📊 Condition Number Calculator

Calculate the condition number κ(A) = ||A|| × ||A⁻¹||

How to Use This Calculator

1

Enter Matrix

Input all elements of your 2×2 square matrix.

2

Calculate

Click to compute κ(A) = ||A|| × ||A⁻¹||.

3

Interpret Results

Small κ ≈ 1: well-conditioned. Large κ: ill-conditioned (near-singular).

Formula

κ(A) = ||A|| × ||A⁻¹||

Where ||·|| is a matrix norm (Frobenius, 2-norm, etc.)

Definition:

Condition number measures how sensitive the solution of Ax = b is to changes in A or b.

Bounds:

κ(A) ≥ 1 (always)

κ(A) = 1 if A is orthogonal/unitary

κ(A) = ∞ if A is singular

Interpretation:

  • κ < 10: Well-conditioned (stable)
  • 10 ≤ κ < 100: Moderately conditioned
  • 100 ≤ κ < 1000: Ill-conditioned
  • κ ≥ 1000: Very ill-conditioned (near-singular)

Error Amplification:

If input has relative error ε, output can have error up to κ × ε

About Condition Number Calculator

The Condition Number Calculator computes κ(A) = ||A|| × ||A⁻¹||, a measure of numerical stability. A large condition number indicates that small changes in input can cause large changes in output, making the matrix "ill-conditioned" and prone to numerical errors.

When to Use This Calculator

  • Numerical Analysis: Assess stability of matrix operations
  • Linear Systems: Predict error in solving Ax = b
  • Inverse Problems: Evaluate sensitivity to perturbations
  • Optimization: Check conditioning of Hessian matrices
  • Machine Learning: Assess numerical stability

Why Use Our Calculator?

  • Complete Analysis: Shows norm, inverse norm, and condition number
  • Classification: Categorizes matrix as well/ill-conditioned
  • Inverse Display: Shows A⁻¹ if it exists
  • Educational: Helps understand numerical stability
  • Accurate: Precise calculations
  • Free: No registration required

Key Concepts

  • Well-Conditioned: Small κ, stable solutions
  • Ill-Conditioned: Large κ, sensitive to perturbations
  • Singular: κ = ∞ (matrix is not invertible)
  • Error Bound: ||Δx||/||x|| ≤ κ × (||ΔA||/||A|| + ||Δb||/||b||)
  • Best Case: κ = 1 for orthogonal/unitary matrices

Applications

Linear Systems: If κ(A) is large, small errors in A or b cause large errors in solution x.

Least Squares: High condition number indicates unreliable solutions.

Frequently Asked Questions

What is condition number?

Condition number κ(A) = ||A|| × ||A⁻¹|| measures numerical stability. A large condition number means the matrix amplifies errors in calculations.

What does a large condition number mean?

A large condition number (κ > 100) indicates the matrix is "ill-conditioned" - small changes in input cause large changes in output, making it numerically unstable.

Can condition number be less than 1?

No, κ(A) ≥ 1 always. For orthogonal/unitary matrices, κ = 1 (best possible). The condition number equals 1 for identity matrices.

What if condition number is infinite?

If κ(A) = ∞, the matrix is singular (determinant = 0) and not invertible. This is the worst case for numerical stability.

Which norm is used?

This calculator uses the Frobenius norm: ||A|| = √(Σᵢⱼ aᵢⱼ²). Other common choices are the 2-norm (spectral norm) or 1-norm (column sum norm).