NATO held its General Membership and Board of Directors meeting in Los Angeles this month, volunteering 2 hours of time to an industry-wide meeting on the transition to SMPTE DCP-formatted distributions. The meeting underscored that SMPTE DCP continues to be a solution looking for a value proposition.
The history of the DCP deserves a recap. The DCP concept was first conceived and documented in December 2001, and presented at NAB in April 2002. Over the course of the year 2002, intense socialization of the concept took place, and DCI was formed. By the end of 2002, DCI claimed invention of the concept – the highest form of flattery (looking on the bright side). However, the SMPTE DCP standard was not completed until 2009, eight years after its concept. This was not timely, as the digital cinema rollout began in earnest in 2005. As the rollout required a distribution format, a working draft of the standard was tidied up and employed. We now call this Interop DCP. By the time the SMPTE DCP standard had issued, of course, it was different from Interop DCP.
There has always been a gap in leadership concerning SMPTE DCP. As if the flaw of making the standard different from Interop DCP was not enough, SMPTE added fuel to the fire by continuing to revise the standardized version. Further problems exist: DCI has never properly embraced SMPTE DCP. It couldn’t have embraced it the 2005 DCI Spec, since the standard wasn’t released until 4 years later. But even with “errata,” the DCI spec has never properly specified core system requirements for audio, such as audio routing. The result is that products that pass DCI testing are not guaranteed to play SMPTE DCP in its full glory, while commercial requirements drive full compliance and interoperability for Interop DCP.
NATO’s open meeting concerning SMPTE DCP was unbalanced. The meeting attempted to position the problem of transition from Interop to SMPTE DCP as an exhibition problem. By not upgrading equipment to latest versions, exhibitors were told that they significantly reduce the odds of being able to play SMPTE DCP. But this isn’t quite true, as discussed in the meeting, even upgraded equipment will not allow SMPTE DCP to play everywhere. There are structural issues that remain between actual hardware and ideal specifications. In fact, the two hour meeting was largely spent describing all of the issues that could occur when one plays SMPTE DCP formatted content. On the flip side, if Interop DCP formatting is used, the movie will play.
Confusing rather than helping, NATO’s technical consultant presented a not-so-valid value chain for the transition to SMTPE DCP:
- Interop not documented. In fact, Interop is fully documented. But every attempt has been made to not improve the documentation over the years in the attempt to bury it.
- Need for better naming and automation. There is nothing structural in Interop DCP that prevents better naming and automation. But there has been a strong political effort to prevent “naming and automation” improvements in SMPTE DCP from being applied to Interop DCP.
- Need for audio routing and external sync. Dolby Atmos content, the only form of released content to utilize SMPTE DCP, mimics Interop DCP audio in its version of SMPTE DCP, to avoid the fancy and troublesome aspects of SMPTE DCP audio, such as audio routing. Synchronization signals can work fine with Interop DCP, but again, politics prevent such improvements from being implemented.
- Need for more reliable subtitles. There are truths and non-truths in this statement. A lot of effort has been placed into SMPTE DCP subtitles to improve consistency of appearance. But these same improvements are not guaranteed to work across all implementations of SMPTE DCP. Again, the improvements could be applied to Interop DCP, but for politics.
- Need for automated key delivery. Any automated key delivery improvements in SMPTE DCP could also be applied to Interop DCP.
Exhibitors were presented with no research that captured the cost of upgrading equipment, including labor borne by the exhibitor, and upgrade fees required of manufacturers. There was no discussion for how SMPTE DCP would be rolled out to the other 66% of screens around the world, which will be very challenging. Why place a burden on US exhibitors to convert quickly when it will take many years to convert the rest-of-world?
Taking a different tact, there is no doubt that a lot of effort has gone into the development of SMPTE DCP, including work undertaken by your author. Is is possible that there are other ways to accomplish a transition? With nearly 130,000 screens deployed around the world working daily with Interop DCPs, it would seem more practical to manage this change in smaller bites. The problem with this thinking is that a fallback is needed, and Interop DCP is the fallback. In addition, managing incremental change is likely to create more problems than it will solve. Mastering is the unruly side of digital cinema, with over 30 DCP mastering systems listed in Wikipedia, none of which are subject to any compliance rules. Even if the vendors of mastering software were listening, their customers are unlikely to upgrade their DCP creation tools with periodic changes. So rather than asking to change something that works, it is better to keep working software in place and introduce SMPTE DCP as a parallel process. Similarly, the same is true on the exhibition side. No one is asking exhibitors to lose the ability to play Interop DCP. The goal is to simply add the ability to also play SMPTE DCP. But someone must absorb pain in the transition process. Either distributors absorb the pain of dual distribution, Interop and SMPTE DCP, for many years until they’re confident of moving solely to SMPTE DCP, or exhibitors must absorb the pain of quickly updating and testing systems to reduce the transition time to SMPTE DCP. The NATO meeting was staged to push the pain off onto the exhibitor.
With the current theme of making the transition less burdensome to distribution, and pushing the burden of transition onto exhibition, it would seem prudent to place some grease on the table. The question that exhibitors ought to be asking is if the transition to SMPTE DCP is so important, why are distributors not providing a financial incentive for its implementation? At a minimum, why are they not covering costs? If a financial incentive from distributors became the lever required by exhibitors, perhaps then we’d learn just how important the transition to SMPTE DCP really is.