[BC] Digital Intgerference (was AM Interference)

[Robert Meuser]

Bob:


I have to agree with Rich for the most part. You and Frank have built some big 
guns and people in the east know how to inflict maximum damage with them.  In a 
number of cases, enough energy is built up in the common multi band sections 
that in  digital there is a very slight change from analog and the CODEC seems 
pushed to its limits. We have one Spanish station that uses separate processing 
and the digital is way too bright.

NPR is also a disappointment, no multicasting on FM and the AM station sounds 
awful a lot of the time, too many cascaded CODECs and that is when the radio can 
decode the digital.

R

Robert Orban wrote:

> My post did not refer to AM. It referred to comparing HD FM to analog FM 
> reception. Rich Wood reported when he listened to HD FM stations in New 
> England, he could either hear no difference between the analog and 
> digital FM signals or the digital signal was badly processed by 
> comparison to the analog. He had nothing positive to say about his HD FM 
> listening experience.
> 
> In contrast, now that I have my Boston Acoustics Recepter, I verified 
> that on many San Francisco HD FM stations, one could hear significantly 
> more transient punch on drums and other percussion in rock and hip hop 
> music when the radio switched from analog FM to HD reception (to the 
> point where it sometimes sounded like a different mix), although the 
> basic tonal balance and non-percussive texture remained similar. This is 
> an entirely expected result if one is using an 8400 or 8500 to process 
> both the analog and digital signals because they share all processing 
> elements up to the final limiter chain. (In the 8500, the mix of the 
> five compressor/limiter bands can be set interdependently for the analog 
> and digital chains, but the default to have them be identical.) The HD 
> channel requires far less peak limiting than the analog channel to 
> achieve loudness parity during receiver crossfades. This translates as a 
> very audible increase in snap and punch on transients, most of which 
> survives the codec.
> 
> Getting equivalent punch from analog FM would require average modulation 
> to be reduced somewhere between 6 and 10 dB. This number is more than 5 
> dB because the FM pre-emphasis increases the slope and magnitude of 
> transient waveforms and make the processor's limiting/clipping system 
> work even harder than it would have to work if there were no pre-emphasis.
> 
> This result was no surprise to me, because I have been comparing the HD 
> and analog FM processed outputs of our processors for over three years 
> in the lab. What is still a mystery is why Rich Wood heard none of this 
> on any station he auditioned.
> 
> Bob Orban
> 
> 
> _______________________________________________
> This is the BROADCAST mailing list
> To send to the list, email: broadcast at radiolists.net
> For sub changes, archives and info on this other lists: 
> http://www.radiolists.net/
>