FilmSkool — a free film school
VFX — the primer
The department that has failed the moment anybody notices it, and which fails far more often in the schedule than in the artistry.
VFX · Post · 89 lessons · 17h 57m
What it actually is
Visual effects is the widest department here by a distance. It runs from a two-minute rotoscope removing a boom from the top of a frame, through digital set extension, to entirely synthetic creatures with their own musculature — and the only thing those jobs have in common is that the audience is meant to accept the result as photographed.
The single most useful thing to understand early is that most visual effects work is not creation, it is integration. Making a convincing dragon is a solved problem. Making that dragon appear to have been in that specific piece of footage — with the same grain, the same lens distortion, the same slightly wrong white balance, the same motion blur from the same shutter angle — is where the work is, and it is why compositing rather than modelling is the load-bearing skill.
The other thing that catches people out is that visual effects is a pipeline discipline before it is an art. Shots pass through many hands in a fixed order, and a change made late has to be pushed back through every stage downstream of it. Almost every visual effects disaster you have heard about was a scheduling and approvals failure rather than an artistic one.
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.
- Practical versus digital — The romantic position — a real explosion, a real model, a real puppet — has a great deal of evidence behind it and a great deal of survivorship bias in front of it: we remember the practical effects that worked and have forgotten the ones that looked like rubber. The counter-position is that the argument is a false binary, since nearly every celebrated "practical" shot of the last twenty years has been cleaned, extended or composited. What is genuinely true is that something photographed gives you a reference nothing else does.
- Whether previs constrains the film — Detailed previsualisation lets a director see a sequence before committing money to it, and lets everybody agree on what is being made. The objection is that it freezes decisions at the point of least information, that crews then spend the shoot recreating an animatic made by someone who was not there, and that the accidents which make sequences memorable get designed out months early.
- Where the shot is finished — A recurring fight is whether visual effects should deliver a shot that is already graded and final, or a neutral element that the colourist finishes. Delivering finished shots means the artist controls the result and the grade cannot break it; delivering neutral ones means the film stays consistent and can be changed late. Both produce a specific failure — a sequence that does not match, or effects work degraded by a grade applied on top of it.
On the day
Everything you fail to record on the day gets rebuilt later at roughly ten times the cost, and some of it cannot be rebuilt at all. Clean plates, reference stills of the lighting, a chrome ball and a grey ball, the lens and its distortion, measurements of the room: twenty minutes on set, or three weeks of somebody guessing.
Where to start
- Compositing — The centre of the department. Everything else produces elements; this is where they become a shot.
- Practical Effects — Worth watching early even if you intend to work digitally, because it teaches what a real element behaves like.
- Camera Tracking — The unglamorous foundation. Almost nothing can be integrated into a moving shot until this is solved.
- Pipeline — The part that decides whether a project delivers, and the part nobody watches until their first missed deadline.
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.