3D, health issues, missing data, and PR

When I started working with the USC Entertainment Technology Center (ETC) in January of this year, I thought the primary problems that stereoscopic 3D would have to overcome in order to become a mass market product were technical.  There were multiple formats for encoding the signal for distribution from a source to a consumer device.  There were competing display technologies in the marketplace, including active shutterpolarizedhead mounted displays, and autostereoscopic (plus the far inferior but cheap red/green anaglyph).

Nine months into this activity, I think those problems are going to solve themselves through market forces and individual company’s desire to have their products win in the marketplace.

The competing encoding formats will either shake out into one or two that organically come to dominate the market, or software that auto-detects the incoming signal and appropriately processes the image data will take over.  That translation tool will feed a signal that is appropriate to the display device.  Consumers will choose shutter glass versus polarized display based on personal preference, just as they decide between DLP, LCD, and Plasma screens today.  Each will have pluses, and minuses, marketing spin, and personal buzz to help inform the consumer’s decision.  But no matter what display technology the consumer chooses, all content will most likely play on all devices once those translators are deployed.

I now think the greatest possible problem facing stereoscopic 3D as it moves into the mass market is the lack of a rational, data-driven response to grass-roots emotional fears about and possible regulatory action against 3D content.  There is no data on the effect of viewing 3D content on a diverse, mass audience.

Vision scientists and practicing eye specialists have been gathering data from small target populations for decades.  The studies cover the ability or inability to see in stereotechniques and exercises for learning to see in stereo, back-of-the-eye / focal length flexibility, the eye-brain relationship, vertical alignment or misalignment of individuals’ eyes, and other topics.

What is missing are statistically valid studies that are directly applicable to the audience for 3D movies and content.  That audience ranges in age from 6 months to 100 years, and includes men and women, people with 20/20 vision, nearsightedness, and farsightedness, people who are stereo blind, people with excellent eye alignment, and people with vertical and horizontal eye misalignments.

We have plentiful data on the demographics of the movie-going audience.  But we lack data on their visual acuity.

The entire industry is unnecessarily exposed to a battle of opinions that it WILL lose without data.  Bloggers are already questioning whether viewing 3D content could be harmful to children’s development.  Some researchers I have spoken to have said that it could actually be good for the development of vision because it exercises the eyes.

At a recent major European media conference, a technical advisor to the European Broadcasting Union called for warning labels on 3D viewing experiences stating that they could be harmful to the viewer’s health. In a follow-up email exchange with him I pointed out that reading while sitting next to a window in a moving car can also be harmful to your health, but no one is calling for warning labels on car windows.  From a policy-making standpoint, the questions he should ask before making policy recommendations should include the severity of the risk and the size and scope of the risk among the general population.

But today there is no general population data.  Any policy discussion would be based on individual case studies, small non-representative studies, and anecdotes.

At the ETC I have proposed developing studies with defensible methodologies on the health impact of both short term (2-4 hours) and long term (weeks and months) 3D content viewing.  We have also proposed population studies on the distribution of people who experience eye strain while viewing 3D content.  These studies would be done on the wide range of people who would typically attend movies; both 2D and 3D.

As with everything the ETC does, the data would be gathered in a defensible, neutral manner, and presenting in its raw form as well as with some statistical analysis done to it.  But without judgment or marketing spin.  The ETC strenuously holds to its charter as a neutral facility, and not an advocacy body for any industry or group.

The  objective would be to be prepared with data when point/counterpoint opinion pieces and debates erupt related to the impact of viewing 3D content on the health of the audience.

I look forward to working with the members of the ETC and other interested parties to develop these plans and fund the studies.  It will be good to advance the consumer 3D experience and enter into debates with a foundation of data.

Consumer 3D Experience Lab

Here is the description of the Lab and program I am building at USC.

The Consumer 3D Experience Lab at the Entertainment Technology Center is a neutral forum for learning about and discussing the latest resources available for delivering a 3D viewing experience to consumers in the home, on personal devices, and in public spaces.  Three dimensional viewing technologies have come a long way since the blue and red paper glasses used to view 3D in the 1950’s.  The recent increase in theatrical 3D releases has proven to be extremely popular among consumers, and so it only makes sense for the creative industries to team with product and service providers to develop ways to move that experience into other environments, enabling the consumer to enjoy 3D experiences anytime and anywhere.

The Consumer 3D Experience Lab showcases the wide range of 3D technical approaches and viewing experiences currently available to consumers.  The Lab is a platform for bringing experts together to discuss the creative and technical issues that will define an enjoyable, long-duration 3D viewing experience.  The goal of the Lab is to accelerate the convergence of ideas, so that consumers looking for 3D content, products, and services will have a consistent, enjoyable purchasing and viewing experience.

The facility consists of a 3D Living Room for focus groups and private meetings, a 3D Home Theatre containing a consumer-grade 3D projection system, and a 3D Sandbox where all the major consumer 3D content, encoding, and display options are demonstrated.

Twitter has two states – ambient and active

After watching twitter for a few months, I’ve come to the conclusion that it has two states. Twitter was built as a short message many-to-many social network. Email does one-to-many easily just by creating lists. SMS already handled short messages. Twitter’s value is its ability to easily form communities and have them interact quickly in a web of short bursts.

What twitter has become is a two state environment.

When nothing of interest to a large number of people is happening, it is in an ambient state. While twitter is ‘idling’, a small percentage of twitter accounts are doing two things; telling their followers what they are doing at the moment, or posting links to something interesting that they have just found on the web. The first group loses me immediately; but then I’m not the gossip girl demographic. The second, when they limit themselves to really useful information (yes, that is a subjective call), are performing a public service. However, too many people in this second group suffer from what a friend of mine calls the ‘ooohhh shiny’ syndrome, which I’ll now rebrand as linkurea – unusually frequent and excessive flow of links. They tweet 30-plus times a day about anything and everything they find interesting. I think it’s nice that they care to share. But the linkurea crowd needs to find some way of filtering themselves, otherwise the useful and useless combine to become background noise. Also, I made the mistake of having tweets go to my phone when I originally set up my twitter account, and was hit the next month with a $200 ‘twitter bonus’ on my bill.

A subset of these ambient twitterers are the entertainment twitterers. They include Ashton Kutcher updating you on his thoughts every 10 minutes, and people who tweet an aphorism or joke once or twice a day. This is an excellent use of twitter as a guerilla-marketing technique – they build and win over an audience through passive engagement, then can mobilize them as part of a marketing campaign or social initiative.

The second state is what makes twitter such a powerful resource as well as an interesting social phenomenon. When something of interest is happening in the world, twitter kicks into ‘active’ mode and is excellent resource for forming instant communities and spreading information. We saw this in Tehran after the election. We saw this after Michael Jackson died. We saw it during the recent fires in LA. When twitter is in active mode it informs large groups quickly, can suppress misinformation before it goes viral (unintended consequence – it can also rapidly magnify misinformation, making corrections extremely difficult), and it can go global fast.

Twitterfall is my favorite tool for watching Twitter in its active state. Twitterfall displays a cascade of tweets from all over the world that contain my search term. When there is a news event that I care about, I can input the search term and see what a broad cross-section of people are tweeting about it. It also has a “trends” list of the most common terms being tweeted at that moment.

A recent Harvard B-school study, which studied over 300,000 twitter users, found that “among Twitter users, the median number of lifetime tweets per user is one.” The top 10% of twitterers account for 90% of the tweets; a much higher percentage, they found, than on other social networks.

An additional interpretation of their data is that when nothing much is happening, twitter is not yet integrated into the fabric of society to the point where it is viewed as a useful tool, the way email and instant messaging are. But when something is happening that a community cares about, people seek out Twitter, and it comes to life.

Gameworld: Hollywood sends 3D home — in videogames

Disney’s G-Force for the X-Box and Playstation 3, and Disney/Pixar’s Toy Story Midway Mania for the Wii will both be released in 3D that uses red/green anaglyph glasses.  While the studios are releasing 3D games and the lesser 3D movie titles in anaglyph (Hannah Montana, Journey to the Center of the Earth), they have not released A-list animated films in 3D for the consumer market.  They are working with CE and media distribution companies to encourage the adoption of a better quality 3D stereoscopic viewing experience at a reasonable price point.   

 Ubisoft has developed a proprietary stereoscopic 3D technology for game platforms that “James Cameron’s Avatar” game will incorporate when it is released this Holiday season.   It reportedly relies on a polarizing display and polarized glasses.  Unlike anaglyph glasses, which strip out some of the red and green (or blue) spectrum in order to create the 3D effect, polarized glasses retain most of the original color.  It is a much higher quality viewing experience.  

James Cameron is quoted in this Reuter’s article as saying ”You just stick your head into the monitor and the world wraps around you. It’s the first time in a videogame that I was afraid…when the hammerhead enemies attack. It’s very frightening.”    

Gamers have an established record of buying peripherals that enhance the gaming experience and/or make new gaming experiences possible.  Both polarized glass and shutter-glass displays provide a far superior 3D viewing experience to anaglyph.  Both are available to consumers today (Japan is currently the main market).  As long as the consumer’s STB/console processor can handle the encoding format for the left-eye/right-eye data, the same content can be displayed on either the polarized or the shutter-glass display.

As I and many others have said before, it is quite possible that 3D movies and 3D televison programs will move into consumer’s homes and onto their devices by following a path blazed by gamers.

Albert Brooks “Real Life” 3D movie trailer

Albert Brooks created a trailer for his 1979 movie Real Life that is a parody of a 3D movie.  It is sort of blue/red anaglyph, but it is just as good without the glasses.

Mitsubishi, 3D TVs, and tracking convergence

Mitsubishi announced last week that it would field a total of eight 3D-ready HDTVs – including what it termed the largest mass-produced 3D television, the 82-inch DLP Model WD-82737.  Much of the press coverage did not mention which of the five non-interoperable 3D technologies, some of which have non-interoperable subset technologies, the sets would use.  The 82-inch uses shutter glass 3D technology.

Is anyone aware of a research group that is tracking product launch and sales by 3D technology, so we can see if the industry is naturally converging on a consumer playback standard outside of all of the discussion groups?

(The five non-interoperable 3D technologies are: Spectral (including anaglyph, colorcode, and a few other two-color glasses solutions), Polarized (orthogonal and circular), shutter glasses, autostereoscopic (many different view-counts), and head mounted displays.)

3D TV Panel for The Caucus


Last Friday night I moderated a panel on 3D TV for The Caucus, an organization of senior level television producers, directors, and writers.

Ray Hannisian, the stereographer on U2 3D, explained the emerging role of the stereographer as someone who manages the dimensionality of the image, and presents it in a way that blends with the overall intent of the writer and director. Ideally, the stereographer is involved in preproduction discussions, on-location activity, and in post. Ray has been working in construction for over a decade, developing his stereography skills on the side, in preparation for this day when 3D appears to be emerging as a key tool on Hollywood’s creative pallet.

Robert Duncan McNeill is the co-executive producer of Chuck. He directed the 3D episode of Chuck that was broadcast the day after the 2009 Superbowl. He also directed a 3D episode of Medium five years ago. He noted that five years ago the 3D camera rig was kluged together and had to often be mechanically realigned. The 3ality 3D camera rig that he used for Chuck had internal checks and adjustments to automatically maintain alignment. Any problem that was missed during capture could be fixed using software in post. 3ality has also developed small 3D handheld rig for shots in tight places.

Ted Kenney has produced a number of live 3D sports events, including the BCS final and some regular season NFL games. He noted that 3D shots at or near field level put the audience right in the action. He uses eight 3D rigs in various locations and holds shots longer than a normal twenty rig 2D sports shoot. Software in the truck allows for smooth live transitions among the 3D cameras.

Sandy Climan, CEO of 3ality Digital, LLC, described how 3ality is both an equipment and a creative talent company. Working with CEO of 3ality Digital Systems and long-time 3D technology expert Steve Schklair, 3ality is developing camera equipment with the features and software tools necessary for any situation.

Eisuke Tsuyuzaki, VP of Corporate Development and GM, Panasonic Research, projected that, out of the 5 distinct and non-interoperable 3D technologies in the consumer market today, polarized lens and active-shutter are the two most likely to succeed in the marketplace. They are the two that are most acceptable to all three key stakeholders; the content community, display manufacturers, and the public.

The evening panel was recorded for Emerson College, Boston. I have asked The Caucus for a copy of the video.

Woot and Scarcity

To my great surprise, I ran into a Woot both at the E3 Convention yesterday.  Woot is the site that sells a supply of one or more items per day, at a fixed low low price, until their stock of that item is gone.   They have succeeded in forming a community, or rather multiple niche communities on the site that provide information and advice about the day’s product.Woot was at E3 exploring the idea of selling digital content through their site.  Jay Johnson, Director of Deals.Woot, said that they had not yet framed up what they wanted to do – physical media versus digital download, long form content versus short form content versus snippets versus image files versus….Woot’s entire ethos is built on the concept of scarcity.  Once their day’s supply of the product is gone, the store is closed although that chat keeps going until the next day’s items are posted.The internet, as a generation of advocacy bloggers will tell you, is based on unlimited supply.  They claim that the Tragedy of the Commons does not exist in the digital world because the original ‘good’ is not impacted by unlimited copying.  (I disagree on their interpretation of what that ‘good’ being enabled in the digital space is, but that is the subject for another blog.)It makes no sense for Woot to get into the digital download business because there is no reason for the sale to end.  The supply is never depleted.  Without depletion, Woot becomes just another store competing with iTunes, Amazon, and the rest.One of the items Woot is offering today is a USB flash drive with a UbiSoft game loaded on it.  Woot should stay in the business of selling digital content on limited edition physical media.  Collectibles, close-out items, warehouse sales, and the like are a solid niche market that will not go away anytime soon.  Even when they can get the content for free, there is always someone willing to pay for the physical media with the picture on it.

File Sharing, Market Search, and Open Culture

Vuze, a leading HD video P2P file sharing network, yesterday released the highlights of a market research report.  The report compared the media and technology consumption/usage habits of file sharers (specifically Vuze users) versus those of the general internet population.  Vuze commissioned Frank N. Magid Associates to do the research.Digital Media Wire reported the story with the headline “Report: P2P Users Are Hollywood’s ‘Best Customers’.” In all the years of market research and claims from both sides of this file sharing debate, when I have been able to see the raw data itself I have not seen a study that did not have built-in biases. Questions have been worded in a leading fashion. Multiple choice questions did not include the option of an obvious answer that would provide data counter to the bias of the study’s sponsor. The group surveyed for the study had a built-in bias. Contrarian data was buried in ‘none of the above’ and ‘other.’In the material that they have made available, Vuze has listed all of the questions, but not all of the possible responses. They have posted selected results, but not all of the findings.Vuze may have conducted a valuable study, or they may have produced more advocacy results. It would be incredibly useful for Vuze to post the raw data, so that those of us who care about these issues can draw our own conclusions about both the methodology and the findings.

3D test/demo reel project – your input please!

I am working with a group of Hollywood entertainment technologists and stereographers to develop a test reel of 3D content. The ultimate goal is to produce a ‘Hollywood quality’ reel that can be freely distributed (no clearance issues).

I would appreciate anyone’s input on these 3 questions;

* What should the purpose(s) of the reel be? (What need(s) should it meet?)

* What 3D characteristics related to capture, processing, and display should some of the clips in the reel be designed to demonstrate or test?

* What keywords should be used to characterize the clips (asset management ‘data dictionary’ terms)?

To get you thinking, below are the notes from our first meeting.


What should the purpose(s) of the reel be? (What need(s) should it meet?)

* demonstrate the impact of various compression and display technologies on the 3D image quality

* showcase examples of “good” 3D technique

* provide material and a situation for testing different 3D eyewear performance characteristics (ex. Reflections from rear light sources)

* test the idea that ‘in front of screen’ works best for big theatrical screens and ‘behind the screen’ works best for small home screens

* provide a self-testing tool for consumers; ‘if you find watching this test reel uncomfortable, than you are among the small percentage of the population known to have difficulties viewing 3D and should avoid 3D movies, videogames, and other 3D experiences that involve viewing 3D for a long period of time.’

What 3D characteristics related to capture, processing, and display should some of the clips in the reel be designed to demonstrate or test?

* conservative versus extreme 3D stereography

* dominant scene in front of, on, or behind the screen plane

* small versus large depth budget

* transparent and solid images

* depth of field; background in focus versus background out of focus

* specular highlights (impact of sequins and reflective confetti on 3D effect)

* atmospherics (impact of smoke and fog)

* divergence

* “miniaturization” effect visible in some wide shots (humans look Lilliputian rather than distant)

* quick cuts that impact resolving the 3D image for some

* dark image vs bright image

* 2D-to-3D conversions of photographed images; roundness v cardboard effect

What keywords should be used to characterize the clips (asset management ‘data dictionary’ terms)?

Do lists of keywords related to 3D already exist?


Thanks for your help.