[opendtv] Re: [lptv] FYI NEWS There Goes Analog
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: "OpenDTV (E-mail)" <opendtv@xxxxxxxxxxxxx>
- Date: Sun, 17 Apr 2005 19:49:13 -0400
John Shutt wrote:
> And you thought 6 towers covering Manhattan in an SFN was
> a nutty and unproven idea...
You're misrepresenting my objections completely, John. I
never said anything about "unproven," for one thing. That
belongs to the crowd that needs to get hit on the head with
a 2 by 4 before they understand whether something should
work or not. Also belongs to the crowd that believes that
"theory" means "unsupported opinion." Count me out.
My objections to the descriptions we get on "well designed"
infrastructures are that they are incomplete and mostly
inaccurate. And they are vague enough that anyone can drive
a Mack truck through the inaccuracies. You can always weasel
out of something if you make it vague enough.
1. The low towers and medium or low power of each
transmitter are not enough to cover typical US markets.
2. Use of on channel repeaters at the fringes, to expand
this inadequate coverage, is inefficient and difficult to
implement properly. I've been over this already.
OCRs are best used in places where the main transmitter
can overpower the OCR signal all around the OCR's coverage
area. If used at the fringes of the market, the power of
the repeater either has to be very low, or with somewhat
higher power, you would have to use a directional antenna to
direct the beam downstream of the main transmitter. Very
little OCR signal can go back toward the main transmitter,
and even in a direction 90 degrees from a radial from the
main transmitter, the OCR signal has to be controlled.
So either of these solutions for OCRs at the fringes will
greatly reduce the coverage area of each OCR. And this true
with COFDM too. The problem is that you cannot exceed the
echo tolerance of receivers.
So that's why in practice you either have to synchronize
all these sticks, which no one seems to want to do, or you
have to use translators. And, of course, the problem
becomes ridiculous if you really want to create the sharp
boundaries some seem to want. But without the sharp
boundary requirement, it's doable. Just expensive.
So this has nothing to do with "unproven." It has to do
with taking tradeoffs into account. You know, looking at
the whole problem.
Bert
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
