Originally Posted by Mendu
Well, not really. Stat muxing just takes the available bandwidth and tries to be smart about allocating it between competing streams.
Right now we're getting HD at ~18.5mbits/sec from KDLT; this will change to about a 15/3 split with WeatherPlus. When the sub-channel doesn't need its full 3mbits/sec, some of that will go back to the HD feed to temporarily give it a bit more than 15mbits/sec. The BEST you can hope for on the HD stream is that you'll get the same 18.5 mbits/sec we're getting now, but that's unlikely to happen as there will always be some bandwidth needed for the sub-channel.
Maybe they'll do better with their 1080i than KELO does, we'll have to see. But the HD will definitely be degraded somewhat from what it is today.
Now if you were to say that they're getting a new encoder as well, that would be more interesting. But a statistical mux plus a sub-channel isn't going to make any positive difference; it will only lessen the negative effect of adding a channel if we're lucky.
You dont go down to a 15/3 split...you go to a 15 to 18/1.5 to 3 split. Big difference.
And its not "when the sub doesnt need a full 3...". The HD has priority so it would be more like "When the HD does not need the 18".
And the HD signal seldom used the full 18 anyway. When it does, the SD would degrade before the HD would.
Also, You do not need a new encoder...you need a new compression scheme..which stat muxing brings.
If you did a recording pre-stat muxing and compare it to after, you probably wont notice the dif.
Of course...this is under the assumption that they are using stat muxing. The NBC suggested package includes a stat multiplexer but they can go with a configuration that does not use it. So you may be right. it may be noticible. In which case....never mind.