A larger image spreads the same delivered light over more area
A projector’s lumen figure alone does not describe the luminance of the image seen from a seat. The light reaching the image, its area and the screen’s directional response all matter. For a flat rectangular full-white image, width W and height H in metres give area A = W × H in m². If Φ lumens reach that entire rectangle, the area-average incident illuminance is E = Φ / A in lux.
This is a flux-per-area relationship, not a claim that every part of the image receives equal light. A centre lux reading is one point; it cannot replace an area-average value without evidence that it represents the whole rectangle. The alternative lux input must describe the projector’s full-white contribution at the screen plane, excluding ambient light.
The luminance budget
Under the declared uniform screen model, L = E × G / π = Φ × G / (π × A), where L is luminance in cd/m² and G is matching directional screen gain. One cd/m² is also called one nit. The model normalises gain to a unit-reflectance Lambertian reference and applies the same gain across the rectangle for the declared illumination and viewing direction. The resulting single value describes that model, rather than measured centre luminance or an actual image-wide brightness map.
Visual Displays’ projected brightness calculator publishes this theoretical gain–lumens–area relationship and its reverse. For a supplied target L, the corresponding screen-delivered flux is Φ = π × L × A / G. That is the light required at the image, before any separate calculation of what a particular projector must emit.
The following invented example holds the declared direction and uniform-gain assumptions fixed. It is not a screen specification or a recommended brightness target.
Doubling both image dimensions quadruples the area, so the same delivered light gives one-quarter of the modelled luminance. Conversely, doubling a supplied target at fixed area and gain doubles the required delivered lumens. These comparisons hold only while gain and the other declared conditions remain applicable; resizing a real image can change optical settings and output.
Screen gain describes a direction, not total reflected light
Elite Screens’ terminology guide defines its gain comparison perpendicular to the screen centre against a NIST-traceable Lambertian white reference with gain 1.0. Its half-gain angle describes where brightness falls to half the reference-direction value. A published peak gain therefore does not establish the gain at a different seat or under a different illumination direction.
Gain can exceed one because a screen can direct more reflected light towards a particular direction than the reference does. It does not create extra light. It is also not interchangeable with diffuse luminous reflectance: the Diffuse Surface Luminance model uses a reflectance fraction from zero to one for an ideal Lambertian reflector. Putting a gain of 1.2 into that reflectance field would change the physical meaning of the calculation.
For the projection model, keep the material’s gain reference and the illumination/viewing geometry with the input. Applying one directional gain to the whole image is a simplifying assumption. Hotspots, angular gain changes across the image, curved screens, rear projection and selective ambient-light rejection require more detailed screen and geometry data.
Delivered lumens differ from a projector’s advertised output
Screen-delivered flux means the projector’s full-white luminous flux incident on the specified image rectangle in the stated operating condition. It is not bare-lamp output, colour-light output or an arbitrary marketing brightness unit. A rated projector figure cannot automatically be relabelled as screen-delivered lumens: image mode, lens, zoom, operating age and the actual optical path must match the evidence used to establish delivery.
No universal loss percentage is applied here. If output has already been measured at the screen, subtracting assumed optical losses again would count them twice. Likewise, a white reading containing ambient light would overstate the projector-only contribution. The Projector Throw Distance and Image Size tool can relate matching throw-ratio data to nominal dimensions, but its geometry supplies no missing output or screen-gain data.
cd/m² is the primary output. A supplemental foot-lambert value uses the NIST SI conversion table’s published rounded factor: 1 fL ≈ 3.426259 cd/m², so fL ≈ L / 3.426259. A foot-lambert is not one candela per square foot. This conversion changes the unit, not the screen prediction or its conditions.
A reverse calculation accepts an independently chosen target; it does not select a cinema, classroom or home viewing standard. Full-white luminance alone cannot determine perceived image contrast, black level, glare, HDR performance or suitability under ambient light. Those questions need measurements and criteria for the actual system and viewing environment.