Re: Phase/skew sensitivity



Hi Jonas,

On 5/26/2012 5:07 AM, Jonas Eckerman wrote:
On 2012-05-24 12:22, Don Y wrote:

audio library AND RUN THE AUDIO SERVER 24/7 without an
annoying fan, etc., doing the same with a video library
would be impractical. You would *want* to power it
down when it wasn't actively serving video!

This suggests audio and video would probably like to
originate from different physical servers.

To me it suggests that the audio and video could probably originate from
different storage attached to the same server, where the video storage
can be external to the server (using a dedicated gigabit ethernet
connection for example).

Or firewire, USB3, SCSI, etc.

When you want don't need the video, just unmount and power down the
video storage system.

The effort required to process and distribute video is
considerably more involved than audio. Rather than
designing *a* box that can handle audio and video
concurrently -- and not use or need the video portion
much of the time (imagine using such a system in audio
ONLY applications!) -- design something that can do each
individual thing, well.

E.g., the video system might want to be able to record video
as well (imagine the video source being an over-the-air
broadcast that you are pushing down the network "live";
yet, also recording to allow things like "pausing live
video"). So, that system has needs that the audio might
not have. Or, on a scale that the audio might not need
to support.

I wouldn't waste any money (personally) ensuring that
I could watch movies during a power outage. On the other
hand, I'd be annoyed if I had to sit in a silent house
until the power returned! I wouldn't want to have to
provide backup power to a system that is *capable* of
processing video in real-time (think: fancier CPU) when
all I really need is something that can push bytes off
a solid state disk (or some other read-only medium)
out to "speakers".

Regardless, though, the video has sound encoded at 48KHz
while the audio has sound at 44KHz. Do the network
speakers support *both* sample rates? Just one? What
if you want to mix "regular audio" with "audio from video"?
Do you do that in the speakers? Or, in the audio server?
etc.

These are the sorts of questions that muddy an "audio only"
implementation!
.