Choosing a lens is not just a matter of deciding how much space fits in the frame. Tests compare how faces, backgrounds, light and movement relate to one another before a piece of equipment becomes a shooting solution.
Comparing images instead of relying on specifications alone
Nigel Galt recalls Kubrick examining lens tests converted into slides and projected for comparison. The approximate number he mentions comes from his memory of the process, not a definitive technical inventory. The most useful detail is the procedure: creating images under test conditions and observing their differences before selecting the equipment to be used. The choice was based on visible results, not just brand names or manufacturers’ specifications.
Two lenses with a similar focal length can produce perceptible differences in contrast, highlight reproduction, sharpness and performance towards the edges. A test allows these characteristics to be compared in a particular situation. It does not determine which lens is universally better. The decision depends on what needs to be preserved in the face, how much of the surroundings should remain legible and what response is considered suitable for the intended lighting.
Projecting the images also changes the viewing scale compared with a small photograph. A detail that seems acceptable in a reduced reference may become more apparent when enlarged. Comparison makes it possible to recognise variations that are not perceived in the same way during a quick inspection. This explains the usefulness of the procedure without attributing every possible criterion to an undocumented session. The testimony establishes that comparison took place; optical characteristics help explain the kinds of decisions it can support.
Preparation is connected with the material development of the shoot, but it is not the same as its complete chronology. An initial choice may require further testing when the set, a light source or the intended movement changes. The camera does not work separately from these elements. A lens accepted for one situation may present difficulties in another, so discussing selection means considering specific needs rather than an abstract preference for a particular model.
Available light and freedom of movement make different demands
Larry Smith explained that the special lenses used for Barry Lyndon were considered, but were not suitable for the mobility and framing sought in Eyes Wide Shut. This comparison does not establish a hierarchy between the two films. It shows that a solution developed for one exposure problem does not automatically answer another, where the movement of actors, the visible area and changes in camera orientation matter.
A large aperture makes it possible to work with less light, but it also affects the available margin for focus in certain conditions. An actor moving closer or further away requires that relationship to be controlled during the movement. The need to keep a performance legible can conflict with the desire to use low-level light sources or preserve the background in a particular way. Tests help assess these compromises before they become irreversible during a take.
The forced processing used in the film is part of the same decision-making system, although it belongs to a different stage. A lens’s response cannot be fully evaluated separately from the film stock, exposure and laboratory treatment being planned. A reflection or highlight may take on a different significance when viewed in the processed material. A useful test aims to approximate the conditions that will later need to work together.
The photographic look makes it possible to study the result of these relationships. It does not, however, allow every shooting parameter to be recovered automatically. The impression of proximity depends on camera distance, framing and reproduction size, as well as the lens. Without documentation of a particular shot, assigning it an exact focal length by intuition can create false precision. It is better to describe the perceived spatial relationship and the technical information that is actually known.
Choosing a set does not mean having a shot-by-shot list
A kit can include lenses with different characteristics to meet different needs. Selection does not require a single lens to define the entire film. General testimony about Kubrick’s preferences can help identify tendencies, but it cannot replace camera reports or production records. Expressions such as ‘most’ or ‘approximately’ should be retained when the source uses them: turning them into exact figures changes the nature of the evidence.
The images combining the actor with street backgrounds also raise their own integration challenges. The relationship between scale, perspective and movement needs to be consistent with the background material. A choice of lenses can contribute to that solution, but the result also depends on the process used and how the shot is organised. Knowing that rear projection was used is not enough to deduce which lens was used for each take.
The tests retain their value even if a complete record of every choice has not survived or has not been consulted. They help establish that the look did not arise solely from adjusting the camera on the day of the shoot. Alternatives were compared and a decision was made about which ones best met certain needs. This is different from assigning symbolic meaning to a focal length or imagining that every lens change corresponded to a psychological change in the character.
The process can be understood as a chain of verifiable questions: what was filmed for comparison, how it was examined, what criteria were discussed and what selection was made available. Galt’s testimony covers part of that chain; Smith provides context about lighting and movement. The film shows the result, but not all the tests. Keeping these sources distinct makes it possible to explain the optical preparation without inventing an exhaustive inventory or turning an approximate recollection into a laboratory record.
References: Stanley Kubrick, Eyes Wide Shut (1999), original English-language version and film credits. Nick Newman, interview with Nigel Galt, The Film Stage, 20 November 2025. Stephen Pizzello, ‘A Sword in the Bed’, American Cinematographer, October 1999, interview with Larry Smith.
