EA – I’ll remember March 2020 forever, as the moment when the virus COVID-19 changed my professional teaching practice and workplace, virtually overnight. In the first weeks of March staff at Griffith Film School (GFS) listened closely to every announcement from the Federal Government, anticipating that universities would soon be declared closed. At the time, like most film schools, GFS had few fully online courses.
The second year students were right on the brink of starting to produce their major assignment – eight sixty-second television commercials (TVCs). Ideas had just been pitched, projects selected and teams formed. There was a real buzz of excitement about the projects with some ambitious plans on the table.
On 19 March, GFS Faculty took the very difficult decision to halt all face to face classes and access to cameras and lights. Effective almost immediately, and certainly within two weeks at most.
On 20 March, an urgent meeting with the eighty odd cohort was held to explain the new circumstances and the way forward. I outlined the new plan. No TVCs would be produced this term. But all TVCs would be developed, designed and pre-visualised to a high level, ready to go into production later after restrictions were lifted. The deliverables required from the producing and the design students were mostly unchanged and they could still make most of their course requirements.
The cinematography students, however, could not deliver the principle item; finished TVCs. Indeed, without confirmed locations, they couldn’t even do final drafts of equipment and lighting plans.
The new plan, to enable them to effectively create high-level realistic pre-visualisation was to train them to use Cine Tracer, the cinematography simulator in a game engine created by Matt Workman. I’ve been aware of Cine Tracer for some time but it has not been included formally in the GFS curriculum. A few third-year students used it in 2019 to great effect on graduate projects.
We invited one of those students and a production designer to demonstrate the power and the incredible usefulness of the program, particularly as a tool for communication between cinematographer and designer. There were mixed reactions to the plan, as everyone was still in shock. Students were trying to work out if they’d stay enrolled in the course, and what the consequences might be if they stayed or left. The anxiety and stress levels were high.
Behind-the-scenes, a lot of researching, up-skilling, juggling and pressurised lobbying commenced. This was a new resource that wasn’t budgeted for. How many cinematography students had computers and networks capable of handling the program? How would they share and communicate with each other and their design and producer peers? Students without adequate computers had to be supplied with GFS computers. All of which had to happen very fast before the building shut down. It was full on.
Having announced the plan, we now had to implement it. I’ve not used Cine Tracer before, and few of my colleagues had either. It is very much still in development, and changes are updated frequently. So I called Ben Cotgrove.
BC – It was a cold and miserable day in London on 23 March when Addis called me. It was a significant day as it was London’s first day of lockdown due to the worsening outbreak of COVID-19. Teaching a practical cinematography course online? The prospect is so foreign and displaced from my education at both GFS and at the Australian Film Television Radio School (AFTRS) which couldn’t have been more practical and hands-on. How do you practically learn cinematography without even holding a camera? This wasn’t just an interesting opportunity for me but a relief from the abundance of free time I found myself with as the film industry slammed on the breaks here and, seemingly, worldwide.
At the end of 2017, I graduated from a Masters in Film Production completing a thesis titled ‘3D Pre-visualisation Methodology for Cinematographers’. Here’s the short version: hundreds of hours spent learning and experimenting with 3D programs, specifically Cinema4D and a plugin made by Matthew Workman called Cine Designer, integration with industry-level motion capture studio tools and UnrealEngine4 to create live pre-visualisation using virtual actor characters and a virtual camera. Through all of this experimentation, I developed a simple methodology that would structure how cinematographers could use 3D pre-visualisation in their own pre-production and test it by pre-visualising a short film before shooting it for real.
Flash-forward to 2020 and this thesis made me uniquely, and conveniently, qualified to help Addis virtualise the system of teaching practical cinematography. I had already discovered the huge potential 3D pre-visualisation could provide to cinematographers in that you can effectively build, block, light and shoot a scene virtually and have many elements in relation to camera and lighting be near photo-accurate. However back in 2017, this took a huge amount of resources and time. Luckily for us, and unsurprisingly to cinematographers, technology changes fast. We now had Cine Tracer. A program that came out of Cine Designer which I used in my Masters but operated on a games engine and made virtual cinematography, real-time, fast, easy to use and easy to learn. Perfect!
Not quite. Cine Tracer isn’t your ordinary simulator, it’s complicated in that it’s still early days in its development and is technically pre-release software or alpha software. It definitely doesn’t yet have the same polish of a video game you might buy at JB Hi-Fi. It can be buggy, it can change significantly between updates, and without a doubt requires a decent understanding of lighting physics and a basic understanding of 3D terminology and tools in navigating its operating system and building system. Nothing exists quite like it on the market, particularly something so focussed on the requirements and needs of a cinematographer. A quick YouTube search will provide you with many examples of cinematographers using it to pre-visualise incredibly interesting and detailed scenes.
So came the hard part; getting students from downloading Cine Tracer for the first time to fully pre-visualising a TVC production for assessment in a matter of weeks and most importantly, understanding the technical inaccuracies of the program so that they could still effectively implement their camera and lighting choices on set when the time eventually came to shoot these commercials for real.
Addis and I worked together to map out a framework for a series of five tutorials that would achieve this while combining the original outcomes from the original course framework with the new Cine Tracer learning environment. Everything from genre lighting to camera movement in commercials.
The first tutorial felt like it went by in the blink of an eye. Students who knew nothing about Cine Tracer going in hopefully now knew how to build a complete set, navigate the program, and use its sometimes complicated time-of-day and global illumination system to begin. Hopefully, the students were able to keep up with the technical information dump. The homework would be telling.
Students were asked to build their own scenes and sets for the following week and they definitely didn’t disappoint. We were presented with sprawling open-plan homes, glass atriums, nightclubs, dingy motel rooms and even car-crash scenes. It was varied and diverse, with great ideas and different levels of execution. But significantly, all showed that the software and skills needed to operate it were approachable and easily acquired.
Week two we rounded out the base learning by placing virtual characters, lighting our scenes using replicas of SkyPanels, Source4s, frames, Par HMIs along with practical lights. We integrated art department choices that are key to lighting including tweaking the reflectance, surface texture and colour to change effect and result with the lighting. We even simulated wet downs on roads by using reflectance, rain emitters and lighting.
Students were beginning to see the power of the virtual space to test and experiment with ideas of their own. We had to continually speak to the limits of the program and the elements that were not or are not yet photographically accurate, particular with lighting, but we made it an important note to discuss how you could work around these limitations and then factor in what you would actually have to do in the real world.
Such limitations included the games engine providing a base level of ambience that wasn’t necessarily realistic to the real-world conditions. We had to remind students that they would need to cover themselves on set with a way to create ambient fill light and not rely on natural ambience as it was presented in Cine Tracer.
In the final thirty-minutes of the class, we demonstrated the construction and lighting of a scene from No Country for Old Men (2007, cinematography by Roger Deakins CBE BSC ASC) with Tommy Lee Jones and Woody Harrelson in a motel room. We both found recreation a great teaching and learning tool and a students ability to analyse a frame from camera to lighting is key in knowing how to implement those choices in their own work.
Students watched as practical light sources were moved around to find the right angle of light and shadow direction, as distance between characters were changed physically and through focal length choices and how film lights were used to augment the practicals to give more control and definition to the justified lighting.
The test got us close, in a matter of thirty-minutes to what the scene looked like on the screen and was definitely passable as a form of pre-visualisation. Suddenly it felt like we were teaching a real-life physical lighting class and not a how-to use a computer program tutorial. In reality, it was 3am in London, 12-midday in Brisbane, and we were all seated in our bedrooms staring at our computer screens.
We began to focus on self-learning as a way to further develop the students ability with the program and cinematographic skills. Exercises were set where students had to light scenes and show how they would relight for continuity as the camera placement moved around a scene. Students tested recreating the same shot using a hand-held camera, jib, crane and dolly to create discoveries around camera movement and how it affects storytelling. Students were now using Cine Tracer to experiment, analyse and draw conclusions on which cinematographic choices they might implement in real world productions.
It was empowering to provide feedback that students could immediately employ in their work and decide for themselves if it improved their sets and scenes. This instant learning gratification that is objectively harder to replicate in a physical learning environment. Some students flourished and took to the program quickly while others required a few weeks of exercises and feedback to find their footing.
Students who have embraced working with CineTracer have demonstrated massive learning about importance of scale, textures and colour within a location; they’ve investigated styles of camera movement and use of big lamps not available to them within the course, big camera crane moves for example. The depth of their learning about lensing and lighting is far greater than is possible in the limited time of normal hands on classes. Most importantly they were telling and developing a sense of story in these scenes using the virtual camera and lighting.
Off the back of this learning, some students had already used Cine Tracer to pre-visualise a mood-film to pitch on a real television commercial project and others had recreated scenes from previous projects they were unsatisfied with to discover how they might have lit or shot those scenes in retrospect. The benefits of Cine Tracer and the ways it could be employed by the students had become a run-away train if a run-away train could be good.
As with anything, there are some students who have not embraced it. The final measure of that will be evident when they all have to submit individual work made in Cine Tracer for an
assignment separate to the TVCs. We eagerly await their reflections on the assignments, as some will undoubtedly describe the difficulties and challenges of the program, as well as the enormous benefits of being able to adjust everything in a set, so quickly and see what works, and change what isn’t working.
Will CineTracer become part of the assets used to teach Cinematography at GFS? Yes. Is it a replacement for learning hands on? Absolutely not. However, it’s clear that tools are quickly emerging in the field of virtual cinematography that will become integral to the landscape of teaching cinematography and working as a professional cinematographer, and it would be remiss to ignore that significance even in these early stages of development.
The true magic of Cine Tracer is that it is a brilliant tool for exploring possibilities, in preparation for real world challenges, and the proof is in the pudding. Cine Tracer helped close the distance between students and their access to a practical education in cinematography during the time of COVID-19.
Erika Addis is head of cinematography at Griffith Film School and Australian Cinematographers Society (ACS) Queensland president, national vice-president and was founding Chair of the ACS’ Women’s Advisory Panel.
Ben Cotgrove is an award-winning cinematographer based in London by way of Australia.