Percussa micro Super Signal Processor Eurorack Module (mSSP)
Percussa micro Super Signal Processor Eurorack Module (mSSP)
This project will only be funded if it reaches its goal by Sun, December 23 2018 4:00 PM UTC +00:00.
The Percussa micro Super Signal Processor (mSSP) Eurorack Module is a powerful Eurorack DSP module based on one of the fastest ARM processors available, the quad core Cortex A17 at 1.8GHz, and high end AKM audio converters offering 8 in / 4 out with up to 192kHz sample rate and 32-bit resolution.
Scroll down for more demos. The mSSP comes with 50 modular synthesis patches offering polyphonic wavetable, granular, sampling and traditional synthesis techniques, multi-channel audio processing and much more, all in only 26HP of rack space and for only $549.
The mSSP will also come with a Mac and Windows editor allowing the user to create and edit patches. The editor will connect with the mSSP hardware for real-time updates to patches and will support the SSP as well.
Key features and specifications
Audio and Control Voltage I/O
- 8 inputs via high end AKM ADC with 32 bit resolution supporting sample rates of 192kHz, 96kHz and 48kHz (AK5558)
- 4 outputs via high end AKM DAC with 32 bit resolution supporting sample rates of 192kHz, 96kHz and 48kHz (AK4454)
- DC coupled inputs and outputs, eurorack voltage compliant (-5V to +5V)
- All inputs and outputs support any signal type (audio, control, gate, trigger, etc)
- Low-noise power supplies and op-amp stages, featuring 0.1% resistors
- Quad core CPU running at 1.8 GHz per core with NEON SIMD parallel processing
- 32-bit floating point ADC/DAC input and output with 64-bit DSP algorithms
- Capable of running multiple DSP modules in parallel on multiple cores
- Mac and Windows editor software for full patch creation and editing
- Preloaded with 50 modular synthesis patches covering a wide range of synthesis techniques and audio processing methods (half of the patches are chosen by Percussa with the other half to be voted on by backers)
- Custom DSP software with more than 25 built-in synth modules such as wavetable oscillators, granulars, samplers, filters and effects
- Patches can use combinations of many fundamental DSP modules allowing multichannel and multivoice architectures
- Supports loading your own WAV files into patch modules via USB flash drives (e.g. load your own samples in the wavetable oscillators, granular processors, samplers, etc)
- Modify, store and recall patch module parameters as presets (limited only by size of the USB stick)
- Storage and recall of presets and patches with the press of a button
- 100+ pre-recorded samples / wavetables with royalty free license for (m)SSP users
- Supports loading synthesis patches created on the SSP by Percussa or the SSP community
Development Kit / VST support
- Support to develop and load your own Linux/ARM VST plugins (requires linux knowledge and compiling code for Linux/ARM)
- Example Quad VCA VST plugin code written in C++ using the JUCE framework already available for download on github and ready to go, including documentation in the SSP forum
Display, Graphics and User Interface
- 320x240 pixel high contrast full colour IPS display (wide angle visibility)
- 4x High quality Aluminium Machined Knobs
- 4x Endless and Click-less High Quality Encoders
- 8x Industrial Temperature high quality Switches
- Shallow menu system (no menu diving)
- Large fonts for parameter display for instant readability
- Signal input / output Oscilloscope features
- 2 USB host ports supporting USB MIDI (MPE) controllers and USB flash drives (for storing samples, patches, presets and for loading software updates)
- Support for USB ethernet adapters for developers (see instructions in SSP forum)
- Serial linux console connector on back of PCB
- 4 GB flash
- 2 GB RAM
Dimensions and Front Panel
- 26HP width, 3U in height
- 50mm depth behind front panel (estimated)
- 3mm thickness CNC machined aluminium front panel, brushed and black anodized
- Passive aluminium heatsink
- 800mA @ +12V and 100mA @ -12V draw (estimated)
- Standard 10-pin eurorack power connector
- Premium low impedance eurorack power cable included
The demo videos below show two patches running on the mSSP functional prototype:
- 4-voice sequencing with per voice step sequencer, microtonal quantizer (jazz scale from the SCALA archive), sine osc, envelope, bus routing + global stereo delay & reverb and frequency modulation of all sine oscillators via the input module. This patch contains 23 modules total. We use a make noise STO and function to frequency modulate the sine oscillators in this 4-voice patch. Various parameters in the patch are adjusted using the #mssp encoders, such as attack time for the envelopes and clock division factor for the step sequencers.
- 8-voice wavetable synthesis patch, controlled via MIDI, with each voice consisting of a wavetable oscillator with 3D morphing, ADSR envelope, input modulation, MIDI module, and a bus module. The patch also has a delay and reverb module before going to the output module where the signals are routed to the output channels.
- granular patch with input (start/length) modulation, and a delay and reverb processor, right before the output module. Sample used for the granular demo is choir sings -Magnificat- 140928.wav by klankbeeld / capella ducis on freesound.org.
All sound was recorded on the Percussa SSP, using its built-in multichannel recorder, and was then synced to video afterwards.
Why this powerful Eurorack module?
The mSSP comes with one of the most powerful ARM processors you can get in a eurorack module: a quad core A17 at 1.8GHz, which provides over 20,000 DMIPS of processing power. The A17 processor we use in the SSP and the mSSP is about 40x more powerful than the Cortex M processors popular in other eurorack modules.
By putting more processing power into the user’s hands, we empower them to create better sound and do more within a smaller eurorack system:
- Run more sophisticated and computationally expensive DSP algorithms
- Run DSP algorithms at higher sample rates, supporting up to 192kHz, with 32-bit resolution
- Run more DSP algorithms in parallel, enabling more voices and more channels (e.g. 8 voice granular with 200 grains per voice)
Easy to use
The mSSP will come with a Mac / Windows software editor allowing patch creation and full editing. Besides that the mSSP will be preloaded with 50 modular synthesis patches which lets users easily create powerful sounds:
- Modular synthesis patches can be switched on the fly using the up/down button
- Parameters of every synth module in the patch can be changed via parameter pages
- Samples/wavetables can automatically be loaded into patch modules
- Changes in parameter settings can easily be stored in a preset and recalled at any time
- Ability to change the internal routings between modules in patches using toggle settings
- Ability to mute/unmute a module in a patch
- Patches come with visual overview of input and output jacks which makes it instantly clear how the patches can be integrated in your rack
Overview of included synthesis patches
Below is a preliminary list of patches (subject to change) that will be included with the mSSP. Of the 50 patches we will provide, 25 patches will be based on suggestions and votings of the mSSP backers. So you get a chance to suggest patches and vote on them!
- 8 voice wavetable synthesis with 3D morphing, with 2 oscillators per voice (16 oscillators total), with MIDI support, XYZ modulation, envelopes and delay/reverb processors
- 8 voice granular synthesis with 200 grains per voice (live input and samples supported)
- 32 oscillator additive synthesis
- 8 SVF bank (modeled state variable filters)
- 4 voice physical modeling membrane synthesis
- 4 voice karplus strong (string synthesis) physical modeling patch
- 8 channel delay and reverb processor
- 8 channel comb filter processor
- 8 channel mixer with morphable levels
- 8 channel signal follower
- 4 voice modal synthesis (seperate exciter sections, filter bank and delay/reverb processors)
- 4 channel algorithmic pattern/curve generator
- 8 voice multichannel sample player with envelopes (up to 8 channels per player)
- 4 channel 4-operator FM with envelopes
- 4 channel physical modeling resonator
Any patch that has been created on the SSP, can be loaded onto the mSSP as well. So besides the patches we will develop you will have the chance to load patches from the SSP community.
Workflow of the Percussa mSSP Module
The UI of the Percussa SSP will use the same philosophy of the Percussa SSP, where only one “level” of menus is used, to avoid menu diving. Instead of 4 menus as is the case on the SSP, there will be only 2 menus on the mSSP:
- Global: to switch patches, make high level (macro) parameter changes to the patch, change signal routings in patches via simple toggle settings, and to save/recall presets (parameter settings of patches)
- Processor: to make changes to the parameters of the individual modules used in the patch as well as to load samples or other files
User Interface Design of the Percussa mSSP
- (Top left) Change parameters of synth modules
- (Top right) Load different WAV file into Sampler
- (Bottom left) Switcher module to change internal routings in patch
- (Bottom right) Switch presets/patch
- (Top left) Switch modules in patch
- (Top right) Input output overview of a patch
- (Bottom left) Visualization ADSR envelope
- (Bottom right) Select start and length of sampler
- (Left) Visualisation input scope
- (Right) Visualisation 3D Wavetable Oscillator
What is the difference between the mSSP and SSP?
Both the mSSP and SSP are equally powerful in terms of DSP power and sound quality, since they share the same CPU as well as the same family of high end audio converters. The table below shows the differences between the two modules in detail:
Appendix - DSP modules to be used in the mSSP patches
The mSSP software will use the DSP engine we developed for the SSP, featuring more than 25 different synthesis modules. The patches shipped with the mSSP use combinations of these modules, and through the menu system of the mSSP it will be possible to change the parameters of the modules used in the patches.
- wavetable oscillator with 3D morphing, stacking, loading of arbitrary WAV files and realtime recutting, etc
- granular processor with all the usual controls including start and length of grain selection window, duration, pitch, randomization, grain trigger freq, nr of grains, etc.
- comb filter
- SVF (modeled)
- envelope generator with log/exp stages
- physical modeling membrane percussion module
- envelope follower
- reverb and delay processors
- apreggiator of arbitrary signals with clock divider
- step sequencer
- sampler module supporting up to 8 channels per sample and loading of 8 samples per module with start/length, crossfading, etc.
- simple delay line
- LFO (also usable as normal oscillator)
- microtonal quantizer with SCALA support
- noise generator with variable freq
- DC generator
- VST loader module
- quad VCA (VST module)
- output and input modules (with gain/bias per channel on the input module)
- MIDI module with MPE support
- bit crusher
- bus module for summing and routing signals
- signal switcher (audio rate)
Selection of upcoming modules and features on the SSP:
- looper module
- time stretching for the sampler module
- live input and loadable (WAV) windows for the granular processor
Additional features and modules requested by SSP users are described in the Percussa SSP forum. Since we shipped the SSP we have averaged 1 software update every 1-2 weeks.
What is the status of the project?
The Percussa mSSP module is currently a functional prototype, based on the Percussa SSP, which was successfully funded on Kickstarter in September 2017, with first units delivered in June 2018.
Here is what we’ve already done development wise for the mSSP:
- Designed and manufactured mSSP front panel prototypes
- Designed and manufactured a prototype PCB for mechanical fitting and testing
- Connected switches, encoders and display to our prototype PCB for UI testing and connected the prototype PCB to the SSP’s expansion connectors so we can develop for the mSSP, on the SSP platform
- Updated the SSP’s software and drivers to show graphics on the mSSP’s display
- Designed and tested the first demo patches for the mSSP featuring wavetable and granular synthesis
The mSSP module will be based on the carrier PCB design of the SSP, but needs several hardware engineering changes and optimizations in order to produce it at a lower price and to make it fit within 26HP of eurorack space:
- We need to redesign the PCB of the SSP to make it smaller. To do so, we’ll need to split it into multiple PCBs which will be stacked on top of each other (audio and USB I/O, audio converters, CPU and flash/RAM module)
- the ARM system on module which we used on the SSP will be replaced with a module we will design and cost optimize in-house, featuring the exact same processor as on the SSP.
- We’ll design a custom passive heatsink for the mSSP just like we did for the SSP, keeping the CPU cool, and letting us run it at maximum speed.
- The audio I/O section of the mSSP will be based on the SSP’s I/O section but will have less channels, so we will be redesigning the that part of the PCB. We will be using the same high quality components from the SSP so there is no difference in sound quality.
- The display of the mSSP will be connected via a different bus than the wide format display of the SSP which means the display section of the PCB will be redesigned.
Besides these hardware modifications, there is also the following to be done software wise:
- we need to build a UI specifically for the mSSP that sits on top of our existing SSP software code base.
- we need to develop the Mac/Windows patch editor software supporting the mSSP and SSP
- we need to develop the patch collection for which part of the patches is listed above
- we need to add additional features and modules to the existing SSP software to support the patch collection we'll be developing
- Dec - Ordering parts
- Jan - PCB design + order front panels
- Feb - First prototype assembly and testing
- Mar - Software engineering + patch development
- Apr - Second (final) prototype assembly and testing
- May - Final Software and Hardware testing & Manufacturing
- June - Assembly and shipping Normal Backers
Taxes and Duties
If you are located outside the United States, please note that import duties, taxes or VAT is not included in the pledges. When your shipment clears customs, you (the backer) might be liable for paying import duties or taxes to the customs office in your country.
If you are located in California, please note that sales tax is not included in the listed pledges, and depends on the county / city we are shipping to. At the time of shipment, we will calculate the correct CA sales tax and send you an email with a secure payment link asking you to pay the correct sales tax amount for the address we are shipping to.
Outside California, within the United States, sales tax is the responsibility of the backer.
The HS code for customs classification of the module will be 9209.94.80 00 (parts for musical instruments for which the sound is produced electronically).
This kickstarter project would not have happened without the support of our SSP users and community, as well as close friends in the electronic music and music technology communities, and enthousiastic eurorack users. You know who you are.
Risks and challenges
The main challenge of the mSSP project is designing and manufacturing the mSSP PCBs and adapting the existing SSP code base to the unique hardware user interface (small screen, different button layout, ...) of the mSSP. Given the successful SSP Kickstarter project we recently delivered and our history of designing and shipping multiple products since 2004, we’re confident we can design, manufacture and deliver the mSSP.Learn about accountability on Kickstarter
- All gone!