ALGEBRAAlgebraMathematics Calculator
🔬

Numerical Root Finding

Newton: x_{n+1}=x_n−f(x_n)/f′(x_n) — quadratic convergence. Bisection: halve interval where f(a)f(b)<0. Secant: uses two points, no derivative. Quadratic formula for degree 2.

Concept Fundamentals
x_{n+1}=x_n−f/f′
Newton
c=(a+b)/2
Bisection
No derivative
Secant
Exact for deg 2
Quadratic

Did our AI summary help? Let us know.

Newton's method has quadratic convergence near a simple root. Bisection always converges but linearly — halves interval each step. Secant avoids derivative but may not converge if initial guess is poor.

Key quantities
x_{n+1}=x_n−f/f′
Newton
Key relation
c=(a+b)/2
Bisection
Key relation
No derivative
Secant
Key relation
Exact for deg 2
Quadratic
Key relation

Ready to run the numbers?

Why: Many polynomials have no closed-form roots (degree ≥5). Numerical methods find approximate roots. Newton converges fast near root. Bisection is robust. Used in engineering and scientific computing.

How: Newton: need f and f′. Iterate until |f(x)|<tol. Bisection: bracket root with f(a)f(b)<0, halve interval. Secant: use two previous points, no derivative. Quadratic: use formula for degree 2.

Newton's method has quadratic convergence near a simple root.Bisection always converges but linearly — halves interval each step.

Run the calculator when you are ready.

Find Roots NumericallyNewton, bisection, secant

Real-World Scenarios — Click to Load

e.g. 1,0,-2 for x²-2
polynomial_root_finder
CALCULATED
Polynomial
x^2 -2
Roots found
1
Iterations
4
Convergence
Quadratic (Newton)
Roots
1.414214
Share:

Root Values (real)

Iterations per Root

Calculation Steps

Polynomialx^2 -2
Methodnewton
Roots found1
Total iterations4
ConvergenceConverged

For educational and informational purposes only. Verify with a qualified professional.

🧮 Fascinating Math Facts

Newton: x²−2=0, start x=1.5 → 1.4167 → 1.4142 (√2).

— Example

📐

Bisection: if f(1)<0 and f(2)>0, root in (1,2). Midpoint 1.5.

— Method

Key Takeaways

  • Newton's method converges quadratically near a root but needs a good initial guess and derivative.
  • Bisection always converges (if sign change exists) but is slower (linear). No derivative needed.
  • Secant method approximates the derivative; superlinear convergence, no derivative formula.
  • Analytical formulas exist only for degree ≤ 4. Degree 5+ requires numerical methods (Abel-Ruffini).
  • Convergence criteria: |xₙ₊₁ − xₙ| < ε or |f(xₙ)| < ε. Max iterations prevent infinite loops.

Did You Know?

📐x² − 2 = 0 gives ±√2. The Greeks proved √2 is irrational using geometry.Source: History
🔢Wilkinson's polynomial has roots 1,2,...,20 but tiny coefficient errors make roots very sensitive.Source: Numerical
📊Newton's method can diverge or cycle with a bad guess. Bisection is robust but slower.Source: Convergence
💹IRR in finance: find r where NPV = Σ CFₜ/(1+r)ᵗ = 0. A root-finding problem!Source: Finance
⚙️Control systems: poles of transfer functions are roots of the characteristic polynomial.Source: Engineering
📉Abel-Ruffini: no general algebraic formula for roots of degree ≥ 5 polynomials.Source: Algebra

How It Works

Numerical root-finding iteratively refines an approximation. Newton: xₙ₊₁ = xₙ − f(xₙ)/f′(xₙ). Bisection: halve the interval [a,b] where f(a)f(b) < 0. Secant: replace f′ with (f(xₙ)−f(xₙ₋₁))/(xₙ−xₙ₋₁). For polynomials, deflation removes found roots to find the next. Analytical: degree 1–2 use direct formulas; degree 3–4 use Cardano/Ferrari (complex).

Expert Tips

Good initial guess

Newton and secant need a guess near the root. Plot the polynomial first.

Bisection for robustness

If Newton fails or oscillates, try bisection with an interval [a,b] where f(a)f(b)<0.

Tolerance trade-off

Smaller tolerance = more iterations. 1e-8 is usually sufficient for most applications.

Multiple roots

Use deflation or specialized methods. Newton converges slowly for multiple roots.

Method Comparison Table

MethodConvergenceDerivativeRobustness
NewtonQuadraticRequiredSensitive to guess
BisectionLinearNot neededVery robust
SecantSuperlinearApproximatedModerate

Frequently Asked Questions

When does Newton fail?

Newton can diverge with a bad guess, or cycle. It also converges slowly for multiple roots (f'(x)=0 at root).

Why use bisection?

Bisection always converges if f(a)f(b)<0. No derivative, no guess—just an interval. Slower but reliable.

What is deflation?

After finding a root r, divide the polynomial by (x−r) to get a lower-degree polynomial for the next root.

Can we find complex roots?

Yes, but Newton/bisection on the reals only find real roots. Use complex initial guess or specialized methods.

What is Wilkinson's polynomial?

∏(x−k) for k=1..20. Roots are 1,2,...,20. Tiny coefficient errors cause huge root errors—ill-conditioned.

When do analytical formulas fail?

Degree 5+ have no general formula (Abel-Ruffini). Degree 3–4 formulas exist but are complex; numerical often easier.

Quick Reference

xₙ₊₁=xₙ−f/f′
Newton update
c=(a+b)/2
Bisection midpoint
2ⁿ
Bisection halvings
n≥5
No formula (Abel)

Disclaimer: Numerical methods give approximate roots. For ill-conditioned polynomials (e.g., Wilkinson), results may be sensitive to tolerance and coefficient precision.

👈 START HERE
⬅️Jump in and explore the concept!
AI

Related Calculators