alastairhmoore / jmatconvol

Mirror of code from https://kokkinizita.linuxaudio.org/linuxaudio/index.html

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Release 0.5.2 (5 Apr. 2022)
----------------------------

* Fixed allocation in fmatconvol.


Release 0.5.1 (1 Dec. 2021)
----------------------------

* Added input and out gain factors.


Release 0.4.2 (29 Jun. 2021)
----------------------------

* Added -T option.

* Fixed typos in Makefile.


Release 0.4.0 (31 Jan. 2021)
----------------------------

* Added file processor, fmatconvol. This accepts
  the same config files as jmatconvol, and any
  reference to port names etc. are ignored.

* Added commands to read the entire matrix
  from a single audio file, see README_CONFIG.

* Bugfixes, empty connection names are now
  handled correctly.
 

Release notes for jmatconvol
----------------------------

Jmatconvol is a real-time convolution engine. In
contrast to jconvolver it uses a single partition
size equal to the Jack period, and is optimised
for dense matrices of short convolutions, e.g.
for processing signals from spherical microphones
such as the Eigenmike. The maximum convolution
length is limited to 4096 samples in this release.

For any other type of use (sparse matrics, longer
convolutions,...) jconvolver is probably a better
choice.

Jmatconvol can compute up to a 64 by 64 convolution
matrix (i.e. 4096 simultaneous convolutions) as long
as your CPU(s) can handle the load. The computation
work is performed by a user configurable number of
parallel threads, allowing to distribute the load
over all available CPUs.

For configuration details, see README_CONFIG

-- 
FA

About

Mirror of code from https://kokkinizita.linuxaudio.org/linuxaudio/index.html

License:GNU General Public License v3.0


Languages

Language:C++ 94.0%Language:C 3.1%Language:Makefile 2.9%