I've been wowed by 4K in stores, but it was probably more the HDR aspect than
anything else. Maybe the color gamut too, but for sure the HDR. Mind blowing
night scenes, in short.
It would be fun to see the relative improvement caused by each new feature
separately.
Bert
-----------------------------------
http://www.tvtechnology.com/news/0002/qa-mark-schubin-on-4k-uhd-hdr-and-more/278419
Q&A: Mark Schubin on 4K, UHD, HDR and More
Schubin provides a preview of NAB Super Session discussion
April 11, 2016
By Deborah D. McAdams
LAS VEGAS: There are few set standards for 4K, UHD, high dynamic range, high
frame rate or wide color gamut. Each has a particular path from capture to the
home through an HDMI cable into a 4KTV set from one of several manufacturers
offering a variety of technologies. So how will adoption be achieved?
That's what four top broadcast engineers take on during the NAB Show in a
Tuesday, April 19 at 2:30 p.m. Super Session entitled, "4K, UHD, HDR and
More-The Future of Video." TV Technology's Deborah D. McAdams will grill video
expert Mark Schubin, SMPTE President and CBS Vice President of Engineering and
Advanced Technology Robert Seidel, broadcast veteran Jim DeFilippis and
Ericsson's Matt Goldman with crowd-sourced questions from their peers.
Here, Mark Schubin provides a preview.
CROWD MEMBER: What provides the best bang for the buck: 2160p resolution, high
dynamic, wide color gamut, or high frame rate (120 fps)?
SCHUBIN: In my opinion, HDR and HFR offer the most bang for the buck. HDR might
be easier to implement, but it can increase motion artifacts, bringing us back
to HFR.
C.M.: The Canadians are now doing baseball, basketball and hockey in 4K, and
distributing it via cable. What is holding up the U.S. broadcasters?
SCHUBIN: They're more sensible.
C.M.: What are their immediate challenges?
SCHUBIN: Viewers.
C.M.: What about standards? What's needed?
SCHUBIN: Agreement (as usual in the case of standards).
C.M.: Beyond sports, what other genre will benefit and have a business ROI?
SCHUBIN: HFR is primarily for sports. HDR works on everything; 4K and wider
color gamut are tough sells to viewers for anything.
C.M.: Will broadcasters be "forced" into 4K/UHDTV adoption similar to the
evolution of HDTV?
SCHUBIN: I don't think so. There is a much smaller perceptual improvement from
HD to 4K than there was from SD to HD.
C.M.: What are the key dominos in the chain that need to be knocked over for
the consumers to feel this is real and start spending? Is it availability of
"great titles," consumer devices, branded premium channels, marketing or
something else not yet concocted?
SCHUBIN: 4K, again, has very limited perceptual grab. Anyone who doubts that
should visit Bob's demo at CBS with side-by-side giant screens. Wider color
gamut sells only in a side-by-side demo; consumers would have to see it in
stores that way, and store shopping is on its way out. HDR and HFR are both
perceptually significant.
Some fear HFR is perceptually significant in the wrong direction for scripted
programming. HDR might be easier to implement, but it depends a lot on the home
viewing environment. HFR needs agreement on a bit-rate reduction system
(compression).
C.M.: The young adults of today are notorious for consuming social content on
small, portable devices. They are the future money for this. Why do they need
it?
SCHUBIN: HFR is significant on any size screen at any viewing distance but
primarily for fast-moving content. HDR is going to be tough on a handheld
screen outdoors. 4K is largely irrelevant on a handheld screen, and WCG
probably won't be noticed. But even young adults are using larger screens for
longer programming.
C.M.: How do we as an industry start our advancement to UHD and HDR?
SCHUBIN: Standards.
C.M.: What is the immediate opportunity for a TV group with regard to these
technologies?
SCHUBIN: PR.
C.M.: Are we now headed toward 1080p60 with HDR and wide color as a first step,
with 4K used in production and on-set display etc.?
SCHUBIN: From your words to the powers that be! Except maybe it'll be just HDR
on regular HD, and forget the 4K.
C.M.: What are your thoughts on HDR HD facilities versus native 4K HDR?
SCHUBIN: HDR works just fine on HD.
C.M.: Is the notion of HDR for HD now officially dead?
SCHUBIN: No one told me!
C.M.: Given the file sizes, do you think we should be significantly compressing
in acquisition and post, just as we did in the early stages of HD?
SCHUBIN: For 4K, it's probably necessary. For the other improvements, no.
C.M.: HDCam, while not perfect, was a pragmatic way of getting to HD and
getting material out to audiences. What's the equivalent for UHD?
SCHUBIN: HD.
C.M.: 4K, 8K, UHD, and HDR require a very large network topology to support
file transport...
SCHUBIN: I don't understand that statement. What does "topology" have to do
with anything? And the increased data rate for HDR borders on the
insignificant; for the Philips version, it's just 35 bytes per scene (not per
frame or per second).
C.M.: ...how do we justify that investment when the delivery to the home will
not support the transport without heavy compression for years to come?
SCHUBIN: As you can tell from my other answers, I'm not a big fan of a move to
4K, but, if I were, I would point out that, for entropy coding, the sharper the
pixels and the less change between pixels, the easier the compression. So raw
4K might be 8x the data rate of HD (1080i or 720p), but the
compressed-data-rate increase would be a small fraction of that. Gary Demos
showed some compression at the HPA Tech Retreat in February with extraordinary
spatial resolution and dynamic range and not a high bitrate.
C.M.: Will 4K become a standard transmission?
SCHUBIN: Not on my account
C.M.: What are your views on affordability of transmission methods for 4K, that
is, broadcast, broadband and/or satellite?
SCHUBIN: See my comments on "heavy compression" above.
C.M.: Do you think there is benefit to using HDR and/or 4K to author a better
HD product that can be delivered to consumers without significant changes to
current infrastructure, instead of racing to provide 4K?
SCHUBIN: In the case of HDR, absolutely. Grass Valley likes to show an image of
a combined HDR/SDR sports production in which the SDR video controllers are
constantly making adjustments (as usual) and the HDR video controllers were
leaning back with their hands behind their heads. In general, the more you can
offer post, the better. As for 4K, there have been mixed opinions about
extracting an HD image from a portion of the frame. Potentially, increased
sensor resolution would improve sharpness even for HD.
C.M.: Shooting UHD side-by-side with HD is too costly. How can we reduce the
cost of producing sports in UHD and HDR?
SCHUBIN: For 4K, relatively simple downconversion is probably fine. For HDR,
many people have been working on some form of algorithmic downconversion; at
the moment, human intervention might be necessary (no worse than the job of a
video controller).
C.M.: How can we do live production of HDR sports without needing separate
"shaders" for an HDR/wide color gamut output and an SDR/normal color gamut
output for legacy TVs?
SCHUBIN: See answer above. At worst, it would be one "shader" converting the
HDR to SDR.
C.M.: Do you think that high frame rate is suitable for all genres?
SCHUBIN: Tough question! I think it's great for anything intended to look
"live" (sports, news, talk, game shows, concerts, operas, ballets, etc.). For
scripted programming, everyone alive today has been watching it largely at 24
or 25 fps, and they are accustomed to that look. Is it just that they are
accustomed to it, or is there some psychophysical reason that 24 or 25 fps
helps suspend disbelief? I don't yet know.
C.M.: Given the propensity of advertisers for brighter brights and whiter
whites, how do we avoid the CALM Act for brightness?
SCHUBIN: Those helpful algorithms, again! But there is a real issue associated
with long-term exposure to very high light levels: retinal "bleaching." If
you're watching skiing for half an hour at high brightness and change channels
to a night scene in a noir movie, it might be like driving into a tunnel at
noon on a sunny day.
C.M.: How do we integrate interstitials into programming without brightness and
color wars?
SCHUBIN: Those helpful algorithms, again!
C.M.: How well can we produce for both HD and HDR without serious compromises,
particularly in graphics or saturated colors?
SCHUBIN: Keep the graphics in the SDR subset of HDR, and don't sweat the
saturated colors. As George Joblove pointed out at the HPA Tech Retreat, we
should be dealing with gray as a reference, not black or white.
C.M.: Given the constraints of the television, cable and satellite system, how
comparable will the broadcast experience be to the ultra Blu-ray?
SCHUBIN: Could be the same; could even be better.
C.M.: Given the wide disparity of displays and capabilities on the market, how
well can we author once and use everywhere, and how much will the user
experience vary?
SCHUBIN: Never mind the equipment! If the viewer turns lights on or off or
opens or closes drapes, the experience changes. But that's always been the
case. Viewer experiences today vary greatly; that's why The New Yorker was able
to run a cartoon by Joe Dator on Sept. 8 showing a doctor telling a patient
that he'd like to see about fixing his aspect ratio. Plenty of viewers are
watching stretched people today.
C.M.: As HDR displays get better and do a nice job of presenting SDR-graded
content better than it would look on a legacy UHD display, does this challenge
the value of specific HDR grading? What percentage of average consumers will
appreciate the difference?"
SCHUBIN: I haven't yet seen HDR upconversion; I have no idea what it might look
like.
C.M.: When will lower-end 4K cameras offer servo lenses?
SCHUBIN: Define "lower end." AJA's "RovoCam" 4K camera has a built-in servo
lens and costs $2,495 including the lens; is that "lower-end" enough?
----------------------------------------------------------------------
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.
