warren.mcdonald

Hi Miska, Do you have a debian arm package for v3 that I can download. I have all of my cubox setup and was running v2. I just need the v3 package now. I can't seem to find it in the download file structure. Thanks, Warren

Hi all, great thread, interesting, open and respectful treatment of ideas. On the un-bundling of complex audio files:- This may not be as kooky as it sounds. Firstly let me agree that the file as collections of bits are not being changed. What may be changing is underlying computer activty. What we have to consider here, and Peter would, I think, agree, is that we are dealing with computers that represent complex systems. Complex systems, like people, exhibit behaviour that cannot be explained by simple normal rules that we expect they should follow. We are not aware or able to completely assess the impact of what is going in the background. In this case it is quite likely that computer systems internal behaviour, in memory management particularly, could impact the quality perceived by the listener. Repetitive use of an application and accessing of a data file will, by design, likely result in the computer providing a lower latency, higher priority treatment for this activity. The irregular performance of PCs and Macs, that do not expect to run in real time modes, involve swapping applications around in memory and loading files off disk instead of out of cached memory, can impact the playback in subtle ways that are difficult to describe in words. Playing a track even twice in a row can change the performance for that task and therefore, quite possibly, the integrity of the output. Repetitive use of an application may invoke "most commonly used" status which is respected in some memory handling algorithms. Depending on the systems base performance and other running software this affect may be totally unnoticeable on some systems but glaringly obvious in others. Applications like XX high end incorporate this knowledge in their design to offer more consistent listening results. The use of Kernel Streaming drivers is another technique used to avoid run time variations in quality. On the impact of ultrasonic signal components:- The presence of ultrasonic signals and their inherent sub-harmonics most definitely contribute to the character of musical instruments. The ultrasonic and sub-harmonics modify the audible components that we can hear. If they are absent then the overall signal has lower fidelity, in the true sense of the word. This is another complex area where the original recording microphone has a lot of responsibility. Some would argue that any impact of the ultrasonic is already captured in the audible components as measured at the microphone, I would say it is not as simple as that. On the 176k or 192k sample rate argument:- Yes please. Pushing any digital artifacts well above the near ultrasonic range just helps this problem. Cheers, Warren