Numerical Methods Lab
Root finding (Newton–Raphson, bisection, secant), ODE solvers (Euler, Runge–Kutta), and numerical integration & differentiation — each with a full iteration table and error estimates. Type a function, watch every step converge.
How to Use
- Pick a method, type the function — <code>f(x)</code> for root finding / integration / differentiation, or <code>dy/dx = f(x, y)</code> for the ODE solvers — and set the parameters that appear.
- Root finders need a start (Newton: x₀; secant: x₀, x₁; bisection: an interval [a, b] where f changes sign). The example <code>x³ − x − 2</code> with Newton from 1.5 converges to <strong>x ≈ 1.52138</strong>.
- Every run prints a full <strong>iteration table</strong> — each row is one step with all the intermediate values (e.g. xₙ, f(xₙ), f′(xₙ)) so you can see exactly how it converges.
- An <strong>error column</strong> tracks the step size or the gap to the exact value; integration compares Trapezoid, Midpoint and Simpson against the exact integral when one exists, and differentiation shows the central-difference error shrinking like h².
- Newton–Raphson uses the <em>exact</em> symbolic derivative of your function — no finite-difference approximation of f′.
Methods & formulas
About the Numerical Methods Lab
Meet the Numerical Methods Lab: a free, no-fuss tool for everyday maths and number work with nothing to install and no sign-up. Root finding (Newton–Raphson, bisection, secant), ODE solvers (Euler, Runge–Kutta), and numerical integration & differentiation — each with a full iteration table and error estimates. Type a function, watch every step converge.
How it works
Put each value in its box and read the answer as you go. Because it recalculates live, you can play with the inputs to see how each one moves the result — handy for checking your own working or planning ahead. Everything happens on your device, so it is fast and private.
Want the deeper story? The Knowledge Base explains the ideas behind the tools in more detail.
Frequently Asked Questions
Which methods are included?
<strong>Root finding:</strong> Newton–Raphson (uses the exact derivative), bisection (bracketing, guaranteed for a sign change), and the secant method (no derivative needed). <strong>ODE solvers:</strong> the Euler method and classical fourth-order Runge–Kutta (RK4). <strong>Quadrature & derivatives:</strong> numerical integration (trapezoid, midpoint, Simpson) and numerical differentiation (forward, backward, central differences, plus f″). Each one shows its iteration/convergence table and an error estimate.
How does Newton–Raphson work here?
It iterates xₙ₊₁ = xₙ − f(xₙ)/f′(xₙ). The tool computes f′ <em>symbolically</em> (the exact derivative), so each step is as accurate as possible and the table shows f(xₙ), f′(xₙ) and the update. Convergence is quadratic — the number of correct digits roughly doubles each step — which you can watch in the |Δx| column. If f′ hits zero the method stalls and the tool tells you to switch to bisection.
What's the difference between the root finders?
<strong>Bisection</strong> always works if f(a) and f(b) have opposite signs, but converges slowly (one bit per step). <strong>Newton</strong> is the fastest (quadratic) but needs a derivative and a good start, and can diverge. <strong>Secant</strong> is nearly as fast as Newton (superlinear) and needs no derivative — just two starting points. The lab lets you run the same equation through all three and compare the iteration counts.
Euler vs Runge–Kutta — which should I use?
Both step an initial-value problem dy/dx = f(x, y) forward in steps of h. <strong>Euler</strong> is the simplest (yₙ₊₁ = yₙ + h·f) but its error per step is large — O(h). <strong>RK4</strong> samples the slope four times per step and is dramatically more accurate — O(h⁴). For y′ = y, y(0) = 1 with h = 0.1, Euler reaches 2.594 at x = 1 while RK4 reaches 2.71828 — essentially e.
How is the error estimated?
For integration the tool compares Trapezoid, Midpoint and Simpson and, when an exact antiderivative exists, reports each method's true error. For differentiation it shows the central difference at a sequence of shrinking step sizes so you can see the O(h²) convergence (and the eventual round-off floor). For root finding the |Δx| (or half-interval) column is a direct error bound on the current estimate.
Does it run on a server?
No — parsing, the symbolic derivative, every iteration and the tables are all computed locally in your browser. Nothing is uploaded and it works offline.
How do I use the Numerical Methods Lab?
Just type your numbers. The answer shows up right away — there is no button to press. Change anything and it updates by itself.
Do I need to install or sign up for anything?
Not at all — it runs in the browser with nothing to install and no account. After it loads once, it even works without an internet connection.
Is my information private?
Yes. Everything happens in your browser. Nothing you type is sent to a server or saved anywhere.
Common Use Cases
Engineers & coders
Find roots, integrate and step ODEs the way a numerical library would — and see every iteration.
Numerical methods courses
Compare convergence rates of bisection, Newton and secant on the same equation.
Checking an implementation
Validate your own RK4 / Simpson code against a reference iteration table.
Solving f(x) = 0
Get a root to machine precision for any function you can type.
Initial-value problems
Euler and RK4 trajectories with the full step-by-step table.
Error analysis
Watch truncation error fall with step size and method order.
Last updated: