Units and uncertainties
A measurement is not a number: it is a number, a unit, and an uncertainty. CALCULA carries all three through the calculation, so you do not have to track them by hand.
Units
Write the unit after the value, in the cell: m = 2.5 kg, v = 12 m/s, T = 300 K. The calculation takes it into account — an energy computed from a mass in kilograms and a speed in metres per second comes out in joules, and a sum of a length and a mass is refused rather than silently wrong.
To convert without computing, “≡ Selection” ▸ Units offers the conversion of the quantity written in the cell.
Uncertainties
The ± sign (super keyboard, ∞≠∈ tab) attaches an uncertainty to a value:
Any expression using these symbols comes out with its uncertainty propagated. The propagation mode is chosen in ⚙ Settings:
- Linear — the classical propagation formula, first-order partial derivatives. Fast, and exact as long as the uncertainties stay small compared with the values.
- Monte Carlo — the quantity is drawn at random a large number of times (10,000 by default, adjustable) and the spread of the result is measured. Slower, but right even when the function is strongly curved or the uncertainty large.
The Steps panel details the contribution of each measurement: this is what tells you which of the two you would need to measure better to tighten the result.
Significant figures
“Significant figures” (⚙ Settings) does not change the calculation, only its display. The calculation is always done at full precision: rounding for readability therefore introduces no error into the rest of the document.
Try it: the “Error propagation (lab)” template
Simple pendulum, g = 4π²L/T² with L = 1.250 ± 0.002 m and T = 2.24 ± 0.03 s. The result comes out with its uncertainty; switch from Linear to Monte Carlo in the settings and compare: on this example the two agree, which is precisely the argument for using the fast method.
Where to go next
To fit a model to measurements that each carry their own uncertainty, see data and fitting — the “Regression with uncertainties” template gives the complete case.