[opendtv] Re: 3-D TVs for industry: $64B; health research: zero
- From: Mark Schubin <tvmark@xxxxxxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Thu, 04 Mar 2010 12:12:35 -0500
On 3/4/2010 11:19 AM, Stessen, Jeroen wrote:
I find the convergence-accomodation conflict the weakest argument
for explaining issues with large format 3D TVs, it goes to the bottom
of my
to-do list.
As we don't yet really have 3D TVs, we don't yet know what the issues
will be.
There will be more interesting issues (artefacts) to solve first.
In some 3D theaters some horrible mistakes are being made, because people
have not yet been sufficiently trained. That will improve with
experience.
Until then, we should not speak of problems as if they are all
fundamental.
In your previous message you spoke in glowing terms of your local Pathé
cinema:
My friend the operator at the Pathé movie theater (Mark knows who)
says that they receive virtually no complaints, but he has taken great
efforts to make sure that everything is right. Some guys at the lab here
have seen Avatar 3D in Pathé on the opening night, and the quality
was indeed excellent. Also, the brightness was 2.5 times higher than
what it should have been for 3D, I suppose that that helps too.
So that seems like 3D at its finest: Superb source material (/Avatar/)
and superb projection. Yet you seem to report that SOME viewers still
have problems; "virtually no complaints" is different from no
complaints. So, even under optimum conditions at long viewing
distances, at least one viewer in that cinema has had a problem.
Ø And, perhaps Ethan Schur, who spoke at the HPA Tech Retreat about 3D
Ø on mobile devices will disagree with Jeroen's "Duh !" about that
application
Ø of technology.
We were talking of 3D TVs, which in my business are several meters away.
His problems are not our problems. There exist better solutions for
mobile
displays, like dual near-eye displays, i.e. virtual large displays at
infinity.
So, duh, don't put a problem on my plate that is not *my* problem.
Maybe it's just a very unwise idea to watch 3D on mobile to begin with...
Which, unless I am very much mistaken, is precisely the difference of
opinion between you and Schur that I suggested above. He thinks it is
NOT an unwise idea.
Ø But the idea that because some viewers watch TV at a distance of at
least
Ø three meters from the screen there is no need to consider visual
factors in 3DTV is bogus.
Ø Here's a promotional image from a manufacturer.
Duh again, that is a simulated image, and it is about gaming. It is
not at all
representative of how people watch TV. The typical situation is a 32"-47"
screen at 8 to 12 feet distance, and with very few in-your-face 3D
effects.
I'm pretty sure that when Avatar 3D comes out on Blu-ray, we'll all
praise
Sir James for having done another perfect job, and what was the problem
again ? Yes, some people (myself included) will probably need to see an
eye doctor first, but they should have done that sooner or later anyway.
It's that range of "some people" that we need to be concerned about. As
I pointed out, I can fuse almost anything into 3D, so I'm not likely to
have problems with 3D TV. But In-Three, who are in business exclusively
for 3D, posit a home viewing distance of six feet, not three meters, and
there will be some who sit closer than that.
If a stereographer sets something in a scene at infinity, that's a
certain distance apart; let's call it 65 cm. On any screen that's large
enough, the distance can be 65 cm. But video signals don't have
distances. If something is set to be 65 cm apart on a 32-inch screen,
it will be 130 cm apart on a 64-inch screen and 33 cm apart on a 16-inch
screen. At the HPA Tech Retreat, as you know, three of the prizes were
seven-inch TVs, and, in the Panasonic truck, there was a 103-inch
plasma. Perhaps Philips will come out with a 3DTV that automatically
adjusts disparity for screen size. I look forward to it.
However... the intermediate case is probably more interesting and
relevant:
watching 3D content, particularly games, on a typical (24") computer
screen
at arm's length distance. Then there will be accomodation-vergence
issues.
There will also be frequent window violation errors, which are even
worse.
Luckily such content is generated in real-time, so it is easy enough
to adjust
the depth to your individual liking. nVidia has put a controller dial
on its IR
emitter (for the shutter glasses) to do precisely that. AFAIK you can
dial it
down to almost 2D, if you wish. It is much harder to do that for a TV,
you
would need a separate depth signal and supplemental occlusion data. This
has been proposed in research circles, but that don't make it any
feasible yet.
Indeed.
So are we in agreement that there are physiological issues related to 3D
TV that need work?
TTFN,
Mark
Other related posts:
