In terms of numbers of audio channels, there is no theoretical limit within any single Telestra 4 Target platform, since our internal architecture scales more or less infinitely, the practical limit being imposed only by processing power. However, this factor is mitigated by the design of the ASTiOS realtime framework, in that it scales across multiple processing cores taking advantage of the latest multi-core CPUs. Hence, realistically the number is very high, certainly hundreds of channels is not at all unrealistic.

Furthermore, the ACENet audio distribution architecture allows any single Target to link to any number of other Targets to form a processing cluster, and the channel counts begin to become so large that it is almost a meaningless question. More practically, we would prefer to develop the most elegant solution in terms of architecture and processing configuration, in order to keep the raw channel count low, since the higher the channel count, the more demanding the test plan must be to prove the solution is working under all combinations of signal path.

We do not have the concept of a narrowband channel - all audio channels are sampled at 48kHz/16bit. We can easily represent narrowband channels through the use of filtering to achieve the required simulation effects.