Preface

The idea of of an experimental up-to-date audio system for computer, which the main processor relieves, and a hardware environment which becomes pleasant for various applications in the real time processing (Live performances) as computer music workstation or as measuring system for audio technique, resulted not only from the many inadequacies, which the hardware offered at present, but also to make a radical step for the integration of audio into the computer. Particularly for the hardware development the missing transparency in the form of documents for programming existing audio hardware and for a dynamically software development made the decision for this development.

A further goal of this hardware is in the education of audio engineering students to build experimental INTERFACES, for installation in the area of the sound art (sonic art) with extreme request on channel numbers and automation.

The system should be both an audio INTERFACE and an acceleration card for audio calculations. An attention to modularity and different stages of development was put, which one permits ''Low Cost'' - solution and highly qualitative products for the area, measuring technique and performances with many channels as well as adapted hardware for installations within the area of the computer music and new media.

The audio hardware is not necessary necessarily and their function could also by other sound cards be partially fulfilled. It is replaced, hopefully in the future from developments at the market, which offers its possibilities. The software represents a modular kernel, particularly for Linux (Unix) systems from the i386-Rechnern up to Dec alpha workstations, which can be controlled and cared for from a management program.

  

Figure 1: structure of MAAS

The total system represented in figure1 shows the structuring of software and hardware, whereby non-existent DSP modules can be simulated by the computer under application (in a device driver) taking a part of the host cpu-performance.

The hard- and software is released for not commercial purposes for free, to participate in order to thus ensure an uncomplicated distribution and to enable a dynamic development of the system for all users and developers, offering the possibility of a reproduction or a further development to everyone.

Software and hardware modules should be simple to integrate and not disturbing the rules of the system, so in addition guidelines are created, in order to guarantee this development.

Here it is still mentioned that again and again the emerging criticism ''nowadays hardware develop is unreasonably, since it is easy to buy soon appropriate cheaply hardware'', in this sense is surely correct. To do this nevertheless results among other things from the following considerations:

• from the view that knowledge over problems and power limits in the computer music arise only as a result of the process of development.

• and this project mainly is target for the development of this area for lectures on the technical University in Graz and the University of music Graz.

• that this project mainly serves as basis for special solutions in the research and in the art.

  For example a continuous installation with 64-Kanaelen audio (loudspeaker matrix) or 172 8-bits channels for telephone speakers or with special input modules that anyway have to be built (e.g. large switches, sound cabs, etc..). Standard modules to usually fail and only by the self-development the project shows up the point for what is really meant und where it becomes the Unikat. So ALL-TOO much compromises in the hardware decision, particularly in the art, leads usually to spongelike, no longer comprehensible solutions.

• over one year long unsuccessfully search for appropriate hardware, which could be programmed free according to my request (programmable DSP system), because of

  • inflexible hardware
  • missing documents over the hardware, or no warranty that the hardware in the next versions is compatible and the upgrade software does not become an endless search.

  • complicated not acceptable licenses of softwarepackages, where the expenditure brings more administration waork during distribution of own software (as PD) than the effort of developping itself or the user of the PD software must buy some software to use this.

  • intolerable prices with professional systems for application to universities with limited budget or without budget.

If there is no according hardware in the future to buy, then everyone can build itself these to be independentf of others.