Netflix has more than 150 open source projects on its github repository, and while most are primarily of interest to engineers, two recent projects are very relevant to indie filmmakers. The first is a film—though if you stumbled on the short noir Meridian in your random Netflix browsing, you were likely very confused. The short doesn't make a tremendous amount of sense, but it's not supposed to.

It's a smart move for Netflix to invest in increasing adoption of the technologies it is going to depend on in the near future.


Meridian serves a purpose other than story. It's about staging scenes that are technically challenging to stream. The 4K 60 FPS HDR project was specifically designed to continue the process of developing release codecs for encoding high-bandwidth projects. To serve that purpose, the short film has been shared under the creative commons license; its source .mxf file is available here. (Fair warning, though: it's a 90 GB file, as befits a 4K HDR project, even though it's only 12 minutes long.)

Meridian_sky_treeshake_0Meridian shot showing smooth sky and shaking treesCredit: Netflix

Why share it? Netflix has a vested interest in improving technical standards in the industry. As the company moves toward primarily 4K distribution, and as HDR takes off on home televisions, Netflix is going to need more powerful encoding algorithms to ensure that the video quality stands up to audience demands—while also creating small enough file sizes to fit American bandwidth. The film itself is deliberately difficult to compress: dancing smoke, archival footage at various qualities, shaking trees, water droplets on glass, and smooth sky gradations (which often show banding) are among the host of fine detail that is often lost when streamed.

Netflix is hoping that the project will be used by other tech leaders as they develop the next wave of compression and distribution technology.

By streaming the film, Netflix will be collecting data on the performance of various networks. But the company is also hoping that the project will be used by other technology leaders as they develop the next wave of compression and distribution technology.

Meridian_smoke_2Meridian by NetflixCredit: Netflix

IMF delivery format

Netflix is making another push that indie filmmakers should know about: it's releasing software translation tools for the IMF delivery format.

For 20 years, DCP has been the standard for theatrical distribution, but DCP isn't the ideal format for online outlets like Netflix, Hulu, Amazon, or iTunes. For that, SMPTE (Society of Motion Picture and Television Engineers) has created an evolution of the DCP known as the IMF, or the Interoperable Master Format. The goal of the IMF is to make delivering individual versions to each outlet—known as versioning—easier by designing the format with multiple versions wrapped in one container. By creating an IMF, you should be able to deliver to a wider variety of outlets without having to create fresh, individual versions.

Netflix has brought its resources to the creation of a sophisticated test film deliberately challenging the technical limits of internet distribution.

SMPTE only has the power to create the standard; it can't force it on the industry. (They are an organization of engineers, not the industry technical police.) Unfortunately, Apple does not support the IMF format for iTunes, which can be particularly frustrating for independent producers and smaller distributors who don't want to spend the money to remaster for each format but would like iTunes revenue. To solve problems like this, Netflix has developed an open source IMF-to-iTunes translation tool.

While this would seem to benefit one of Netflix's competitors, it's actually a savvy move for the company itself. The easier Netflix makes it to use IMF, the more producers who will use it as their master—and the happier Netflix will be in the long term. Giving away tools to help IMF work better with all outlets will drive faster IMF adoption.

Meridian_endThe End of MeridianCredit: Netflix

Sidestepping "versionitis" 

Netflix is particularly interested in wider adoption of IMF from content creators to avoid what it calls "versionitis," or receiving assets from a variety of sources that need to be aligned together for release.

For instance, Netflix will always default to the highest quality original image possible, which is often 24fps film. If the film was released on foreign home video at 29.97, it's possible that the only foreign-language dubbed audio Netflix can get is 29.97... which won't match well with the 24fps picture master Netflix would rather use. If the producers had created an IMF master, they would've ensured that domestic and foreign versions all matched and were playable from the same deliverable. This not only gives the producer control over how their work is presented, but it also makes life much easier for Netflix.

It's a smart move for Netflix to invest in increasing adoption of the technologies it is going to depend on in the near future. And it's especially useful that the company has brought its resources to the creation of a sophisticated test film deliberately challenging the technical limits of internet distribution. (Demonstration projects have long been a staple of the film industry, as the ability to have a consistent source with known elements is helpful.)

Netflix has said there will likely be a Season 2 for Meridian to test evolving technical problems. We hope the company will take a cue from Kodak in the 1990's and feature a more racially diverse cast next time around to ensure that the full spectrum of humanity is reproduced well in future 4K HDR encoding formats.