This evening I attended an introductory webinar on colour grading using the Academy Colour Encoding System (ACES). Colour Grading and VFX is something I’ve always had an interest in, however it’s not something I’ve ever done professionally (I’ve worked on a few short films for free in hopes of building a portfolio and client base, but never for money).

In essence, the ACES colour management system which ‘removes’ the colour science of any supported camera and provide a non-destructive and consistent workflow for handling several different colour spaces/gammas. It does this by using a camera/picture-profile specific mathematical transform to convert the image into a super wide colour space, and then converting out of that colour space for whatever viewing device/delivery format you will be using. The advantage of doing this with a mathematical transform instead of a technical LUT is to avoid any clamping of the raw data. Using an ACES based workflow ensures the colourist has access to the full dynamic range of the original footage at every step of their process, and their grade can easily be delivered in an alternative colour space. Currently film is the gold standard for archiving, due to its wide colour gamut and dynamic range, ACES aims to eventually become the standard (it is still in development). Deliverables will still be made for todays delivery standards, and the ACES process gives consistency between different standards, meaning you don’t have to create separate grades for DVD/theatrical/web release. While we currently use delivery standards like Rec.709 or DCI-P3, what happens in the future when new standards are invented that allow for more data? If our archival format is limited to today’s standards, we won’t have sufficient information for remastering in the future when new standards arise.

This is totally unrelated to drama coverage, but thought I should write a reflection about it anyways, as it’s something I have spent a considerable amount of time researching and experimenting with this week.

The other thing I’ve learn about, while I’m on the topic of colour management, is the Cineon Log gamma developed by Kodak, which is historically used for digital storing of film scans. I’ve been researching the traditional colour grading/colour timing process for film formats, and have been amazed by what I’ve found. I’m so glad that I have access all of today’s digital tools, as they seem like they are real time savers compared to the analogue process (although I would love to try the analogue methods if I ever get the chance).

Left: Standard Rec.709 transform, Right: my film emulation process (Click to expand)

To put some of this into practice, I’ve put together a basic film emulation grade using these tools/techniques. Rather than trying to be creative in coming up with something that “looks cool”, I’m trying to be as technically accurate as possible here. The process I have used is as follows:

1. Transform into Cineon Log to ‘remove’ my camera’s colour science, then…
2. Adjust each RGB channel’s exposure using virtual printer lights, then…
3. Simulate halation using a linear gamma transform and RGB channel blurs, then…
4. Give my footage a film response using a scientific film stock emulation LUT, then finally…
5. Add texture using 16mm film grain scan.

Anyway, I won’t bore you with nerdy technical details anymore (who really cares about colour management anyways?). Thanks blog, for allowing me to indulge in some obscurities. (Lately I’m treating this page more and more like a diary?)

—–

Just as an afterthought… maybe theres no point in caring about doing things the “proper way”. If it looks good, isn’t that enough? The first counter argument that comes to mind, would be that you want something that looks consistently good, no matter which platform or year it’s being watched in. I think I need to be careful about getting too technical with what I make, and just throw some paint around on the canvas from time to time. So what if it’s not perfectly sharp, hey?