[opendtv] Re: Optimizing the system
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Fri, 17 Dec 2004 09:37:35 -0500
At 3:15 PM -0500 12/16/04, Manfredi, Albert E wrote:
>It's your next two points that continue to be non
>sequiturs.
>
>> 2. More than 90% of viewers will still be
>> watching the SDTV down conversion;
>
>Which is irrelevant by itself, even if true (which
>it is not anymore). If the DVD plays in their
>DVD players, or the signal fits in the 6 MHz TV
>channel, no need to limit the quality of the signal.
It IS still true, and will continue to be true for at least another
decade, if not longer. All that will change is that the percentage of
homes with HD capable displays will grow from the current <10% to
perhaps 25-30% within one to two decades.
You may well challenge this, however, keep in mind that any display
smaller than 40 inches does not need 720P resolution, much less
1080P. The percentage of homes that will buy sets smaller than 40
inch versus larger, is going to grow significantly, as people begin
to replace old NTSC receivers with new flat panel LCD TVs. But these
sets will not benefit from HD resolution because of the size of the
screens.
> > 3. Unless your display is larger than 100 inch
>> diagonal you WILL NOT see any difference.
>
>Which continues to be irrelevant to the problem
>of optimization.
NO BERT. THIS IS why it is relevant to the problem of optimization.
There will be other ways to get high quality HD source for the
handful of homes that can afford a dedicated theater with a screen
larger than 100" diagonal. We are talking about distribution of media
to the masses, and optimization of the source signals to assure high
quality for the displays that will be used by the masses. This is
equally true for DTT, DBS. Cable, IPTV, HD-DVD, and future downloads
of content via the Internet.
You cannot prove that there will be any perceptible benefit to
encoding of 1080P for emission, on sets smaller than 100 inch
diagonal; on the other hand, you can prove that the channel will be
stressed more for 1080P than 720P, IF the source content has levels
of detail that are significantly higher than the 720P source.
But even this is irrelevant, as you will not find any RPTV or panel
displays even this large - the current max seems to be about 72" for
RPTV, and these sets tend to look washed out.
So the only situation that is relevant is for the small number of
homes that will create the proper environment for a large (>100")
front projection display. That's a screen that is nearly 8 feet wide
by more than 4 feet high.
>"Optimization" means maximizing (or minimizing)
>an objective subject to a list of constraints.
>Aside from your #1 point, the other two points
>aim to reduce quality (i.e. the objective)
>possibly below that allowed by the constraint set.
>There's no valid excuse for this.
NO BERT. They DO NOT aim to reduce quality. They aim to match the
level of quality delivered with the upper end of products that will
be purchased by consumers to view digital television. Note I did not
say the high end. The upper end of the product curve will be HDTV
displays in the 50-70 inch range. Beyond this, we are talking about a
VERY tiny niche market, and even here, upconversion from 720P will
produce VERY HIGH QUALITY images on those big displays.
>
>Since the highest quality of HD has been defined
>to be 1080 lines, that is where HDTV will
>migrate over time. The Europeans are exactly
>correct to question any pre-conceived one size
>fits all 720p notion, especially if they are
>not constained to MPEG-2 for HD.
This is not even true today. Hollywood is still holding out for 4K x
2K. And the Japanes are already developing Ultra High Definition
formats. In SOME peoples minds, HDTV has been defined as 1080 lines.
Back in the '30s the same type of people were calling 480 lines HDTV.
In other words, this is a moving target.
But there is a problem with moving targets. They only need to move so
fast, and so far. We could build all motor vehicles to cruse at 150
miles per hour, but you can't drive that fast. The history of the CE
industry suggests that the average product will improve in quality
over time; it also suggests that the industry is very pragmatic about
performance, and will not burden its products with unnecessary
overhead, especially as they try to drive new technologies into the
mainstream of the marketplace.
The truth is that many CE vendors took the easy way out, with 1080i
displays, which are technically less challenging to build than 720P
displays. These displays will NEVER work as well as a 720P display,
because they are ALWAYS interlaced, regardless of the source format.
It will be MUCH MORE DIFFICULT to justify 1080P displays over 720P
displays, because the cost differential is much larger than the delta
is performance, and most people will NOT be able to see the
difference on the screens that will be purchased by the masses.
We are still in the political war phase over HD formats. But
progressive formats are winning the day as more and more people
understand the advantages. It should tell you something that all but
one of the networks doing live action sports have chosen 720P (NBC is
still not in the game, other than the Olympics, for which they took
the 1080i video that was being shot for the International
Broadcasters feed).
I do not expect this format war to end tomorrow. There are powerful
forces at work that are propping up an inferior vision of HDTV. A
vision that is seriously flawed. These flaws will not stand the test
of time. But 1080P will find a home for applications where it makes
sense:
1. Image acquisition - for oversampling purposes
2. Digital Cinema display on very large screens
3. Large screen Digital Signage
Just out of curiosity Bert, can you provide any examples of where the
consumer electronics industry has optimized products around the
highest level of performance for a mass market product? Please do not
cite the litany of improved audio formats we have seen over the
years. Each time the industry raised the bar on audio quality there
were higher quality formats available. For example, CD audio is
marginal in terms of quality, but it was much better than existing
analog formats. The mass market is generally satisfied with CD-audio.
But there is a tiny niche market of audiophiles who want something
better; for them the CE industry created DVD-A.
For some reason, I do not see the masses rushing out to upgrade to
DVD-A players...
It's time for a reality check Bert. Look in the mirror - you are an
above average consumer. One who cannot even justify an investment in
HDTV at this time. And please don't try to tell me that your new LCD
panel is HD capable. There's more to the HD viewing experience than
numerology.
Regards
Craig
----------------------------------------------------------------------
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: