On 26 March 2011, Barry Clark posted The Path to 3D4Kp48 – The Next Step for Cinema. While there had been some behind the scenes discussion on the topic at that point, the subject was not really mentioned online. Most any conversation ended quickly since no one really had the answers of what one set of manufacturers could do, or how long it would take, especially since there was no standard to work to. Everything in the dcinema world had been done to a standard, recently ratified by SMPTE, and many manufacturers were still working flat out to catch up to those.
Regardless, it has been a busy 7 months since then – there are still many questions waiting to be answered about the science involved. However, as ShowEast ends with a major announcement or three, perhaps it is time to summarize the situation.
First background point: There has long been agreement that capture should be done at higher resolutions than the final product requires. In the digital world this is most true since most every manner of manipulation of audio and video loses bits. In the press discussions and press releases there has been some slippage about making points that may be important in image capture that may not be as important in exhibition, or vice-versa.
Second background point: On the last day of CinemaCon 2011 (March 31), James Cameron showed experiments that he had done with actors and 3D cameras, captured at 24, 48 and 60 frames per second, each being displayed on a very complicated set up at 24, 48 and 60 fps. The full details can be read here: Cinemacon 2011 – High Frame Rate 3D Demonstration by: Reiner Doetzkies of TI and Geoff Burdick of Lightstorm. MKPE also has a write up at: Cameron, Showscan, and 3-D
Since then, Director Jackson and Director Cameron announced that they will be releasing their next big tentpoles in high frame rate stereoscopic; The Hobbit in 48 fps (December 2012), and Avatar II at 60fps (December 2014).
Where? How? With what equipment? With full encryption? At some standard? For, though this is entertainment, we need a product at the other end that will play the entertainment, probably on a variety of equipment since the studios are always on the lookout for anti-trust problems… and there isn’t a lot of time…and perhaps even people. (Obligatory science statement: Work=Force times distance, but only the magnitude of the force in the direction of the objective is un-wasted and significant.)
In fact, SMPTE has updated the standards for projection in the last year; where they were once exclusively US friendly, they are now world-friendly with 25 and 50 fps marks as well as US television friendly 30, and archival friendly with some sub-24 rates (in proposal/approval stage.) In doing this, equipment manufacturers – notably Texas Instruments (since Sony Projectors already could get to stereographic 60 fps), and the server manufacturers who one by one announced that they could meet the output challenge with full security and with the required open and closed captions.
The question quickly became, what is the input challenge? Neither Cameron’s nor Jackson’s groups appear ready to tell the world what they expect for the post-production deliverable. Critically, all the dcinema equipment in the field is set for the bar of 250 Mbits/second. It is unlikely that anyone would want the added compression that doubling the number of frames would require to fit in the same datastream.
And so, to now. At ShowEast USL announced that their new IMB (internal media block) has a compressed JPEG2000 data rate at 500Mb/s, with a data rate into the projector of 10.0098Gb/s. This produces full frame 2048×1080, 12 bit color planes at 60 frames 3D
Again, to compare, current servers deliver a DCI-specified maximum of 250Mb/s compressed JPEG2000 data rate and deliver a little less than 3Gb/s over two HD-SDI coax cable links to the projector. DCI also allows these devices to be 10 bit.
This will probably be the typical ‘standard’ since the math pretty much points there. So, what the manufacturers are being asked to do is double the compressed data rate from the server to the IMB and more than triple the decoded video data rate to the projector.
Given that there is approximately 50% penetration of equipment in the market, none of which can do this, one speculates what the rest of the industry will do. Will this type of speed require a large bump in costs, or will it become the expected, de rigueur, for systems in the future.
GDC announced that they would be showing something similar, with their IMB and SMS (media server) in a Barco Projector at an August convention in Japan. GDC to Showcase High Frame Rate Digital Projection System at BIRTV 2011. GDC has a relationship with Barco in the Far East.
Then at ShowEast, with USL’s IMB announcement, there was also an announcement from Christie. They will be including an IMB capable of high frame rates. Their release was also more publicity rather than a tech announcement…you know, because all the people using and buying these devices only need publicity because these are all commodity items now.
One thing that we do know is that the 4K TI units needed an IMB. Doremi has one. We see that GDC has one. Dolby showed theirs at CinemaCon. Qube showed theirs at CinemaCon as well, and announced that they are showing a working unit at ShowEast.
Finally, Digital Cinema gets the words “Future Proof”. It is on the GDC PR and on the Christie PR. One can presume with surety that previous units were not future proof. One company gave assurance with a 10 year warranty – which seems beyond reality if it is thought through.
One is also reminded that the VPF agreements are one-time affairs. The studios consider that they did their share and when the first tranche is over, they are on the road to saving distribution costs forever. Yet, here we see the exhibitors, some with equipment that is older than 5 years, getting to a reality junction. The next set of equipment will be bought from cashflow, not partially donated by VPF agreements. Future Proof…The New Black.
Back to high-frame rates: The following was also seen in someone’s PR. No need to mention whose since this style of hype will be with them all.
High frame rate technology also allows audiences to enjoy the 3D movie experience without the visual discomfort that has affected some 3D moviegoers.
So that is the reason that people have had headaches walking from the cinema! Wouldn’t you love to see those research papers? Because one reality can’t be unstated. Typically, only one plane is in focus in the human visual system and things that are moving by in the out of focus plane doesn’t have full motion resolution either…just like 24 fps film representation. So while stereoscopic can use some help, and while 48 fps and 60 fps material did look incredible compared to 24 frame, is it the panacea? Is it worth the cost?
The Cameron material shown at CinemaCon was also shown at 10 foot lamberts. Movies mastered for that level with the equipment tuned to consistently perform at that level sounds like a nice evolutionary step. Associated with that would be that every seat in the house have that level instead of the 50% falloff at 23 degrees off horizontal or vertical centers that high gain and typical silver screens ‘provide’. When most cinema auditoriums are at 3 foot Lamberts and most of the audience is seeing 1.5 or less, that seems like a more effective vector of attack for a larger number of patrons with the equipment available.