Roy Stewart, VP of Technology Licensing and Strategic Partnerships for Technicolor oversaw the live HDR broadcast of the Los Angeles Lakers versus the Denver Nuggets basketball game from the Staples Center in Los Angeles on 31 January. He explains how Technicolor made this possible and details some of the lessons learned from the exercise.
Stewart: At a strategic level we are trying to accelerate the adoption of HDR and we felt the best way to do that was to accelerate live programming for sports in HDR. We wanted to produce an event in high dynamic range video and at the same time we wanted to be able to output standard dynamic range video.
Everybody in the industry has been talking about HDR and there have been a lot of experiments where people have gone in with HDR cameras, but nobody has done a full production the way sports broadcasts are produced today.
The objective of this exercise was to validate that we could do it with existing equipment, that we could do both SDR and HDR at the same time and that we could do it with very little capital, if any, and with very little incremental cost.
The partners involved were Spectrum Sports, the regional sports broadcaster in Los Angeles, and their production company, Indianapolis Motor Sports Productions.
We built a really good partnership with Spectrum Sports. Mark Coleman, Spectrum’s VP of engineering and operations, has quite a few technical Emmys to his credit and loves getting involved in new things that will promote his industry.
With Spectrum Sports we set out to validate that we could do a full production of a game the way it is done today, which is using an outside broadcast truck, multiple cameras, multiple inputs for graphics, score boxes, remote cameras, overhead cameras, PoV cameras etc.
In addition to validating the production we validated broadcast playout. Today a broadcaster plays out a show in master control using a playout server. That playout server is basically just managing a giant playlist. So they are pulling segments of the game program and commercials.
We also wanted to validate that we could play out a HDR show using the equipment that is used to play out content today, including upconverting the SDR commercials to HDR.
We were able to do that. For the HDR broadcast everything was in HDR including the interstitials and the commercials and for the SDR broadcast everything was in SDR.
Stewart: Any market leader has to create the whole solution in order to generate a new market. This has been documented in Silicon Valley for decades. There is a book written by a gentleman called Geoffrey Moore called Crossing the Chasm. It says the company that is first to market, and wants to build the market, has to create the entire solution.
That company doesn’t have to make all the products commercially but they have to validate that the whole solution works. So we started with the studios and we got them to upconvert content to HDR and we perfected that workflow.
Live broadcast was the next one. We felt we had to step in and take the leadership role and validate to the industry that it was possible to do live broadcast in a single production, and that you could do broadcast playout. If we did not do it, we would have to wait until somebody else did it. And we just did not want to wait that long.
It actually took very little effort and it went very well. The benefit was that it became a wake-up call to the industry and a milestone event. Now everybody out there realizes it can be done, and for very little incremental cost; whether they are a broadcaster, a sport league or a manufacturer of equipment, or of the software used in production and playout.
Stewart: They were very interested in anything that will improve the viewer’s experience and that will make their product more attractive to the viewer, especially if it costs little money to implement. That is what we validated with this production. It significantly improves the quality of the video. It significantly improves the viewer’s experience for little incremental cost to upgrade the software.
Stewart: First, Mark Coleman and his team identified all the pieces in production and in broadcast that would have to be upgraded or enhanced with software fixes or software updates to allow this to happen.
Most of the software was pretty straightforward. It just required some patches. The hardware components were the most challenging: the upconverter boxes, which we provided, and the cameras, which Sony provided.
We had a couple of planning sessions where we identified the equipment we needed or the hardware upgrade required and once that was done we put a project plan in place to make sure all the pieces came together.
We met in Indianapolis in December and made sure that everything on the production truck that needed to be upgraded had been upgraded and we tested all the inputs. Then we were ready to go ahead with the production.
Stewart: The camera operators have to know how to shape the cameras. So the camera can capture and shoot both SDR and HDR with excellent quality. They have been shooting SDR for a long time and they have developed some habits that are not necessarily good for HDR production.
There is a skill that needs to be enhanced to shoot both great SDR and HDR. And we found also that when you shoot great HDR your SDR looks better than in the original production. So there is real benefit in shooting HDR.
Stewart: There are some obvious ones and some that are not so obvious. One obvious one is that all the inputs that are not HDR, things like remote cameras have to have our technology so they can output HDR. And the commercials are going to be in SDR for quite some time so the playout server has to be able to upconvert those commercials with minimal latency and then insert them into the playlist.
Those are obvious things. What is not so obvious is where you do the actual integration and, as I said, latency is what kills you.
Stewart: We validated that we can do SDR and HDR production using a single set of cameras, a single crew, a single truck, and we can do the workflow exactly the way it is done today. The only incremental cost would be the new cameras and all the trucks are already being equipped with new cameras.
The second thing we validated is that the way we are playing out today in master control will work for HDR. Any broadcaster can use the same tools, the same equipment and can work the way they do today.
Going forward we have to get the vendors in the industry to integrate the technology where it makes sense and we need to focus more on the distribution piece, pushing the signal through a QAM through a traditional cable network and over an IP connection.