Last year, the American Society of Cinematographers and the Producer’s Guild of America orchestrated a follow up to 2009’s Camera Assessment Series with the Image Control Assessment Series. Unlike a shootout where the relative strengths and weaknesses of the selected cameras were pitted against one another, ICAS was an attempt to unify the different camera systems under a single workflow: the Academy Color Encoding System, or ACES.
ACES is a bit of a boogeyman amongst cinematographers and postproduction houses. People have heard of it, they don’t know a lot about it, and no one seems to want to mess with it. The first reason for this might be that ACES is a new standard being orchestrated from the top down by the Motion Picture Academy “with the participation of more than 50 of the industry’s leading technologists and practitioners” at a time when the industry is becoming more and more decentralized. Red and Sony and Arri all have their own unique workflows that get the most out of their cameras. In order to comply with ACES standards, camera manufacturers must co-develop firmware-specific input device transforms (IDTs) that translate their native camera output encodings into the ACES color space. At the moment, ACES hasn’t been fully embraced by all of the camera manufacturers, and not all of the post facilities have upgraded to the 16-bit 4K pipeline. As cinematographer and chairman of the Camera Subcommittee of the ASC Technical Committee David Stump, ASC notes, “I’d say we’re more than 75 percent of the way there.”
Simply put, the ACES standard is the Academy’s answer to 35mm film. (Official explanation here.) They don’t really care what you do with it, just so long as everything you do is in 16-bit OpenEXR files. (ACES is resolution agnostic.) At that bit depth any ACES-capable post house can pretty much do anything a filmmaker wants to a digital image, particularly if they’re capturing or scanning to 12-bit to 16-bit raw files. So instead of cramming your image into a 10-bit DPX your ACES file grants you the full latitude of your digital sensor or film scan.
“There’s a rule of thumb that in order for a camera to claim a certain stop latitude, it has to have a certain number of bits,” Stump explains. “Number of stops plus 1 equals bit depth. If you’re building a camera with an 11-stop latitude, you can generally fit your full signal into 12 bits. Sony built the F65 to do 14-plus stops which is why they put it into 16-bits. Otherwise the engineering effort of that camera would be ill-spent if they were simply jamming the entire signal into 12 bits.”
So, can we assume that if a camera can capture either 12 or 16-bit color in 4K and if ACES is truly as flexible as they say it is, does it even make a difference what camera we use?
“With all the cameras on an equal footing, the workflow is no longer the weakest link,” he replied. “We start to expose the weaknesses of the cameras. With the ICAS tests we tried to find out what haven’t we been seeing, and what could come out of these cameras in the future. For me, the takeaway is that there’s no free lunch. You get what you pay for, and if you pay more for a camera there’s a pretty good chance that you’re getting more in the camera.”
The takeaway precludes professional digital cinema cameras that still capture 8-bit images, like the the Canon C100, C300 and 1D C, and the Sony NEX-FS100 and FS700. (Not taking into account the recently announced HXR-IFR5 interface unit and AXS-R5 recorder and Convergent Design’s Odyssey 7Q.)
“Right now ACES is a high-end, elite workflow,” Stump remarks. “It’s like that for all new technologies. Maybe the first iPhone cost a billion dollars to build but the 2nd million cost 50 cents each. Once ACES is finished and once it becomes more ubiquitous, then it gets to trickle down and then everyone can avail themselves of its advantages.” He doesn’t think ACES will fully replace the venerable DPX file format–it will “become a middle of the pack tool” available to filmmakers dependent upon budget, schedule, and resources.
So what can ACES offer the independent filmmaker working with a relatively low-end production? As magnetic tape-based LTO storage hits the backend of its adoption curve, even storing 4K masters in ACES isn’t a sure thing. “I’m beginning to hear failure stories where people pull the tapes out to re-release a movie or repurpose the content and they’re finding out that they’re all dead,” Stump says. At the moment, he notes with some bemusement, the best option is still film.