Sensitometry workbench

Import measured density data, choose a fitting interval, inspect a curve and export its readings.

Lower log exposure for fitted intervalUpper log exposure for fitted intervalMeasured base plus fog density

Measurements — logH,density; sample below is synthetic arithmetic datalogH,density 0,0.1 0.5,0.4 1,0.7 1.5,1.0 2,1.3Or load your CSV (maximum 200 KB)CalculateExport these readings

Enter your values and calculate. Read the assumptions below before using a result.

Method and assumptions

Supply two columns: logH,density. The optional header uses those names. The tool sorts the readings by exposure and rejects duplicate exposure coordinates: decide whether repeats should be averaged, or analyse them separately, before importing. The example readings are synthetic arithmetic data and do not represent a photographic material.

Select the interval deliberately. The tool fits a straight line by least squares only to readings within that interval, then reports its slope, intercept and root-mean-square residual. A small residual describes agreement with a line over those points. It does not establish instrument calibration, correct exposure geometry or a straight-line region outside the selected interval.

For the example, density rises by 0.6 for each unit of log exposure. Fit 0 to 2 and recover that slope. Move the boundaries so fewer points remain, then consider why a two-point fit always has zero residual even when the measurements have uncertainty. Residual scatter and uncertainty of measurement are different quantities.

The base-plus-0.10 crossing is linearly interpolated only when one upward crossing is established. It is left unset when the supplied readings do not establish it uniquely. A relative exposure axis remains relative: this tool does not award an ISO speed or call an arbitrary fitted interval H-740 contrast index. Save the raw readings, instrument geometry, calibration, exposure units, fitting interval and any exclusions alongside the graph. Export preserves the numeric readings so another person can repeat the calculation.

Evidence and record

Read the teaching method and its sources. Keep your inputs, the displayed assumptions and the result together; a result detached from its conditions cannot be checked.