[opendtv] Re: 1080p @ 60 is Next?
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Tue, 22 May 2007 09:23:10 -0400
At 10:40 AM -0700 5/21/07, dan.grimes@xxxxxxxx wrote:
By going to 1080p@60, one gets the better spatial qualities of
1920x1080 plus the temporal resolution as well. (And I realize that
there are more qualities than just spatial and temporal.)
Perhaps. Remember, you are giving up sensitivity when you increase
the resolution AND the frame rate, and the higher you go the greater
the impact of noise, while at the same time you lose contrast due to
MTF considerations.
It is interesting to note that many of the newer camera designs are
operating internally at 1080@60p, although there are MANY different
internal processing resolutions. I guess my point here is that the
most important consideration is an all progressive chain. After that
it's all about the perceived quality of the pictures.
Dan [5/21]: Naturally, and improving the distribution chain would
be nice, too. The truth is, the bits can be delivered (at least in
many cases), but for many reasons are not. There is definitely
going to be more debate over the need for better distribution now
that 1080p displays are becoming the standard.
Where it makes sense we will see improved distribution chains. This
is already happening with digital cinema, and you might be able to
deliver higher quality to some of the venues on campus. But further
improvement to the DTV distribution infrastructure are a LOOOOOOONG
way off.
Just because we have 1080p displays does not mean we need 1080@60p to
deliver the best quality images to them. As I have already covered,
we can deliver 1080@24P now. And we can deliver BETTER quality 60P to
these displays using 720P for emission.
It will be a long time before the production infrastructure of
1080@60 exists, especially for live origination. We are just now
getting a meaningful number of trucks that can do the existing HDTV
formats. And TV stations are just beginning to do HDTV production.
The economics alone suggest that 1080@60P is not going to be used in
these applications, and the science tells us that we don't need it
anyway for a mass media distribution infrastructure.
I would like to go further into the State University aspect. First,
our project is based on a donor who wants the highest quality. So I
must take that into consideration. Second, funding for a university
project/entity is unique in that we don't recover our investment in
dollars. We must recover it in education, research and benefit to
society. So our recovery models are based on getting the right
tools we need to do a unique job. For instance, we have the
potential to get research dollars on comparing audience reactions to
ultra-high resolution productions. It is hard to get the grant to
do the research if you can't produce or display ultra-high
resolution. Third, we only get infusions of money once every 10
years or so. So we need to invest in the most reliable and future
proof technology. Otherwise, five years down the road we will look
very much out of date. Would it be better to invest smaller amounts
over time? Absolutely, but it just doesn't happen that way.
Interesting trade-offs. I suspect that you can have your cake and eat
it too. By this I mean that you can put some of the money into the
very high res stuff. But I would suggest skipping 1080@60P cameras
and move into the higher realm of Digital Cinema - for example the
RED camera. You can still process this stuff at 1080@60P, in fact
this would be a good thing to do.
But you also need to provide the students with REAL WORLD training,
and you're not going to be able to put cheap 1080@60P camcorders on
their shoulders.
So set your sights high, but be both flexible and practical.
Dan [5/19]: In theory, I would rather buy 720 equipment that
oversamples versus 1080 equipment that undersamples. I keep telling
myself this and then I visit NAB. I am always disappointed with the
results of 720 equipment when compared to 1080 equipment.
Especially when recording formats are considered. I know that the
720 equipment will be fine for programs delivered via OTA, cable and
the internet. But our customers are more than a public audience.
They are also researchers wanting higher resolution (need
uncompressed at times) and higher frame rates.
I would take another look at some of the 720P camcorders out there. I
think they are improving faster than the 1080 products. Panasonic
introduced a bunch of new products this year, and some of them are
very competitive at the high end. More important, Panasonic is moving
to a more appropriate compression technology, which will improve
delivered quality and keep the storage requirements at reasonable
levels.
Dan [5/21]: We might be unique in the fact that almost none of our
productions are live. And we use a lot of green screen. I wish our
production team would understand that the artifacts they see in our
keys are because of the lower color resolution inherent in all tape
formats.
You could do the chroma keying live and record the results. There are
relatively easy ways to improve the bandwidth from the camera to the
compositing system. But once you commit the source to ANY of the
compressed tape formats you have thrown away a great deal of useful
information.
Dan [5/21]: Yes, I want to build a state-of-the art facility. I am
trying not to let my pride be the guide here, desiring to design a
top notch production facility. I am trying to make my decisions
based on realities and practicalities. Because my applications and
customers are unique, perhaps I am quite off when it comes to
broadcast television. But most of our work is in the broadcast
realm, so I hate to over extend our capabilities where we needn't,
paying more for a production facility than we need to. I am trying
to go to the highest level practical rather than to the lowest.
What ever we do end up picking, we will be stuck with it for a very
long time. This simply must not be in a field that is changing so
fast. But I don't have control of it...here.
Don't let the big numbers have too much influence. Focus on the
ability to maintain high quality levels throughout the chain.
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:
