A basic OpenCL application
A common example of a GPGPU application is one which calculates the Fast Fourier Transform (FFT). This algorithm is commonly used for audio processing and similar, allowing you to transform, for example, from the time domain to the frequency domain for analysis purposes.
What it does is apply a divide and conquer approach to a dataset, in order to calculate the DFT (Discrete Fourier Transform). It does this by splitting the input sequence into a fixed, small number of smaller subsequences, computing their DFT, and assembling these outputs in order to compose the final sequence.
This is fairly advanced mathematics, but suffice it to say that what makes it so ideal for GPGPU is that it's a highly-parallel algorithm, employing the subdivision of data in order to speed up the calculating of the DFT, as visualized in this graphic:
Each OpenCL application consists of at least two parts: the C++ code that sets up and configures the OpenCL instance, and the actual OpenCL...