We’ve just finished our latest Stereo 3D shoot and I thought I’d get my thoughts down on this ever-growing speciality. As 3D TV and films increase in popularity the amount of 3D content is growing, but what makes for good 3D and film makers squeeze the best from this new medium.
There’s loads of technical information that filmmakers need to consider when creating films in 3D, phrases like Interocular Distance and Convergence Points fly around constantly. The web is filled with places to find info on the technical aspects of recording in stereo, but for me the role of a director is to a tell a story and the tools of the trade in 3D are slightly different than in 2D so I’m going to talk through those aspects. Like many in this business I’ve spent my years learning how to use selective focus to draw the viewers eye to the information that i feel is important. The obvious example would be to have an object at the front of frame in focus whilst the background is out of focus… but this is not something that translates well in stereo 3D. The viewer’s eyes naturally scan the whole image and their own eyes do the focusing, that’s why in 3D the game is to maximise the amount of a shot that’s in focus. Without having blurred backgrounds setting up 3D is more like setting up a live play, there’s a greater emphasis on lighting to accentuate your key subject. Like any rule this is meant to be broken but its certainly true that extremely narrow depths of field just don’t seem to work well in 3D.
The main thing to think about when making a 3D film is that it can be a slower process…your moving two cameras around and not one, the mirror rig is delicate and heavy, there are large location monitors to check your depth and are a whole load of new processes that have to be done to make the 3D work. The up shot of all of this is that you can capture only about a third of the footage in 3D shoot compared to a 2D shoot so careful planning of shots and tight scheduling is the name of the game.
But the fact that shoots are slower is not all bad news and that’s down to how viewers actually watch 3D. A good 3D shot, with depth through the frame and lots for the eye to look at takes a lot longer for viewers to assimilate. The up shot of this is that edits are by their nature slower. Cuts that would be a second and a half long will run for 4 or 5 seconds in a stereo film. While talking about edits a good trick is that when looking to inject pace through fast edits, the “3D’ness” of the shots should be reduced. This gives the viewers eyes less work to do and edits flow more smoothly. Equally, if a shot has a large volume for the eye to register it’s worth slowing the pace of the edit to allow viewers to fully appreciate what’s on the screen.
This all brings me finally to the big question – what is good 3D? Well, i’m afraid there’s no straight answer to this one and its all dependent on where the film is being shown. At a technical level the 3D that passes safety guidelines for TV will be a lot less punchy than for a 4 minute film at a live event, a 2% divergence is fine for TV but that doesn’t mean something pushed up to 4.5% at a exhibition is bad. It really just depends on what works when you watch it… and thats something that no amount of guidelines, books and look up tables can prepare you for, the only help is experience.
As a background to our work here’s some background to our latest 3D video shoot. We filmed at a huge mechanical plant which presented lots of challenges, not just how best to create punchy 3D but how to keep top end cameras clean in pervasive dust and communicating with ear plugs in!
We are currently editing a 3D marketing film for use at a exhibition and I’ll be uploading the film as soon as it’s been shown.
Here’s the technical bit: Shot on the PS Technic mirror rig in full 1080p HD on Sony F3’s. The 3D film is being delivered as a single “side by side” HD quicktime file for viewing on HD TV’s at the event.