Error Function Calculator

Enter a value of x to get erf(x), erfc(x) = 1 − erf(x), and the equivalent standard normal probability and CDF Φ(x).

Quick Facts

Definition
erf(x) = (2/√π) ∫₀ˣ e⁻ᵗ² dt
Twice the area under the standard Gaussian curve from 0 to x, normalized so erf(∞) = 1.
Complementary function
erfc(x) = 1 − erf(x)
Gives the remaining tail area from x to infinity.
Symmetry & bounds
erf(−x) = −erf(x)
erf(0) = 0 and erf(x) stays between −1 and 1 for every real x.
Link to the normal distribution
Φ(z) = ½[1 + erf(z/√2)]
Connects erf to the standard normal CDF used throughout statistics.

Your Results

Calculated
erf(x)
-
Error function value
erfc(x)
-
Complementary: 1 − erf(x)
P(|Z| ≤ |x|)
-
Normal-distribution probability = erf(|x|/√2)
Φ(x)
-
Standard normal CDF = ½[1 + erf(x/√2)]

Ready

Enter a value for x, then press Calculate.

Frequently Asked Questions

What is the error function erf(x)?
The error function is defined as erf(x) = (2/√π) ∫₀ˣ e⁻ᵗ² dt. It measures the area under the bell-shaped Gaussian curve e⁻ᵗ² between 0 and x, scaled so that erf(x) approaches 1 as x → ∞ and −1 as x → −∞.
What is the difference between erf(x) and erfc(x)?
erfc(x), the complementary error function, is simply 1 − erf(x). While erf(x) gives the area under the Gaussian curve from 0 to x, erfc(x) gives the remaining tail area from x to infinity, which shows up often in diffusion and heat-transfer equations.
How does the error function relate to the normal distribution?
For a standard normal variable Z, the cumulative distribution function is Φ(z) = ½[1 + erf(z/√2)], and the probability that Z falls within a standard deviations of the mean is P(|Z| ≤ a) = erf(a/√2). This is why erf appears throughout statistics and confidence-interval calculations.
Can erf(x) be solved exactly, or is it always approximated?
erf(x) has no closed-form elementary antiderivative, so it is always evaluated numerically — from tables, a polynomial approximation, or, as this calculator does, numerical integration of the Gaussian curve using Simpson's rule, which is accurate to many decimal places for typical inputs.

Formula and Method for the Error Function

The error function erf(x) is defined as erf(x) = (2/√π) ∫₀ˣ e−t² dt — twice the area under the standard Gaussian curve e−t² between 0 and x, scaled so the total area from 0 to infinity equals 1. It has no elementary closed-form antiderivative, so every calculator — including this one — evaluates it numerically. This tool also reports the complementary error function erfc(x) = 1 − erf(x) and the equivalent standard normal distribution values, since erf is the mathematical backbone of the normal (Gaussian) probability distribution.

How the calculation works

Given x, the calculator numerically integrates e−t² from 0 to x using composite Simpson's rule with 2,000 subintervals, then multiplies by 2/√π to get erf(x). Because erf is an odd function, erf(−x) = −erf(x), so negative inputs are handled by integrating |x| and flipping the sign. From erf(x) the calculator derives erfc(x) = 1 − erf(x), the probability that a standard normal variable falls within |x| standard deviations of the mean, P(|Z| ≤ |x|) = erf(|x|/√2), and the standard normal cumulative distribution function Φ(x) = ½[1 + erf(x/√2)].

Common mistakes

  • Confusing erf(x) with Φ(x): the standard normal CDF is not erf(x) directly — it's Φ(x) = ½[1 + erf(x/√2)], which rescales the input by 1/√2.
  • Mixing up erf and erfc: erfc(x) = 1 − erf(x), not −erf(x); erfc(x) is always between 0 and 2.
  • Expecting values outside [−1, 1]: erf(x) is bounded between −1 and 1 for every real x — if a computed value falls outside that range, double-check the inputs.

Real-world applications

  • Statistics: converting z-scores to probabilities and building confidence intervals for normally distributed data.
  • Heat conduction and diffusion: the temperature profile in a semi-infinite solid after a sudden surface temperature change follows T(x,t) = Ts + (T₀ − Ts)·erf(x / (2√(αt))).
  • Semiconductor fabrication: dopant concentration from a constant-source diffusion follows C(x,t) = Cs·erfc(x / (2√(Dt))), from Fick's second law.
  • Digital communications: bit-error-rate for signals in Gaussian noise is often written as ½·erfc(√(Eb/N₀)).