they are 0v, so A/B are the constant values
Can you add this algorithm?
Possibly... of course this comes down to a couple of aspects.
- do I have time :)
- does this fit SWAT? or would it be better of as a dedicated module?
Note: Whilst Disting may be a source of ideas, I do not plan to implement its algorithms! This would be a mammoth task, and also pretty pointless... there are many functions in Disting that can already be achieved by other modules or by the SSP natively. e.g. you dont need a precision added on the SSP since summing is automatic and lossless on the SSP anyway.
We are not limited by hardware production costs or physical space.. So more complex functions would likely be better implemented as dedicted modules. e.g. having simple sample/looper in SWAT doesn't really make much sense, it'd be better to have a dedicated module to do this.
This document is based on the LATEST version of the plugin
Tips
see individual algos :) most important tip : read the help text on each algo, whilst inputs/outputs are hopefully consistent - sometimes they might be unexpected
Change Log
- 2.0.0 release for new ssp image
- 1.4.1 release
- 1.4.0 early access release - midi learn
- 1.3.1 release, re-write using new infra and ui
- 1.3.0 early access release - add delay (sample)
- 1.0.1 Adjust for pitch cv changes
- 1.0.0 Initial Release