3DTV Over-the-air Broadcasting in a Bind

Rodolfo, thanks for your report and analysis.

Your article could perhaps also have been titled, "How to turn HD back into SD".

You summarized things pretty well here...
> Therefore, overcompressing or butchering resolution to make 2D/3D signals fit in the single 6 MHz channel (option 2) seems inevitable. Since 1998 those that appreciated image quality got accustomed to endure that scenario. <

I find this very disturbing.

However, there are a couple other options which you did not enumerate. One of which would simply be to forgo 3D completely for OTA broadcasting, and focus on providing the best 2D-image quality possible. There's nothing that says that broadcasting HAS to have 3D, and considering the fact that it's at least a possibility that the whole 3D fad will lose steam and die out (like discrete quadraphonic sound originally did), that might be the most intelligent option. But given the propensity for gimmicks over quality, quantity over quality, cost over quality, and every other determinant that constantly sacrifice quality, perhaps that intelligent option is the least likely to transpire.

The other alternative would be to put 2D-to-3D converters into the sets. While the quality for the 3D effect there would not be as high as if it were discrete, it still offers a 3D option for those wanting to avail themselves of it. It would also preserve full HD resolution, not only for 2D viewing, but also for 3D, which is not an option at all with these bandwidth-limited mechanisms. That would be my personal preference, given the constraints that apply.

One thing that should be kept in mind, right behind the fact that the amount of "true" 3D content is currently very low, and growing at a snail's pace. That is whether 3D is artificially created in real-time (a process that could improve over time as GPUs and algorithms improved), or is a "true, real, dual-camera-captured 3D" is that all 3D is an effect. It's not really 3D at all, just a one-position simulation of it. There's no ability to change focus, depth of field, or position. What is being touted as 3D is simply a set of depth cues, which they brain may (or may not) interpret in a visually interesting way.

Whether the presence of this new effect is worth all the sacrifices that are being contemplated is certainly debatable.

- Tim

Another thought providing article with a disturbing
undertow for those few actually concerned
with picture quality.

Thanks Tim for enriching the analysis with two more possibilities.

I did not mention them because the pressure of implementing 3D to compete with subscriber services, pre-recorded media, and local theaters is overwhelming, and broadcasting not doing 3D is very unlikely and will paint them on a corner.

Regarding the second alternative you mentioned of 3DTVs doing the 2D to 3D conversion, the image is not that bad and many 3DTVs have the feature, but the image does not compare to original 3D content, so the pressure still there to offer a competitive product.

One of my Sons, a 38-year old that is also a system engineer and a gizmo/techie, meets me every year at CES to taste the future of technology and exchange ideas over dinner. As a Star Trek buff he sometimes issues a remark from the TV series that makes me smile for its applicability to everything in CE, and that is: “resistance is futile”.

This remark applies very much to what we are facing. 3D will happen; and legacy HD quality is at stake. As it has been since 1998, suffering the compression games content distributors play with good quality content because “people will not notice” (how butchered it is).

Look at the back pages of the ATSC paper, image quality deterioration risk and over-compression are common denominators on the various scenarios. The team is inviting ideas, so go ahead and send them your concerns as one more voice from “WE THE PEOPLE”.

Over the past couple of weeks I exchanged ideas and emails with the European Director of Standards Coordination to find out why in the world they did not choose 2D-plus-Delta for 3D if it was more HD friendly according to broadcasters.

One big reason was lack of compatibility with legacy STBs and HDTVs to extract the HD image from the 3DTV signal stream, and while broadcasters liked it because of the dual 2D and 3D quality, they were looking for someone else to foot the bill on hardware upgrades/new STBs, so the approval body chose not to implement the service compatible format on phase 1 (but may consider it for phase 2 in a few years, but there are no assurances).

We face the same problem. New tuners would be needed if switching to a more efficient MPEG-4 (50% efficient for equivalent MPEG-2 quality) for the 3D signal (and for the HDTV signal if compressed with MPEG-4), even if multicasting the 3D MPEG-4 with an MPEG-2 HD stream within the 6 MHz for backward compatibility of HD content with existing HDTVs.

Upgrading HDTV broadcasts to use MPEG-4 will make the compatibility situation worst considering that the majority of the 180 million DTVs already in households have MPEG-2 integrated tuners (that cannot understand MPEG-4), although the MPEG-4 compression standard was already approved by the ATSC several years ago, perhaps anticipating something like this.

Thanks for your contribution Tim (and Alice),

Best Regards,

Rodolfo La Maestra

3D gives me and others in my family a headache, and then makes some of us quite nauseous. Thank goodness OA and cable TV don't have enough bandwidth to support it. Maybe we'll get a few more HD channels instead.

Thanks, Rodolfo.

> the pressure of implementing 3D to compete with subscriber services, pre-recorded media, and local theaters is overwhelming, and broadcasting not doing 3D is very unlikely and will paint them on a corner. <

Sadly, I suspect you are correct. In spite of the fact that they are not already competing with those alternative sources in terms of picture quality or lossless sound quality, 3D will be seen as something they must provide, regardless of the consequences. That's because we have some very poor thinkers.

With 2D i can enjoy a bright, colorful and detailed image, with plenty of depth, and without artifacting and motion anomalies, as can everyone else in the room, with a selection of thousands of movies and programs. Alternately, with 3D I can have every one of those characteristics degraded, but enjoy the benefit of viewing 2 layers of cardboard cutouts, instead of one. Not a tough call, for me.

As far as theaters and 3D are concerned, that may well be a novelty that few will be willing to pay extra for, for long. For myself, I've only seen one 3D film in the theaters, and that was all it took. While Avatar was very well done in 3D, and an exemplar in it's category, to be honest by the time the film ended, I had almost forgotten about the 3D "effect". What I hadn't forgotten about though was the loss in brightness, color, detail, impact, etc. that I had enjoyed in the scope presentation (2.39) when I watched that version in the theater (vs. 3D at 1.78 AR).

And lest I be confused with a curmudgeon that's stuck in the past, I'll mention that I already have an excellent 61" 3DTV in my home, and a 3D-Blu-ray player, and glasses, and a few movies. So theoretically, I'm all set. Yet I have little interest in buying any more 3D blu-rays (assuming there actually were a significant # to chose from), and definitely no interest whatsoever in degraded content supplied via cable, satellite, or OTA.

> Regarding the second alternative you mentioned of 3DTVs doing the 2D to 3D conversion, the image is not that bad and many 3DTVs have the feature, but the image does not compare to original 3D content, so the pressure still there to offer a competitive product. <

The current real-time converters may well not compare in terms of picture depth, but the "original 3D content" does not compare in terms of dimensionality on the other 2-axes. To sacrifice the primary characteristics to serve a tertiary one is not a good tradeoff, in my book.

- Tim