Celluloid Junkie » Digital Faces Critical (Archiving) Dilemma

Digital Faces Critical (Archiving) Dilemma

Posted by Carolyn Giardina | May 14, 2009 2:34 pm

digital_dilemma_cover1

“The Digital Dilemma”—a 2007 report from AMPAS’ SciTech Council that examines the challenges of archiving and accessing digital motion picture materials—was the subject of the SMPTE’s Hollywood Section May meeting, held at the Academy’s Linwood Dunn Theatre in Hollywood.

The report concludes that there is no long-term strategy for archiving and preserving digital assets that doesn’t require large capital investment.

Andy Maltz, director of the SciTech Council, reported that in 2008 the worldwide box office reached $20 billion, but he estimated that only 0.0015% of that amount was spent on addressing this dilemma. He emphasized that much more needs to be done.

During the evening, Maltz presented an overview of the report and an update on its impact since its release. He generated applause and laughter from an audience that has been inundated with 3D format talk when—in outlining the discussion objectives—he quipped “there will be nothing about stereoscopic movies tonight.”

Actually, he wasn’t entirely correct. He addressed AMPAS’ need to preserve digital content, citing as an example the increasing number of 3D titles that originate from and are/will be released in the digital realm. He also cited the ASC/DCI Standard Evaluation Material (Stem) and all Oscar-nominated films, which are archived by the Academy.

“Current technologies and practices are inadequate,” he said. “Periodically you will need to copy the digital media. Storage technology and operating technology will become obsolete.”

As to current investment, the report suggested that it costs $1,059 to preserve one film archival master for one year. In contrast, it estimated that annual cost of preserving a 4K digital master to be $12,514.

The SciTech Council is now working on what it calls a Digital Motion Picture Archival Framework, which would include the development and standardization of a file format, directed research, education and a follow up report on “The Digital Dilemma” for indie filmmakers and executives at public archives.

A very worthwhile read, “The Digital Dilemma” report can be downloaded from the AMPAS web site (here).

lto 4 backup

Really Storage technology and operating technology will become obsolete

brumeister

I've now read the document and the sad situation is that the technologies discussed for digital data tape is over 8 years out of date. With LTO-4 being available for over 6 years, the consideration and use of LTO-2 media for this type of storage (and the attendant costs involved because of the number of pieces of physical media due to LTO-2's 200GB capacity per tape limitation) are far from correct. As I mention in my first post, the capacity of a single LTO-4 cartridge means that a finished 3 hour 1080p HD film can readily be stored on a single data cartridge. Additionally, the increased performance of LTO-4 technology over LTO-2 (110MB/sec versus 30MB/sec) means that the creation and maintenance costs are reduced since the time required to create, duplicate, or recover data from an LTO-4 is dramatically reduced.

brumeister

Another thought about this -

From the article:

“Current technologies and practices are inadequate,” he said. “Periodically you will need to copy the digital media. Storage technology and operating technology will become obsolete.”

However, while this "seems" to be an issue on the surface, the cycle of jumps is measured currently in the 10-15 year range. We've been working with LTO-1 since the late '90s and we still have technology that allows us to read the original tapes.

[EDIT] - I just checked with my HP rep and the indication is that LTO-1 drives will continue to be available until at least 2015 - that's 20 years since inception.

Also, since we're working with digital data here rather than analog, each copy is as true to the source as the original. Therefore, even though there is a requirement to "refresh" the archival copy from time to time, the result of the refresh does not suffer from degradation or the requirement for restoration after 20+ years of storage.

brumeister

I'd love to know how this research was performed. The sanest method of long term storage for digital assets is on LTO-4 and pending technologies. An LTO-4 tape stores 800GB of data at a cost of less than $50. An average HD film production - including raw, unedited footage is in the neighborhood of 4.5TB. This equates to approximately $300 worth of LTO-4 media at current market prices. If you add in the one time expense of the LTO-4 tape drive or library, you're taking your startup costs to around $3K to $12K - but that is a one time charge that can be spread over many projects.

As to the numbers mentioned in the notes above, is the dollar value listed inclusive of media or storage only? I ask because while my numbers above are all related to the entire production content, a finished project will consume far less in the way of storage capacity with an average HD or 2K finished project fitting onto a single tape while a 4K project would possibly overflow a single tape and require 2 (until LTO-5 ships at year's end.

As for longevity, an LTO tape is expected to be viable for archival storage for more than 80 years - with some estimates when stored in a controlled environment such as Iron Mountain topping 150 years. I agree that technology will progress and potentially leave LTO technology in the dust, but the options for migrating data from an LTO tape to the large-scale storage du jour will be straight forward so long as the data owner keeps up to date on the technology advances.

It seems to me that the technical Illuminati at AMPAS are making a mountain out of a digital molehill.

Digital Faces Critical (Archiving) Dilemma

Related Posts