An APU does for Audio what a GPU does for Graphics

Why do you use a GPU for real-time graphics processing? It’s not because the CPU can’t handle graphics processing. It’s because, when you offload the graphics computation onto a specialized, dedicated processor running in parallel with your CPU, you guarantee high-quality, glitch-free delivery of real-time graphics and (as a bonus) free up compute cycles and memory on your main computer for other tasks. The same is true for audio!

An Audio Processing Unit (APU) delivers high quality, real-time audio that is reliable, sample-accurate, jitter-free, and highly responsive to controller interaction — while at the same time freeing up computing and memory resources on your computer that can be dedicated to other tasks.

Because APU hardware is designed specifically for computing audio signals in real time, and because all of the APU’s computing cycles and memory are dedicated to computing sound, you don’t have to cut corners or compromise the audio quality of your real-time sound synthesis and processing algorithms.

…what seems to have happened is the software manufacturers (plugins etc) have scaled to fit the new CPU capabilities. The upshot is the CPU is still getting under load… Kyma is attractive because, being its own eco-system, it doesn’t suffer from CPU usage bloat in the same way and works reliably irrespective of what the computer is struggling with.

— Alan M Jackson, experimental musician / composer, moderator of the Kyma Kata and personal coach to game & film sound developers

The question is not whether present-day CPUs are fast enough to do audio computation.  But, no matter how fast the CPU, your audio algorithms have to be designed to run alongside an unknown number of other audio applications and plug-ins, not to mention all the non-audio apps and system tasks that are continuously running in the background on your main computer (things like web browsers, background file-indexing, drivers, and other system tasks). By shifting real-time audio computation to an APU designed for and dedicated to that task, not only can you improve the quality of your real-time audio signal processing algorithms, you can also free up memory and computing power on your main computer — resources that you can then use for handling other tasks.

Audio Processing Units are part of the landscape of ubiquitous computing you see everywhere around you — there are GPUs running alongside every CPU, there are 25 to 50 processors in the average car, there is a processor in each audio interface, a processor in each MIDI keyboard or controller; there are server processors dedicated to doing nothing but encryption; there are processors fully dedicated to bitcoin-mining, not to mention the CPUs and GPUs in your phone, your gaming consoles, your watch and in your thermostat. It makes sense to distribute computational tasks across multiple processors rather than expecting a single general-purpose computer to do it all — especially in applications when real-time* performance is required.

Introducing the Pacamara Ristretto

With the Pacamara you have high quality, configurable, algorithms running on dedicated, uncontended hardware at sample rate with high (clock) accuracy. The result is it sounds great and it’s reliable.

— Alan M Jackson

An Audio Processing Unit Powered by Kyma

I can do things in Kyma in minutes that would take days in [another visual programming environment for music]… and the results speak for themselves, with excellent sound quality out the box.

— Geoff Hood (My wires are unconventional), electronic musician

Kyma is a recombinant sound design environment that puts vast libraries of unique audio synthesis and processing modules at your fingertips, where you can combine them to quickly create fresh, original sounds that can’t be found anywhere else…

[Kyma is] frictionless for ideation and research. Then, reliable and consistent at the outcome stage.

— Cristian Vogel, sound designer and musician

Kyma also comes with a reactive parameter language for controllers (Capytalk), Meta modules (modules that construct other signal flow graphs), and generative machine learning (like Hidden Markov Models that can generate new sequences in the style of whatever MIDI files you feed into it). Start by modifying graphical signal flow diagrams from the extensive library. Then, if you want to go deeper, there’s more!


Powered by Kyma

Designed with inspiration in mind…

Learn more

What’s a Pacamara?

Anything you want it to be! Use the Kyma software to configure a Pacamara to function as an outboard sampler, as a synthesizer (including additive and aggregate synthesis patches that you won’t find anywhere else), as an outboard effects processor (including live spectral analysis and processing), or as an interactive sound-effects generator.

But things really start to get interesting when you create combinations of all of the above; that’s when you know you’ve left “plug-in” territory and have started blazing your own trail into the unknown, developing new synthesis and processing algorithms. Using the graphical signal flow editor, it’s easy to create new sound processing and synthesis algorithms that no one has ever heard before.

Develop your signature sound

Having a palette of your own encapsulations to use means that…a team of sound designers…could help a studio build its unique sound. It makes Kyma even more of a “secret advantage”.

— Alan M Jackson

The flexibility and recombinant modularity of the Kyma environment gives you the freedom to create new sounds that have literally never been heard before. Save your new modules in your personal library so you can recombine them into new algorithms when you work on future projects. Over time, your copy of the Kyma sound design environment evolves to represent your signature sound — with unique acoustic qualities and your own exclusive sonic fingerprint.

But Kyma is more than a simple graphical patch editor. With the Timeline or the Multigrid, you can create virtual synthesizer/processors that evolve over time. Imagine performing an instrument that can change its acoustic properties and the way it responds to your input over time or in response to the way you’re performing.

How do I integrate a Pacamara into my studio workflow?

When you connect the Pacamara’s USB-C port to a USB port on your Windows or macOS computer, the host computer sees the Pacamara as an audio and MIDI interface. So you can stream audio and MIDI between the Pacamara and audio apps running on your computer.

This means you can do things like:

  • Play Kyma through the host computer’s audio output device (or internal speaker)
  • Process computer audio through Kyma effects
  • Treat the Pacamara as an external audio effect through an External Audio Effect plug-in in your DAW
  • Send MIDI events from your sequencer to the Pacamara to treat it as an external synthesizer
  • Record Kyma outputs into a digital audio editor/recording application running on the host computer
  • Capture algorithmically-generated MIDI output from Kyma in music notation software running on the host computer
  • Capture a screen recording with audio from the Pacamara
  • Send high quality stereo audio from Pacamara during a Zoom meeting with clients or collaborators

and otherwise integrate the audio signals generated and processed through the Pacamara with any audio application running on your computer.

Kyma erases the boundaries between studio and stage, FX processor and instrument

— Scott Miller, composer-improviser-professor

Alternatively, you can use a USB class compliant digital mixer as the audio interface for the Pacamara. If you connect the Pacamara Ristretto to that USB port, treating the digital mixer as the Pacamara’s audio interface, the Pacamara can access the mixer and all other I/O through the USB interface and internal routing. In this configuration, the Pacamara can function as an 8 channel EFX workstation for the rest of the studio. Connecting any Pacamara USB to a Metric Halo’s SCP (satellite computer port), you can integrate all of your outboard gear (for example, a Buchla system and Kyma) in the MIO Console.

Are you part of a sound design team?

With [other audio software] and plugins you can get into all sorts of versioning and compatibility issues. A setup that’s working fine on my computer might not work on a teammate’s. Kyma is a much more reliable environment. A Sound that runs on my Paca / Pacarana / Pacamara is going to work on yours.

— Alan M Jackson

Game studios, post production teams, research groups, student labs…can use Ethernet to access a shared Pacamara from another office, studio, or classroom or over the Internet from anywhere via remote desktop software.

How do I perform live with a Pacamara?

Pacamara is dedicated, reliable, and takes direction well — in short it is a perfect bandmate to include in your live performances.

Interactive controllers

Perform the Pacamara as a virtual electronic instrument by connecting it to class-compliant USB MIDI keyboards, faders, buttons, pedals, or continuous keyboard for highly responsive, nuanced control over any sound parameter. Use Capytalk (Kyma’s functional reactive control language) to define complex real-time mappings from combinations of controller values to sound parameter values.

Live instrument and vocal processing

For low-latency, high quality, multi-channel audio input and output, connect the Pacamara to a class compliant USB audio interface or digital mixer.

Kyma enabled me to augment the sonic possibilities of classical instruments and singers in a very convincing way. I found no other system that could cope with the expressiveness of classical instruments like Kyma does. Before I knew Kyma it was ‚an instrument and electronic sounds‘. With Kyma it is a convincing partnership and amalgamation of analog and electronic sounds.

— Franz Danksagmüller, live music performer / composer

Ready to tour?

I can bring my entire setup as carry on!

— Anne La Berge, musician, flute-player,improviser & artistic collaborator

Pacamara Ristretto is small (210 X 215 X 45 mm or 8.25 X 8.5 X 1.75 in), it’s light (1.268 kg or 2.8125 lb), it has a ruggedized case and connectors, and it fits into a backpack. There’s even a built-in stereo headphone jack for sound-checks and working in hotel rooms.

Don’t trust the venue Wi-Fi?

When you’re performing with wireless OSC controllers, use Pacamara’s built-in Wi-Fi access point so you’re not at the whim of the shared wireless network in the venue.

Did you forget to pack your power supply?

I’m on vacation right now and it’s such a joy to sit with nothing but a Ristretto, a laptop and headphones and be able to work!

— Bruno Liberda, composer, performer

If you forget your power supply, it’s not a show stopper. Any USB power supply (45 W or greater) will work; in a pinch, you can even borrow the supply from your laptop.

Better yet, use a portable charger/power bank and the headphone jack to continue tweaking on the your way to the gig or to work in the airport during a long layover.

With a portable power bank, you can commune with nature by working outside, or do your creative coding in the local coffee shop, while sipping your favorite pacamara-based beverage.

Improvements over the previous generation

It’s the latest, most powerful, revision of the most important audio DSP environment of the last 30 years.

— Alan M Jackson

The Pacamara Ristretto is 2.5 times more powerful and less than half (44%) the volume of the previous generation APU, meaning that the computational performance-by-volume has increased 5.7-fold.

The Pacamara Ristretto name was inspired by a hybrid coffee plant bred to produce larger beans from smaller plants that can be planted closer together, and “ristretto”, an intensive shot of espresso that delivers twice the caffeine per unit volume.

Hardware that gets better with time

There’s a subtle and important difference between hardware that is upgradeable… and hardware that DOES get upgraded continuously and significantly. Kyma is Symbolic Sound’s only product and they’ve been working on it, passionately, for decades. [They] have a proven history of constant and significant upgrades… a proven dedication, over time, to improving a single product.

— Alan M Jackson

Traditional audio hardware has a fixed function that never changes. A Pacamara acquires significant new functionality over time through frequent software updates. The end result? Your Pacamara actually grows better & better with time!

Pacamara — your secret advantage

Whether you design audio for games, AR/VR, sound for picture, you’re mapping data to sound to better understand the universe…

Whether you’re creating sound art, working as a user-experience (UX) designer, or you design sounds with healing properties…

Whether you produce podcasts, or are creating new forms of audio narratives, or audio journalism…

Or if you compose multichannel digital synthesis and processing environments for live performance…

Your reputation depends, in part, on having the tools you need to create the freshest sounds with the highest audio quality, the lowest latency, the most advanced algorithms, all with rock solid dependability.

How to know when it’s time to get Kyma: You like the flexibility of software but prefer the sound of hardware devices. You like to work with digital modular environments but want to do complex sounds without creating hundreds of wires and modules. You want to mix control and audio rate processing in an easy way. You need a solid low latency system for live performance.

— Christian Schlösser, musician, composer, audio engineer and sound designer

Satisfy your sound obsession

It’s time to explore new sound spaces and to forge a path into uncharted territories of sonic experience. It’s time for a Pacamara.

Place an Order

* In real-time computing, a result is guaranteed to be delivered within a specified time constraint (such as the period of the audio sample rate clock). Examples of applications requiring real-time computing include fly-by-wire aircraft, the control system for anti-lock braking in cars, and digital audio signal processing of a live input. Even a “high-performance” computer is not guaranteed to operate in “real time”; it may operate in bursts — sometimes faster than real time but sometimes slower — and any time it misses a “deadline”, you hear the result as a pop or a glitch in the audio output.