With all the hype surrounding 4K acquisition, I was surprised to hear that David Fincher's The Girl with the Dragon Tattoo would be the first large-scale end-to-end 4K digital cinema release. Previous films captured at 4K were posted and distributed at 2K... which kind of defeats the purpose. In my quest to find a 4K theater to see GWTDT, I got a response from Sony's Digital Cinema Twitter account, yet because I've been catching holiday films with my family (including young ones) I haven't yet had a chance to see the film yet (it seems I'm not the only one in this situation). Regardless, I wanted to share a very interesting article about the film's 4K workflow by Light Iron's Michael Cioni, which includes this interesting nugget on Fincher's approach to reframing in post (his framing chart is pictured):
On The Social Network, David utilized a technique that allowed him ample choices for reframing and stabilization by capturing 4K and 5K images with a 10% look-around pad that was pre-framed in camera. In the case of GDT, because EPIC was used, the look-around image could be increased to roughly a 20% pad... With EPIC, I believe David’s technique of a 20% look-around is something filmmakers should consider on all projects. The ability to take advantage of ample look-around space becomes a key component in reframing and stabilization (which are techniques being adapted by more filmmakers and more departments) but this reframing also allows for a much better transition to varying aspect ratios in different deliveries.
Before anyone calls me a RED fanboy, note that this technique is not specific to RED cameras, but to 4K cameras -- the Sony F65 being another example, along with ARRI's next ALEXA (probably), JVC's erstwhile 4K camcorder, and that mysterious 4K Canon DSLR. For everyone who claims that reframing in post is bad, here's one of the preeminent directors of our day designing his shoot around the technique. Also interesting: Fincher and co.'s choice to crop to 4K instead of downscaling, the latter of which has been lauded as a way to get "true" 4K images. A 4K RED image, after debayering, yields around 3.2K of measurable resolution, whereas a 5K image should equal a true 4K after debayering and downscaling. Perhaps because of TGWTDT's use of the RED ONE MX for 2/3 of the shoot, Fincher chose to crop his 5K files to 4K rather than downscaling. It will be interesting to see what folks shooting 100% on EPIC will do going forward.
Another useful tidbit from Michael's post: this Mac app from Blackmagic Design, which Light Iron uses for measuring disk speed.
As for the SCARLET, which is not really a 5K camera, you theoretically have less flexibility than with the EPIC for reframing while maintaining maximum resolution. But on a recent short, I found myself realizing that, because the SCARLET's native aspect ratio is 1.9:1, when framing for cinemascope (2.4:1) as we did, you do gain a decent amount of "look around" area in the vertical dimensions. Conversely, if shooting for 16:9 release, there is a tiny bit of cropping you'll need to do to avoid pillarboxing (better to just shoot in "4K HD" mode, AKA Quad HD, which yields an image 3840 pixels wide -- exactly twice that of 1080P). I've said on the record many times that I think the importance of resolution is overstated compared to color rendition, dynamic range, compression, and other factors. But I think one of the first things you realize when working with a 4K image (not just from RED) is that you do gain a consequential amount of flexibility.
Check out Michael's full article for plenty more insights about a high-end Hollywood 4K workflow.
