last updated 2006-08-22

Table of contents


AlmusVCU 0.87. Quick and dirty release. I was planning to hold this until I was complete packaging reverb programs and updating the documentation, which has been postponed in, well, two years. Now when people start to play around with 0.86 I thought I'd better release what I have now. I have not done much pre-testing of this, so if it does not work, let me know and/or try out 0.86 instead.

What is still missing in this release apart from up to date documentation is reverb programs and some really good cross-talk cancellation programs. I have the raw material for this, but I need to take some time to test, make a package and apply a suitable license acceptable by the authors of the raw material and me. This time is extremely hard to find, so I cannot make any promises.

Even without this, AlmusVCU can be a very useful program, and the differences from 0.85 to 0.87 is mostly bug fixes, so the documentation is quite ok in its current state. If you want to do high quality Ambiophonics with reverb, you need to wait for the release of reverb programs and cross-talk cancellation programs (or make your own, the filter program format is documented).

AlmusVCU 0.86. Added support for multiple input and output devices. This means that AlmusVCU can handle several sound cards both for input and output.

This project may have seem rather dead, but it has not been. It is true that I have not worked on it much for about six months, but since the last release almost two years ago, a lot has happened. This new release is a totally new product. The reason it has not been released earlier, is that I have not had any reverb programs available that I could release. This is however about to change, and therefore I have decided to make another public release of AlmusVCU. Suitable reverb programs will appear on this page at a later time (hopefully soon).

A serious bug which caused buffer underflow with large I/O-delays was discovered in BruteFIR and has been fixed. The new version is BruteFIR v0.99b.

Released AlmusVCU 0.7a, which fixes a few bugs. It is now also possible to start the convolver after configuration change without re-running the benchmark, if you think you know better. Additionally, a status section was added to this page, so I don't need to explain that the reverb engine is not very usable yet in every news event.

A major new release, AlmusVCU 0.7 is out. There is now support for Ambisonics and dynamic equalisers. The user interface has been improved, the source code structure as well, and there is an option for low latency operation. Still working on the part of providing good reverb programs to use though, so the reverb engine is not very usable yet, but all other components are.

A new release, AlmusVCU 0.6 is out. Still it is a very early release, and the stability may even have been reduced since the last version due to large feature extensions. Built-in equalisation design is now implemented, and it is now possible to have more than one ambiopole. There are still no good reverb programs to use with the software yet though.

Added a wish list, so I can remember which features I should implement in future releases. Still, crushing bugs is the most important though, and some features in the current release are not yet fully implemented.

AlmusVCU v0.5 has been released. This is an early release, meaning that the current state of the software is quite unstable. There is also not any good reverb programs available yet, so the software is not very usable at this moment.

Current status

The current release of AlmusVCU is a development release, meaning that not all functionality have been thoroughly tested, and thus may be broken.

Since the start of 2004, not much development effort has been put into the project. The software is fairly complete in terms of functionality, what is needed now is more users and discovering and fixing bugs.

Some ideas of future developments of AlmusVCU is work in the direction of wave field synthesis, optimal source distribution binaural systems, impulse response measurement and room equalisation. There is currently nothing planned though.

What is AlmusVCU?

AlmusVCU is an open-source software which makes a computer equipped with a multi-channel sound card into a real-time versatile convolver unit. The main area of use is as a surround sound processor in multi-channel high fidelity systems such as: The software aims at being useful for music lovers interested in the superior Ambiophonics and Ambisonics surround formats. However, rather than being tailored for a few specific uses, AlmusVCU is feature rich and very flexible, and also tries to be a valuable tool for audio hobbyists, students and researchers. AlmusVCU can be configured to collaborate with other audio processors, even be used as a reverb processor in recording post-production, or simply be used as a generic convolver unit for arbitrary audio convolution tasks. However, this documentation concentrates on its use as an Ambiophonics or Ambisonics surround sound processor, by providing tutorials on how to build such systems from scratch.

AlmusVCU runs on Linux-based operating systems [3]. It can either be installed and run on an ordinary Linux workstation, or on what it is designed for, a purpose-made embedded computer platform, forming a HiFi-component of the highest grade, at an unbeatable price.

The power of a modern standard computer platform allows AlmusVCU to outperform most custom-made DSP platforms in terms of throughput. At the cost of a small I/O-delay (about 20 -- 200 ms depending on application, configuration and hardware), AlmusVCU gives the audio processing power and sound quality of commercial hardware solutions costing 10 times more than the computer it requires. The software itself is free, licensed through the GNU General Public License [4].

The main features are:


The latest version of the manual is AlmusVCU user manual 0.85 and is available in PDF format. All documentation for the software is contained in the manual; this webpage only serves as a brief introduction, and a place to download from.

The manual contains the following chapters:


The following files are available for download:

If you have any questions just email me.


  1. Ambiophonics Institute website.
    The original Ambiophonics website.
  2. web site.
    An enthuisast website which promotes Ambisonics.
  3. Linux Online website.
    Linux is a free Unix-type operating system originally created by Linus Torvalds with the assistance of developers around the world.
  4. Free Software Foundation GNU General Public License.
    One of the most common free software licenses. Its main purpose is to make sure that the software is kept free and open source.
  5. A. Torger BruteFIR.
    A generic convolution engine which employs the partitioned convolution algorithm.

(c) Copyright 2002, 2004, 2006 - Anders Torger