[BC] Seedy Quality in IBUZ Ads

[Robert Orban]

At 07:19 PM 8/26/2006, you wrote:
>From: "Mark Humphrey" <mark3xy at gmail.com>
>Subject: Re: [BC] Seedy Quality in IBUZ Ads
>To: "Broadcasters' Mailing List" <broadcast at radiolists.net>
>Message-ID:
>         <74b029b80608261649p72e20fbldd1224677537bce1 at mail.gmail.com>
>Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>On 8/26/06, Richard Fry <rfry at adams.net> wrote:
> >
> > If HD FM sounds better than analog FM, isn't that related at least in part
> > to the relative distortion of the modulation processors used to feed the FM
> > analog and HD exciter inputs?
>
>That's exactly the problem -- most FM processors must employ some form
>of clipping to maintain high legal modulation, as necessary to
>maximize the audio signal-to-noise ratio at the receiver.  To me,
>clipping is sort of like capitalism -- it has a lot of drawbacks, but
>I guess it's the best alternative we've found so far!
>
>Most FM processors can be set up to eliminate clipping, and the
>pre-emphasis limiting can also be defeated.  However, if the output is
>also set to keep peak modulation within legal limits, the "apparent"
>noise floor on a typical receiver will increase *at least* 10 dB, and
>with some program material as much as 15 or 20.  This wouldn't be a
>major problem if all receivers were fed a nice, strong,
>dead-full-quieting signal, but sadly that's not the case in the real
>world.
>
>When I was attending high school near Syracuse, back in the early
>'70s, the local 107.9 was a "purist" classical station called WONO.
>They used no processing whatsoever -- the studio console fed directly
>into a pair of equalized phone lines, which fed the stereo generator/
>transmitter several miles away -- which meant their modulation
>(including stereo pilot) typically hovered around 15 to 25%.   With an
>outside antenna and high-quality tuner, their signal could sound
>nearly as quiet as the recording --- but on many portable receivers or
>car radios, the background noise was quite objectionable.
>
>Nice idea in theory, but it just didn't work in practice.

Yep.

Even with minimal peak limiting, the average modulation of a typical major 
market FM station would have to be decreased anywhere from 6-10 dB (program 
material dependent) to achieve no audible HF loss caused by preemphasis 
limiting. No commercial broadcaster is going to take that kind of hit.

Then there is the whole multipath distortion issue. There is a reason why 
it was virtually impossible to do an FCC mandated proof of performance from 
a remote receiver back in the day. Even when the signal is not falling 
apart audibly, measurements typically show surprisingly high THD. The fact 
that FM listening subjectively satisfies a lot of people says something 
about the human ear's insensitivity to small amounts of nonlinear 
distortion. But if FM were the "new technology," people on this list would 
be making sarcastic cracks like, "That newsfangled FM system works really 
well -- as long as you can connect your receiver directly to the 
transmitter's sampling loop! But in the real world of mobile, there's 
nothing but spitting, picket-fencing, random blasts of multipath 
distortion, and a stereo image that's constantly expanding and contracting 
as the receiver blends. Great system, that FM!"

Bob Orban

Bob Orban