There are several different low pass filters, and as many high pass, band pass, band stop… filters. In Audio toolkit, there are different usual implementation available:
and it is possible to implement other, different orders as well…
I’m happy to announce the release of a side-chain stereo compressor based on the Audio Toolkit. It is available on Windows and OS X (min. 10.8) in different formats. This stereo compressor can work on two channels, left/right or middle/side, possibly in linked mode (only one set of parameters), and can be set up to mix the input signal with the compressed signal (serial/parallel compression). The side chain channels can be used to steer the gain stage (the same setup will be used, right/left or middle/side).
Focus on this release was on performance. As such the core functions were optimized, as well as some tools and EQ.
A new filter dedicated to fast convolution (using a fixed-size partition with a mix of FFT convolution and explicit FIR filter) with 0 latency was added.
After my post on HPCToolkit, I felt that I prefered QCacheGrind as a GUI to explore profiling results. So here is a gist with a Python script to convert XML HPCToolkit experiments to callgrind format: https://gist.github.com/mbrucher/6cad31e38beca770523b
For instance, this is a display of an Audio Toolkit test of L2 cache misses:
Enjoy!
I’ve started working with the HPC Toolkit, especially the part where it can use PAPI to read hardware counters and display the result in a nice GUI.
Convolution is an algorithm that is often used for reverberations. If the equation is easy to understand and to implement, the implementation is costly. The other way of doing it is to use Fast Fourier Transform (FFT), but the direct/crude implementation requires latency. If it is possible to optimize the basic convolution code, it is sometimes more interesting to use a different algorithm, as it is the case here.