Re: Will broadcast TV ever go away?



On Jun 27, 9:26 pm, phil-news-nos...@xxxxxxxx wrote:
On Thu, 26 Jun 2008 10:48:22 -0700 (PDT) dmaster <dan....@xxxxxxxxxxxxx> wrote:

| Interesting, Phil.  While you indeed may be able to do so, I read an
| article about a double blind test done with experienced HDTV
| reviewers.  Using 1080p and 720p set pairs from major manufacturers,
| the reviewers were unable to reliably identify the 1080p display, even
| though all the content was Blu-Ray 1080p.  I'm afraid I don't have the
| link with me anymore, or I would supply it.

I've never heard of a "720p" set.  But I can't rule out that maybe some
company actually makes one.

Ah, you're being a bit pendantic. Yes, I realize that most sets sold
as 720p are actually something like 1366x768. Some of the early 720p
TVs were actually 1280x720, if I recall correctly. In any case, TVs
are sold as "480p" (EDTV), "720p" (any HDTV with 720 or more vertical
pixels, but less than 1080 vertical dots), "1080i" (CRT HDTVs), and
"1080p" (1080 vertical pixels.). So, the test in question compared
what were likely 1080p sets and 768p sets from the same manufacturers.


I've personally seen sets running some kind of high quality video, quite
possibly a Blu-Ray disc.  I was able to quickly tell the difference
between a 1366x768 set and a 1920x1080 set.

I can also see the better resolution between my 19" 1440x900 native set
and my brother's 32" 1366x768 set.

With the size difference (even assuming the same maker), you'd have to
nomalize the viewing distance to get a valid test. I'd be truly
surpized that anyone could actually see the difference between
1440x900 and 1366x768. Either the distances aren't normalized, or the
color/contrast/etc is different, or its an issue of expectations (the
danger of a non-double blind test), but it is surely not the tiny
difference in pixels.


| Another study I saw from last year showed (at the time) that no 1080p
| set tested could actually produce 1080p output on a still picture
| (thought the good plasma sets came close).  All displayed a real
| resolution well below 1080p with motion video, with 3rd tier LCD sets
| often well below 720p.

There is no such thing as a 1080p output of a still picture, at least
with an LCD.  The picture will be static.  There should be no motion.
I've used the "freeze" button on a couple sets and it freeze very static.

You misunderstand. A 1080p set can display a moving picture or a
still picture. The reality of today's TVs is that the pixel
capability of the display panel is usually not the limiting factor in
resolution, but rather the TV processing. The study showed that 1080p
sets did *not* actually show 1080 vertical pixels of information, even
with a still image. For moving images (which of course requires far
more processing), the short fall was mild (something in the 900s for
good plasma sets) to extreem (something like the 600s for 3rd tier LCD
sets).


| Third, remember that the human eye's limited resolution requires
| either an unusually close viewing distance or an unusually large
| screen before the human can have any hope of discerning the additional
| resolution.

And the point where people can see the difference varies from person
to person.

Admittedly.


| Finally, consider the lessons from digital photography.  With good
| quality components, it often requires at least a doubling of
| resolution *in both dimensions* before an average observer clearly
| sees a difference.  A real 1920x1080 vs 1378(?)x768 is (other than for
| marketing reasons) a distinction that is probably too small to notice
| except under the most ideal conditions.

But I do try to create ideal viewing conditions.  Under adverse viewing
conditions, we probably don't even need HD.

Very true. For many people, there conditions prevent them from seeing
any difference between 480p and HDTV. And only a very few are likely
to be set up to see the difference between 720p (768p as it were) and
1080p.


| As others have stated, things like black level, artifact-free motion,
| real contrast ratios, actual motion resolution, and the like are
| probably far more important to perceived picture quality than the
| small step from 720p to 1080p.  (In my opinion.)

They are all factors, but they vary from person to person.  I actually
like a lower contrast, myself.  I like less saturated color, closer to
what is real life.  With older color CRTs (not the recent ones made in

Same here. I prefer a natural looking picture. My neighbor seems to
love his plasma set on "stun". To each his own, but his looks like a
cartoon to me.

the 1990s and 2000s) I would adjust them for very good black level,
linear contrast, and life-like color.  Viewed in a dark room they were
great.  While the Tekronix in the TV studio would still beat it easily,
it was at least closer to that target than almost all home TVs of that
day.

| I'm not saying not to buy a 1080p set.  I'm not saying that you might
| not see the difference under some ideal conditions.  But I'm
| suggesting going into the buying process with some realistic
| expectations (along with a knowledge of the dangers of non-double
| blind comparisons).  In the end, buy what looks good to *you*, not a
| number.

I suggest people view the TVs under a variety of real viewing conditions.
These include home-like lighting conditions, as well as real video sources
(e.g. DVD, Blu-Ray, Cable, Satellite, AND Over-The-Air, with all formats
of sources) and see what display lets best.  Unfortunately, you cannot get
this exercise in a store showroom.

Too true.


--
|WARNING: Due to extreme spam, googlegroups.com is blocked.  Due to ignorance |
|         by the abuse department, bellsouth.net is blocked.  If you post to  |
|         Usenet from these places, find another Usenet provider ASAP.        |
| Phil Howard KA9WGN (email for humans: first name in lower case at ipal.net) |

.


Loading