Use Your Own Location Photos to Test a Film Idea, Not Pretend It Is Shot
Use Your Own Location Photos to Test a Film Idea, Not Pretend It Is Shot
A photograph of a room is real evidence about a small number of things. It can show what one person standing in one spot could see at one moment under the conditions that were there. For a film, that is often exactly what you're unsure about: whether two people can see each other, where an arrival first becomes visible, whether a doorway separates a public area from a private one.
It is not evidence for the finished scene. A still cannot show you the light at the hour you intend to shoot, the sound in the room, whether a performer crossing the space reads the way you want, or whether anyone will let you film there. Owning the file means you own the file. It does not clear the room or prove anything about the shot.
So give the picture a question to answer, and then write down which parts of the question it answered. That's the method. What follows is how to choose the views, how to caption them, and what to do when a photograph disagrees with the scene you already like.
Turn the scene into a question you can inspect
"Find a cinematic corridor" is not a question. It's a shopping instruction, and it produces a folder of appealing rooms with no scene in them. The question comes first, and it's made of named positions and a proposed action.
Rewrite the scene as a spatial uncertainty. Not "a tense meeting in a shop" but: a worker is behind the counter, a visitor comes through the door, and the beat needs the worker to see the visitor before the visitor sees the worker. Now you have a criterion for choosing photographs. You can say which views matter and which are merely pleasant.
An inspectable question states a position and leaves room for the answer to be no. "Can a person standing behind this counter see the door?" can come back no. "Does this place feel like the film?" cannot come back anything.
The shop in this section is invented. No such shop has been visited, photographed, or arranged, and the view schedule below is a plan rather than a record. The two outcomes I describe are hypothetical readings of images that do not exist. That's deliberate: you'll point the same shape at a room you can actually stand in, and the labels are what keep it honest.
Notice that the repair-shop beat is really two questions, not one. When does the visitor become visible to the worker? When does the worker become visible to the visitor? Those are two separate sightlines that can fail separately. Photograph one and you have half a study.
The last step is deciding what kind of question you're asking. Spatial questions — who can see whom, where the reveal lands — behave well in stills. Questions about timing, motion, and sound do not, and it's worth knowing which one you hold before you start.
Choose views that reveal the relation
A useful study usually has three parts, and none of them is the beauty shot.
One view establishes the space: where the door is, where the counter is, what sits between them. Two more views are made from the consequential positions the scene proposes — one from the worker's post aimed at the door, one from the threshold aimed at the counter. Where the relation is hard to hold in your head, add a rough plan with the positions marked. The plan isn't decoration. Drawing a line from one position to another is a claim about what lies between them, and it's the claim your photographs are supposed to test.
These views work together as an argument, which is why one selective angle is dangerous. A photograph from the worker's post that happens to frame out the doorway will look like a clean sightline and prove nothing. The views also have a hard limit: a set of stills has no time in it. Three frames taken minutes apart cannot show that a visitor's walk takes the eight seconds the scene needs, or that a reveal lands on the beat you wrote. Movement needs a different test, and stills should not be recruited to pretend otherwise.
Two things about the camera change what a view can claim. Height matters — a phone held at chest height is not the worker's eye, and a step to the left can open or close a sightline entirely. Field of view matters too. A wide lens contains more of the room than a person in that spot will notice, so a doorway sitting inside the frame is not the same as a doorway a visitor sees. "In frame" and "noticed" are different findings, and only the first one is in the picture.
Here's the schedule for the invented shop.
A — the room. Camera just inside the entrance, raised above head height, wide, framed to include the door, the counter, the shelving unit, and the aisle between them. This frame's job is layout: what exists, and roughly where. It's a diagram made with a lens. Don't ask it for a sightline.
B — the worker's post. Camera at standing eye height behind the counter, pointed at the door.
C — the threshold. Camera at standing eye height just inside the entrance, pointed at the counter.
The plan draws the door, the counter, a shelving unit along the aisle, and two dots for the positions: a solid line from the worker's eye to the door, a dashed line from the threshold to the counter, with the shelf crossing it. That dashed line is the hypothesis — the shelf may be what hides the worker from someone walking in.
If your visitor's arrival point is the threshold, A and C share a camera position. That's fine, but it means the shared position can't yield one frame doing both jobs: at a raised wide angle it maps the room, and at eye height it argues a sightline. A is the map; C is the sightline. If you want a hero wide from the entrance, you're adding a fourth view, and you should know that you're adding it for the pitch rather than the question.
Separate observed conditions from proposed treatment
Every view needs a caption, and the caption needs two halves.
The observed half records what the image actually contains: where the camera stood, at what height, at what time, in what light, and what was visible. The proposed half records everything you intend and nothing the room has shown you — where the performer will stand, what gets moved, what gets removed, how the light will be changed.
B — worker's post. Observed: camera at standing eye height behind the counter, pointed at the door, overcast afternoon, room empty. The door opening and the street beyond it are visible; the near edge of the counter fills the lower frame.
C — threshold. Observed: camera at standing eye height just inside the entrance, pointed at the counter. The shelving unit occupies the middle of the frame, and the counter sits behind it. Proposed alteration: the aisle shadows lifted on a copy so the shelving unit's position reads against the stock behind it; nothing in the room moved or removed.
That second caption is the important one. The editing was done on a copy, and without the note a reader would take the lifted aisle as the light the room actually gives. Any manipulation that changes the inference needs to be disclosed in the caption — a lifted shadow, two frames stitched to get the whole room in one picture, a coat rack removed because it would have stood in the visitor's way. A retouched study can still be useful. It can't be evidence for the thing you retouched.
Some questions a photograph simply cannot touch, and they should be marked untested rather than left to look settled by association. A picture carries no sound. A quiet-looking, empty shop is a picture of an empty shop, not a finding about whether a door chime carries to the back room or whether the street is usable to record on. The same goes for the parts of the room your frames miss, the dimensions you paced rather than measured, and every hour of the day other than the one you stood there.
Watch for the finish of the image, too. A well-exposed, carefully composed photograph feels like it resolved questions it never approached. The polish of the picture is not support for its claims, and it is the easiest thing in the world to let it carry a caption it hasn't earned.
Let the study change the scene or the claim
Now run the two outcomes.
In the first, view B shows the doorway clearly, and view C shows the shelving unit standing between the threshold and the counter. The asymmetry the scene wants is visible at those two positions and those two heights. That's a real result, and also a narrow one. Write the claim at the size it actually is: from the worker's post, the door is visible; from the threshold, the counter is not. What the pictures still don't establish: that the worker notices rather than merely could notice, that the visitor stays hidden through the walk, that the light at the intended hour preserves either reading, and whether the shelf will be there on the day.
In the second outcome, both views are obstructed. B shows a column and the edge of the shelf where the door should be; C shows the counter buried behind stock. The hypothesis has failed, and now there are two honest ways forward. Change the blocking — move the worker out from behind the counter to the bench beside the aisle, which is a change to the scene. Or change the beat — the worker registers the arrival by the chime, the change in light, a key in the lock, and only then comes out. That second move converts the question into a sound question, which is a genuine result: the study has told you what kind of test you need next, and it's not another photograph.
The failure layout is worth naming, because it's the one that looks best. A single flattering wide of the shop from the entrance, warm light, no people, captioned as though it demonstrated that the timing and concealment work. It demonstrates that you found an attractive room and that the light was good that day. It has no second position in it, no time in it, and a lens that sees more than a person standing there will notice. The two claims it's being used to support are precisely the two it can't reach.
When a study is inconclusive, the temptation is to polish the same photograph again. Resist it. Specify one small next step instead: a stand-in walking the route while someone films on a phone and calls out the moment the visitor's head crosses into the worker's line; or a listening test at the intended hour if the beat has turned audible. One test, aimed at the unresolved part.
Keep the project matters separate. Whether you may return with a crew, and whether you may publish the study image, are questions about access and rights, not spatial findings. Your ownership of the file doesn't clear the people in the background, the signage on the wall, or the property itself. Those get sorted outside the study, before anything goes into a deck.
Label the study, then name one next test
A study that stays honest is short and a little repetitive. It says what it is, what it shows, what it proposes, and what remains open:
Study — repair shop concealment. Photographed [date], from three positions, no site access secured and no shoot arranged. Observed: at the worker's post, the door is visible; at the threshold, the counter is not, because the shelving unit sits between. Proposed: worker positioned behind the counter rather than at the bench; shelving unit repainted, not moved. Supports: the concealment beat can be played as a spatial asymmetry rather than a performance trick. Does not support: the timing of the reveal, the light at the intended hour, or any sound. Next test: one take of a stand-in walking from the door while a second person marks the moment the threshold clears.
Two things are true of everything above. The method is a proposal about how to work, not a report of work completed — no location in this piece was visited, no photograph was made from these positions, and nothing here settles whether any real room behaves the way the invented one might. And the useful result is rarely the layout you wanted. If the photograph shows the shelf exactly where you needed open air, you have learned something about the scene for the price of a walk and a few frames. That's the trade worth taking. The image that argues with you did more for the film than the one that agreed.
Frequently asked questions
What can a location photograph actually establish for a film?
It can show what one person standing in one spot could see at one moment under the conditions that were there. That is useful for sightlines, where an arrival first becomes visible, or whether a doorway separates public and private. It is not evidence for light at the intended hour, sound in the room, whether a performer crossing the space reads as wanted, or whether anyone will let you film there. Owning the file does not clear the room.
How do you turn a scene into a question a photo can inspect?
State named positions and a proposed action, and leave room for the answer to be no. Instead of a tense meeting in a shop, specify a worker behind the counter, a visitor through the door, and a beat needing the worker to see the visitor before the visitor sees the worker. That yields two separate sightlines. Spatial questions behave well in stills; timing, motion, and sound do not.
What views should a location study include?
Usually three: one establishes the space—the door, the counter, what lies between them; two come from consequential positions, such as the worker's post aimed at the door and the threshold aimed at the counter. Add a rough plan with the positions marked. One selective angle is dangerous because a frame that crops out the doorway can look like a clean sightline and prove nothing. Stills have no time; camera height and field of view matter, and being in frame is not the same as being noticed.
How should captions separate observed conditions from proposed treatment?
Use two halves. Observed records where the camera stood, at what height, at what time, in what light, and what was visible. Proposed records performer placement, moved or removed objects, and changed light. Any manipulation that changes the inference—a lifted shadow, stitched frames, a removed coat rack—must be disclosed. A retouched study can still be useful, but it cannot be evidence for the thing you retouched.
What if the photos show the intended blocking does not work?
Two honest ways forward: change the blocking, such as moving the worker to the bench beside the aisle, or change the beat, such as having the worker register the arrival by a chime, a change in light, or a key in the lock, which converts the question into a sound question. If inconclusive, do not polish the same photograph again; specify one small next test, such as a stand-in walking the route while someone films and calls out when the visitor crosses into line, or a listening test at the intended hour. Keep access and rights separate from spatial findings.