A linear camera response preserves proportional changes in incoming light over its usable range. A nonlinear response changes that relationship, so an increase in recorded value need not represent the same proportional increase in light. This distinction matters when interpreting relative brightness levels. It is separate from the timing question of when a particular exposure began or ended.
An obvious disappearance can remain recognisable even when the camera response is not linear. IOTA's analog-camera guidance says that nonlinear response at high intensities is not a problem for most occultation observations. This is a useful, limited statement. It does not establish the response of an individual configuration, nor does it make detailed brightness measurements independent of camera settings.
Locate the measurement question
Timing a strong drop asks where the signal changes relative to the exposure sequence. Measuring the relative depth of a smaller change asks how the recorded levels relate to the incoming light. The latter task puts greater weight on response information. Before testing, decide whether the intended analysis needs only a clear event boundary or also a trustworthy relationship between brightness levels.
For a lunar occultation involving an unresolved pair of stars, the recording may contain an intermediate level as the components disappear at different times. A faint component's contribution can be harder to judge than a complete disappearance. Wikipedia's light-curve article identifies stepped curves among the possible forms when double objects are involved. Exposure blending, background and noise also belong in the interpretation, so a step should not be treated as self-explanatory.
The page on reading a light curve introduces brightness plotted against time. A response test examines the brightness relationship while keeping time and illumination conditions controlled. A separate timing test remains necessary if the resulting curve will be used to assign precise disappearance or reappearance times.
Keep the configuration fixed
Use the camera, recorder and settings intended for the observation. Fix gain, integration and gamma; disable automatic adjustments where the equipment permits and where their role is understood. Record the complete configuration. A response curve measured under different processing settings may say little about the signal saved during the actual event.
Wikipedia describes gamma correction as a nonlinear brightness operation. A gamma setting can therefore change the response being tested even when the sensor and light source remain the same. Distinguish a display transformation from a transformation applied before storage. The test should use the saved values that will enter analysis, not an attractive preview whose processing is uncertain.
Exposure and recorder processing are equally important. Integration changes may alter the light gathered, and a recorder may apply adjustments while converting the signal. Treat the assembled recording chain as the device under test. Keep the unaltered trial recording so that later inspection can separate camera behaviour from a change introduced during export or analysis.
Use a stable scene and a known light change
A general response test records a steady, evenly illuminated scene while changing the incoming light in a controlled and independently understood way. The illumination needs to remain stable during each measurement. A source that flickers or drifts can make the camera appear nonlinear when the source is changing. Avoid relying on an untested lamp setting as if it were a calibrated brightness scale.
Keep framing, focus and the measured region fixed. Exclude reflections, moving shadows and bright areas that spread into the region being measured. Check for uneven illumination across the scene and choose an area whose behaviour can be described consistently. Record a dark or background reference in conditions relevant to the trial, and retain it with the illumination sequence.
Progress through a range that includes the signal levels expected in the observation, without driving the stored values against their limits. Revisit earlier illumination conditions to see whether the results remain consistent. Such repetitions can reveal source drift or an automatic adjustment that escaped the initial setup. They support the test only when the lighting and camera state remain documented.
Interpret the pattern cautiously
Compare the measured output with the independently established change in incoming light. A straight proportional relationship after accounting for the background is the intended linear pattern. Curvature, flattening or irregular jumps call for investigation. Do not force a curve through doubtful measurements simply to obtain a convenient correction.
Flattening at the bright end may indicate that the stored signal is approaching its limit. The dark end can be dominated by background and noise. Restrict any useful interpretation to the region actually supported by the trial. A correction derived from a narrow range should not be extended to brighter or fainter signals without evidence.
Changing a setting to obtain a different response also changes the configuration that must be documented and checked. Computer camera controls make deliberate settings easier to manage, while camera selection places response beside sensitivity and exposure behaviour. Keep the trial recording and describe the usable range in plain terms.
A response test cannot make a poor signal or uncertain time reference precise. It answers a particular brightness question under stated conditions. Use it with the exposure information and the original event recording, retain ambiguous results, and report the limits that matter to the measurement. The useful outcome is an understood configuration, not a universal label declaring a camera suitable for every occultation.