> > That sounds like you'd like more than one audio mixer system wide, one >> for each audio output, with only one visual interface. Though the API >> isn't meant for this use case IMO. Moreover one media node is slaved >> to only one time-source, which contradicts with being synced to more >> than one physical output/time-source. >> > Windows 7 has separate user-visible mixers. I don't know about under the hood. I have separate mixers for my HDMI Audio, and for the normal HDA Audio (headphone/mic jacks). Each has a master level, and then the per-app volume controls. This, to me, seems like the right approach. In Haiku, I even get two HDA Audio nodes on the left panel (although the HDMI audio itself is still broken/unsupported). So in theory, Haiku should be able to do all of this. What would be neat, however, is to be able to drag'n'drop an output from one device mixer to the other. If you wanted a UI like Dane's mockup, you'd have to provide the master controls from all audio outputs on that mixer pane; and then you'd need the mixer to no longer belong to a device node as it currently is. Anyway, most of the issues here seem to be mostly UI-centric to me (for example, the Media prefs doesn't handle display of multiple mixers properly).