FilmSkool — a free film school
Unreal Engine — the primer
The only department where one person does every job on the unit, which is its whole promise and the entire reason it is hard.
UNR · The New Wing · 173 lessons · 60h 24m
What it actually is
Unreal is a game engine that the film industry borrowed, and almost everything confusing about learning it follows from that sentence. It was built so that a world could be explored in any order by somebody holding a controller, and a film is the opposite: one path through that world, decided in advance, seen once. So the tool will happily let you build something magnificent that is not a shot, and it will never mention the difference.
What it actually gives a filmmaker is a set that costs nothing to strike and light that costs nothing to move. That is not a small thing — it is most of what a budget buys. You can put a camera at a height no crane would reach, run the take again with the sun two hours later, and change the building in the background between takes. The constraints that shape ordinary film craft are simply absent, and the first thing everybody discovers is that those constraints were doing a lot of work.
Which is why this wing is organised as eight departments rather than as a tour of the interface. Inside the software there is no line between building the set, lighting it, operating the camera and rendering the master; you are one person in one window doing all four badly at once and unable to tell which one is going wrong. Every professional pipeline built on this engine puts those lines back. Learning it as a film school subject means learning them first, and the panel layout second.
One more thing worth saying early: real-time is a claim about the renderer, not about you. The picture updates while you move the camera, which is genuinely revolutionary for blocking, previs and anything shot on an LED volume. It does not mean the work is fast. A shot that looks finished still took somebody a week, the same as it always did.
What is still being argued about
Live disagreements between people who know what they are doing. You will find confident advice on both sides of each.
- Whether an LED volume is a set or a very expensive rear projection — The case for is that the light is real — a face lit by the wall is lit by the environment it appears to be in, which is the thing green screen has never been able to fake and the reason actors and DPs both prefer it. The case against is that a volume costs more per day than most films have in total, locks your background at the moment of the shoot, and has a hard limit on how far the camera can move before the illusion falls apart. Both sides are right, and the honest version is that it is a tool with a narrow and real sweet spot — vehicle interiors, windows, anything where the background is far away and the camera is not.
- Whether to light it like a set or like a render — One school treats the engine as a stage and brings a gaffer's habits to it: motivate every source, work in real-world units, place instruments where a rigger could actually have hung them. The other points out that there is no rigger, no ceiling and no budget, so refusing to put a light where a light could not go is a superstition imported from a world with scaffolding in it. The first produces images that read as photographed; the second produces images that read as beautiful and, quite often, as not quite real. Almost everybody good at this can do both and chooses per shot.
- Whether any of this belongs in a film school at all — The objection is serious and is usually made by people who teach: that a student who learns to build a world before they can stage a two-hander has learned the expensive half first, and that engine skills date while staging does not. The answer from the other side is that virtual production is now on the call sheet of a large share of episodic television, and a school that will not teach it is choosing to be behind. This wing takes the objection seriously enough to sit last in the prospectus, for the same reason the AI wing does — but it is here, because the jobs are.
On the day
On a volume, the thing nobody warns you about is that the engine is now a department with a crew, a call time and a failure mode. Somebody is running the wall, somebody is tracking the camera, and when the scene will not load you are not waiting on a lamp, you are waiting on a build. Budget the same contingency you would for a location you have never seen — because in every sense that matters, that is what it is.
Where to start
- Unreal Basics — Do the boring part on purpose and do it once. Everything downstream assumes you can navigate a viewport and organise a project, and nobody teaching lighting is going to stop and explain it.
- Virtual Camera — Go here second rather than to environments, even though environments is the fun one. A cine camera with a real focal length is the piece that turns a 3D scene into a shot, and it is also the piece your existing film knowledge transfers into directly.
- Unreal Lighting — The largest single difference between work that reads as a film and work that reads as a game, and the department where a gaffer's instincts are worth the most.
- Unreal Environments — Now build something to point the camera at. Later than you want it, and in the right order.
- Made in Unreal — Finished films, taken apart. Watch these early as well as late — they are the only honest answer to "how long does this actually take".
Read next
This is the plain index, shown because JavaScript is off. Every lesson here is a free video published by the channel credited on it and plays through that channel's own embed — FilmSkool hosts nothing.