[opendtv] Re: Public M/H information
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Fri, 19 Dec 2008 07:24:08 -0500
At 11:19 PM -0500 12/18/08, Tom Barry wrote:
Interesting. I'm not sure 240p will be enough even for mobile. It
has to compete with what I can get already as part of my Sprint PCS
Vision plan and is I think already less than the rez of an iphone.
Supposedly at a viewing distance of 3 screen heights the sweet spot
for human vision is about 1080p.
This is only true for a screen that is VERY LARGE. In tests of
preferred viewing distance the RAI found that most human observers
would not choose to sit 3 picture heights from a screen until that
screen was about 117" diagonal.
I guess I view my 2" high phone screen (sideways) at maybe 8 screen
heights viewing distance so it would peak out at about a resolution
of 1080p * 3/8, or about 405 lines even for that very small screen.
I'm guessing even smart phones will soon have an average resolution
of above 240p.
This is the correct form of analysis. Human visual acuity is
typically measured in cycles per degree of viewing field. 22 CPD is
considered to be a sharp image. The spread of human visual acuity
across the population puts the upper end for normal vision in the
range of 22 to 30 CPD. In a few pathological cases, a stimulus as
small as 40CPD may be observed. A classic example is a red blinking
light against a black night sky in the distance when you are driving;
the temporal component is probably important in the detection of such
a small stimulus. Compare this to the same blinking light against the
normal blue daylight sky.
So what is critical is the delivered CPD at the desired or typical
viewing distance. This is the main reason that 720P is more than
adequate for most DTV displays - i.e. people typically sit 9-10 feet
from the display so it takes a very large display to see the extra
detail in a 1080P display.
I would also note that we are now moving into the era of display
oversampling. This not only hides the raster, but it also makes
content that is NOT Nyquist filtered look less aliased. Thus we see
many computer displays moving to 110 DPI or greater and small
displays such as smartphones with even higher pixel densities. For
example, my iPhone has a 3" X 2" display that is 480 x 360, or 160
DPI.
That being said, video is Nyquist filtered and typically not a sharp
as other sources. A 240P video clip looks fine on my 360P display, IF
it is not falling apart with compression artifacts. YouTube is 240P
but their default player scales to 480 x 360. Hard to argue that
people will watch THIS.
But maybe a free advertising supported mobile simulcast of the
network HD broadcast could still be viable. ??
Why a simulcast. Yes for sports, but hardly anything else. Virtually
all prime time programming can be viewed on demand or downloaded.
Prime time for mobile applications is between 7AM and 6PM.
The time and channel business model of broadcasting is already dying.
Extending this to the new mobile service is not going to change that,
but it could cause the new service to be still born. If this service
is to be successful it is going to take some INNOVATION...
AND MUCH MORE attention to locally produced content and the needs of
local advertisers.
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:
