[BC] Sangean HD component tuner and Infusion portable Internet radio

[Dana Puopolo]

Bob,

I used to be CE of a beautiful music station with an 18 share. We did not have
to run the processing "wide open". We used an Optimod 8000A with a Compellor
in front of it set for a minumum of processing. We used a composite STL with a
MSI clipper set only to get overshoots in the STL. Every piece of equipment
was tweaked for the ultimate possible quality. We probably had the first 100%
transformerless radio station in the country. Paul Temple, Grady Moates and
myself all made sure that EVERYTHING was always 100% "on spec". The reason we
used any processing at all was because in cars we needed our average
modulation to be high enough to overcome road noise. 

We were perhaps 3 db less loud then the loudest station on the dial there,
WPRO-FM. 3 measly db. Pro-FM NEVER 'jumped' out at you when switching between
stations. They were louder, but square waves always are (they were proud of
the fact that they cranked the level into the safety clippers on their 8100 up
3 db). They sounded like CRUD, had less then HALF our numbers, and while their
cume was good, their AQH sucked!

The notion that someone should take a medium that has perhaps 80 db of dynamic
range and compress and limit it until is runs within a couple db of the
CRUSHED main channel is preposterous! Why bother HAVING all that range if
you're not going to take some advantage of it?  The one GOOD thing I've heard
from FM HD is the lack of noise. To me at least, peak limting (and clipping)
does the most damage to audio of anything. Or do people buy 100 watt per
channel amps for nothing? TRANSIENTS perhaps more then anything are what adds
to the realism of music!

Yes, cutting peak limiting back from perhaps 6-10 db to five will make things
have more 'punch'. BUT...eliminating it all together will make it even
punchier! Not to mention that conceptual coding algorithyms work best with
uncompressed audio as their source.

Now, I understand how look ahead limiting works - and it's a good invention
and innovation...BUT just like any good thing, it can be overused. 5 db of any
limiting in a medium with a wide dymanic range like HD is about 4 db too
much!

The bottom line is this: I have heard all the HD stations in three major
markets at length (NYC, Philadelphia and Boston). With a few standout
exceptions, most HD audio sounds unimpressive. Why? BECAUSE of the
processing!!  There's no "WOW!" factor!

AND...In the case of the standout stations, EVERY ONE is a few db softer then
their analog main channels. Secondaries too. I'm talking about audio that
sounds noticeably BETTER then the main analog channel. Who cares if I have to
turn the radio up a bit? I'd do it anyway, because the music DRAWS ME IN when
it sounds really good, instead of pushing me away, as crushed, clipped analog
FM does.

If we are going to claim 'CD like audio' for HD FM radio, we'd better deliver
- and the crushed, limited MESS I hear on most stations just doesn't cut it!

-D

 
------ Original Message ------
Received: Wed, 20 Dec 2006 09:15:31 PM EST
From: Robert Orban <rorban at earthlink.net>
To: "Broadcasters' Mailing List" <broadcast at radiolists.net>
Subject: Re: [BC] Sangean HD component tuner and Infusion	portable	Internet
radio

At 04:50 PM 12/20/2006, Dana Puopolo wrote:
>Bob,
>
>Why would anyone employ peak limiting or any kind of aggressive processing
on
>HD? It makes absolutely no sense! Run the gain down 8-10 db and keep the
peak
>limiting OFF!

If you did this on the HD1 channel, the crossfade between the analog and 
digital channels would cause the digital channel loudness to drop 
dramatically and annoyingly and the analog channel to blast at the listener 
who had just readjusted her volume control for the digital channel.

If you're talking about an HD2 channel, maintaining approximate loudness 
parity between the HD1 and HD2 channels avoids annoying volume changes when 
the consumer switches between channels. Having a station "blast" at you is 
particularly annoying. We know this from experience with television, where 
commercials that are significantly louder than program material have 
generated spontaneous consumer complaints to the FCC as early as the 1950s. 
"Joe Sixpack" prefers comfortably consistent program loudness and does 
*not* want to have to constantly fiddle with his volume control.

Fortunately, the amount of digital-channel peak limiting required to 
maintain parity between the analog and digital channels is not very large. 
If you use a good look-ahead technology, you will hear no overt artifacts 
and only the slightest loss of transient punch compared to the source. To 
put this in perspective, the amount of digital channel peak limiting 
required to achieve analog/digital loudness parity is significantly less 
than the amount of peak limiting routinely applied to today's CD releases 
in mastering.

Bob Orban 


_______________________________________________

The BROADCAST [BC] list is sponsored by SystemsStore On-Line Sales
Cable-Connectors-Blocks-Racks-Wire Management-Test Gear-Tools and More! 
www.SystemsStore.com       Tel: 407-656-3719    Sales at SystemsStore.com