The Future of 3D

Will Strauss#

Author: Will Strauss#

Published 1st January 2013


Over the last 3 years I’ve worked in more than ten different countries on stereo 3D; and no matter where in the world I am I’m continually running across the rumour that 3D has done its dash and will soon be heading the way of the dodo.
As with most rumours though I have yet to see any evidence of this. One of the most anticipated releases for France this year has undoubtedly been Asterix & Obelix. Pixar and Dreamworks have proven the suitability of 3D for animation, and it would be a very financially suicidal gambler to bet against the box office success of the recently premiered The Hobbit. Even India, which for economic reasons I would have thought would eschew 3D, has come to the party with the production of its first 3D dance film Any Body Can Dance.
If 3D was going to fail it would have done so years ago when the tools for creating it were still in their infancy. Since then though every major television manufacturer has invested the equivalent of the Greek national deficit in developing the technology to support 3D for their products (and continue to with Sony announcing this month the release of their latest screen that doesn’t require glasses). As for interest, Sky Digital’s 3D channel is moving from strength to strength. In fact the only thing slowing down its growth is the lack of 3D content to expand to further channels. Sky certainly have a strong dedicated audience for their 3D channel.
So if the demand is there why are there not more film makers running off to make stereo 3D content? For distributors it is much easier to generate interest and sell a 3D film. A 3D release will always generate more revenue than it would have done in 2D alone. In increasing cases films that were originally only intended for a cinema release also get asked for IMAX deliverables as well.
For many of those considering a 3D project their hesitation stems purely from fear. Fear of an initially steep learning curve and, more importantly for any investor, fear of the perceived increase in cost that moving a prospective project from 2D to 3D will incur.
Today I would argue that this fear is there to be overcome. Granted there are extra variables to take into account whenever a project is being considered for 3D production but this information is easily available and accessible. 3D is far from the dark art it was as little as five years ago; and there is now a wealth of experienced stereo consultants in the world to advise first time 3D film makers.
This leaves the financial considerations. As with most issues in film making though much of the expense and pain can be minimised through a bit of extra thought and planning in pre-production, and 3D is no exception. All of the common sense rules of 2D apply to 3D, the most obvious being to remove everyone from the planning process who regularly uses the phrase, ‘We’ll fix it in post.’
Certainly if you have the money there’s next to nothing that you can’t fix in post, particularly with the right tools; but if you want to come in under budget I would recommend thinking about your entire pipeline from the start and getting one or two 3D specialists in from day one. Those who have a ‘We’ll cross that bridge when we get to it,’ mentality are not the ones that you want hire, particularly since a great deal of the additional considerations for 3D aren’t rocket science - many are as simple as adding a depth script to the storyboard so that you know how ‘big’ you want your 3D to be from shot to shot.
So how can we make 3D affordable? The initial cost that makes first time stereo producers go white at the thought of a 3D job is the extra cost for cameras and rigs. Granted, having to budget for at least twice the number of cameras (and the storage to go with) will always tickle in a budget meeting but, to be fair, it’s often a relatively small percentage of the budget in the grand scheme of a film over-all.
No, the real cost here is in time. A single rig can take as long as half an hour to correctly calibrate and most rigs will require recalibration as often as the lenses are changed. Not good when the rest of the crew are charging by the hour to watch the camera department set up. A good crew can reduce this from 30 minutes down to 10 (Martin Scorsese’s Hugo and Ridley Scott’s Prometheus have done wonders for the professionalism of 3D camera teams in the UK) but that’s still time that piles up as the day goes on. This is where a savvy stereographer can save the day. Rig technicians are pedantic creatures, at least the good ones are. They want that rig perfect from shot to shot. The trick is knowing what is best to be done on set and what is faster (and cheaper) to do in post. If I’m on set and we’re fussing with a scaling error while the sun is setting then please do carry on shooting, it should take less than half a minute to fix in post. Most other simple geometry errors like rotation are also the same. The trick is to know when to take the extra time to get it right on the day, no matter how scared of the 1st AD you are. Lens flares and polarisation errors are two examples that are often better addressed at the time; although it would take a word count great than the one afforded me for this article to list them all.
But no matter how well calibrated a 3D rig is, or how experienced the crew, I have yet to come across a native 3D image that was shot 100% perfectly. It just doesn’t happen. There will always be something, whether relatively simple considerations like scaling and rotation, to more complex issues like differing lens distortions between the two cameras. These can be left to be corrected later in post but often they will be addressed in the dailies process as well, for several reasons. One is to see whether it actually can be corrected and to reassure the director that the shot is salvageable - an absolutely invaluable process as minimising pick-ups is essential to affordability. Another is to make viewing more comfortable for the director and editor in the off-line edit, something you will appreciate after your second hour of straight editing in 3D.
This is where the importance of considering your pipeline from day one comes in. If you have to correct the 3D in the dailies how do you use this in post? On far too many jobs I’ve seen these corrected rushes then recorded on to HDCamSR or rendered to MXF/Prores and that’s the rushes done. This means all that work has to be re-done on the uncorrected original files in the post process after conform as the tapes contain no history. How do we save the meta data? For meta data is gold in 3D.
The simplest way to deal with this is to use the same system for the dailies/rushes/onset visualisation as will be used for the final post production, a workflow that has obvious advantages. In fact in many cases 3D projects will blur the lines between production and post-production as file based workflows allow for post work to commence while production continue to shoot.
Granted, many of the dailies specific solutions available are cheaper than their post production cousins but, if you have to make these corrections anyway think how much time/money will be saved if you could use your dailies correction in post. Not to mention the fact that a high end finishing system will often have better tools anyway. This workflow has the added benefit of carrying any grading corrections through for the final colourist to access, remarkably useful when the director likes what the dailies colourist has done and wants the finishing colourist to match to the off-line quickly. The dailies grade will of course merely be a place to start, there may be considerable matching still to be done between shots or, in the worst case scenario, the director may just change their mind and throw it all out in favour of a new look. However, while the creative grade is arbitrary, good 3D is good 3D. Does the pixel in the left eye match its corresponding pixel in the right eye in Y disparity? Yes? Then you’ve got a geometric match between the eyes. Now do the same for any colour inconsistencies between the left and right eye and that’s the majority of your technical pass complete. Why do this all over again in post with a different system than the one used for the dailies?
Shooting natively is of course not the only option for producing live action 3D. I have begun to despair from the number of times I’ve been discussing technology with a thrifty producer and they’ve looked me in the eyes slyly just said, ‘Don’t worry, we’ll just convert the 2D to 3D in post,’ with the certainty that they are the first film makers with the cunning to hit upon such an idea. Just about every producer when looking at purchasing post production equipment will ask to see the ‘2D to 3D’ button with the absolute conviction in their heart that of course this system will have such a thing and that’s all that’s done in a 2D to 3D conversion.
And in some ways they are correct. There are many boxes out there that claim to convert 2D to 3D; what is debatable though is how well they achieve this. Every major post production house in LA that bills for 2D to 3D conversion work is looking for a way to fully automate the conversion process, which is what the conversion boxes claim to do. The problems with automation though are obvious as soon as one delves into the topic, a topic I don’t have time for today. Suffice to say that 2D to 3D conversion done well is no mean thing – you get exactly what you pay for and there are very few short cuts that meet with favourable results.
One further caution on conversion – a 2D to 3D conversion is not magic. Considerations still need to be made in pre-production for how the film will be shot. To make the most of 3D you must shoot with 3D in mind. Over the shoulder shots are not your friend, nor is shallow focus. Jump cuts aren’t usually a good idea either, and the lighting definitely needs to be considered with 3D in mind during the shoot. Long story short - you still need to shoot for the 3D that you plan on creating in post if you want to make the most of the medium, and make something that people will want to see.
Always remember that roughly 20% of people are extremely sensitive to 3D. That’s 20% of your possible audience. Which is why Sky have such high quality control standards for the material that they broadcast, standards that I agree with. If someone’s first 3D experience is a bad one, if they get a headache for instance, they will remember that and likely avoid 3D for good. That’s future viewers that you’ve lost for good due to a lack of technical pride.
Fortunately this is happening less and less as 3D literacy expands and film makers become more confident with the medium. With the right additions to your team and the right kit used making extremely good 3D is easily achievable. The financial pros certainly outweigh the financial cons if you have an eye for it and are savvy in your planning. Talent helps too of course.
In the end 3D is a tool available for you to use. If you think that 3D will suit your project then I highly encourage you to use it. With the right subject matter the results created by this extra dimension to your film can be spectacular. A film maker should never avoid a tool out of fear, and there’s nothing to fear about 3D anymore.
Sam Sheppard is an international film consultant specialising in script to screen workflows, post production supervision and stereo 3D with a minor in film-making in general. Sam was originally trained in New Zealand where he worked in post at Peter Jackson’s Park Road Post Production and on set for such films as The Lovely Bones and Avatar. As well as working internationally as a technical specialist for SGO’s Mistika, training and advising in digital workflows for 2D and 3D production, Sam is also a founding partner of Silverscreen Pictures.

Related Listings

Related Articles

Related News

Related Videos

© KitPlus (tv-bay limited). All trademarks recognised. Reproduction of this content is strictly prohibited without written consent.