Danny’s comments about the Input device are on point…
From a product standpoint, I want what you’re suggesting, and also more.
There are some technical questions about how to handle latency and video sync. RAAT does do a fair amount of buffering. These aren’t an issue for radio or phonograph sources, but for computer audio or TV’s, it’s more difficult. So is emulating sound cards in a way that is convincing to a wide range of mainstream apps.
There are some really interesting product possibilities that come from supporting mixing in addition to custom input options.
For example, for a desktop/office endpoint product, I really want 2 inputs and 2 outputs:
Inputs:
- Computer/notification sounds/youtube
- Music
Outputs
- Speakers
- Headphones
I’m not sure this product needs output matrix functionality (i.e. using speakers + headphones at the same time), but I definitely want to hear both notification sounds and music at once. And I want the notification sounds to have no impact on SQ unless they are actually happening. The usual compromise is to use the “OS Mixer” and run the whole computer at one sample rate/configuration, and then normalize everything to that, which means you’re living in a degraded state 100% of the time just so you can hear skype ring a few times a day.
Slaving the notification sound parameters to the format configuration of the music would be an improvement, but the most elegant solutions only become possible when you start architecting hardware around them.
I have thought through another product that does want to use headphones + speakers at the same time for different streams: a bedside radio with integrated headphone amp. If I fall asleep listening to music in headphones and forget to turn it off, I still want the alarm to play through the speakers in the morning to wake me up, even if I left the system “switched” to the headphones zone the night before, and even if the headphones are still playing.
Complicated little problems that require end-to-end product design. The “input device” is a small step in the right direction, but the more flexible possibilities are (to me) more interesting.
(As a side note, everything in my office is plugged into a 16ch mixing board that drives a pair of powered near-field monitors. So everything comes out the same speakers with no switching hassles…but this isn’t a practical setup for most…it has sated my appetite for more polished solutions).