Our technology provides seamless support of real-time preamplifier, equalizer, compressor, reverb, multi-effect, stomp-boxes, cabinets, microphones and tapes emulations on Intel based machines, both for Microsoft and Apple OS. It can also take advantage of CUDA parallel GPU computing architecture. The Acustica Acqua Engine is a combination of multiple advanced technical processes that are unique to the Acustica team. Effect devices can be succesfully sampled without further editing or adjustment, and then immediately processed and reproduced via the same engine, where the sampled data is stored and available for recall, loading, saving, and advanced editing when desired. The quality of reproduction of sampled sources are nearly indistinguishable from the originals. The following techniques are merged into a single model.

Nonlinear convolution: The Volterra theorem is applied as a complete re-thinking, replacement, and generalized substitution of the convolution algorithm that is commonly used in audio applications. Accurate sampling, processing, and reproduction of up to 9 harmonics is possible with our proprietary technology.


Vectorial Volterra Kernels Technology (V.V.K.T.): Volterra kernels are stored in tree data structures (managing up to 100000 elements in real-time using a CPU Pentium IV 3 GHz). The Acqua Engine is capable of implementing a list of modules commonly used in audio synthesizers (LFO, envelope followers, dynamic modules, FUNS). Multiple combinations of these processes may be applied to control sources and destinations.


Dynamic Volterra Kernels Technology (D.V.K.T.): Volterra kernels are applied in real-time within the Acqua Engine, depending on the assessment of input and output audio streams. Processing can be applied down to a single audio sample. This is similar to algorithmic convolution, but much more advanced in that it may be applied to varying, constantly updating impulse responses.

Time Varying Models (T.M.V.): A collection of kernels collect data using an advanced sampling technique, creating a multi-dimensional snapshot of a nonlinear/time-varying system. Multiple recordings are interpolated in order to mimic the time evolution and response to external variables such as user parameters and input/output assessments (e.g. time-varying cyclic effect processors , stomp-boxes, digital multi-effect units).


The Acqua Engine introduces further optimizations and improvements on the basic design and architecture. Our attention is constantly being directed towards unexplored subjects, maintaining our vision for providing the ultimate sampling-based processing approach.

Split Hybrid (SPLIT-H.): Different algorithms can be used for early reflections and tails. The application of FFT or CUDA improves processing performance, disclosing to emulation of heterogeneous chains of reverb/effects processors.

Smooth3: Artifacts caused by a fast kernel switching are minimized using different types of interpolation, which is executed per sample even when block-based algorithms (e.g. FFT) are involved.

Stone: Minimum phase filters are applied on sampled kernels in order to minimize high/low frequencies roll-off, to fix issues caused by poor AD/DA converters or by the intrinsic behaviour of the target system (e.g. tape sampling).


Raw Data Management: Raw user data (e.g. generic envelopes, envelope followers) can be stored in tree data structures and used as control sources and destinations, allowing complex program-dependant emulations.

Multi-Threading Support

To get the most out of modern multi-core processor architectures, CPU intensive sub-systems of the Acqua Engine Core have been rewritten in order to support multi-threading.

Cuda: Nvidia CUDA technology enables the use of their GPU (graphics processing units) to solve complex computational problems. CUDA (Compute Unified Device Architecture) technology gives computationally intensive applications access to the tremendous processing power of Nvidia GPUs through a revolutionary new programming interface. Providing orders of magnitude more performance and simplifying software development by using the standard C language, CUDA technology enables developers like Acustica to create innovative solutions for data-intensive problems. For advanced research and language development, CUDA includes a low level assembly language layer and driver interface.


The Acqua Engine VST Plug-in is "Designed For CUDA", which means that it is able to take advantage of this technology to provide enhanced performance for those who use Nvidia graphics cards that are CUDA enabled. It is an advancement in GPU development, software technology, and a partnership where everybody gains something valuable from this process. Nebula has a source beyond limitations of the host computer to draw extra processing power from, and the customer has more options and flexibility as a result. It is just one more example of how Acustica is always working to take advantage of the best technology that is available.

http://www.acustica-audio.com

To learn more, please visit our Designed For CUDA technology