Understanding Epsilon Delta Proofs
An epsilon delta proof turns a limit idea into a testable statement. The goal is simple. For every positive error band epsilon, find a positive input band delta. When x stays inside delta of a, the function value must stay inside epsilon of L. This calculator helps build that chain with clear bounds.
Why Bounds Matter
Most proofs fail because the bound is not controlled. A useful proof does not only test points. It shows why every nearby point works. Linear functions use a slope bound. Quadratics use a local factor around the chosen point. Polynomial and rational options use derivative style bounds. These bounds support a formal inequality path.
Construction Context
Construction work often uses tolerances, margins, and acceptance limits. Epsilon delta logic is similar. Epsilon is the allowed output tolerance. Delta is the allowed input tolerance. A drawing, cut length, slope, or material measurement may need controlled variation. The calculator can show how a small input change protects the final value.
Proof Workflow
Start with a function model. Enter the approach value a. Enter the claimed limit if you want to test it. Leave it blank to use the computed limit. Choose epsilon and a local radius. A smaller radius usually gives a stronger, safer proof. The safety factor shrinks the final delta, which adds a practical margin.
Interpreting Results
The result gives the computed limit, claim error, bound constant, raw delta, and final delta. If the claim does not match the computed limit, the proof may fail for small epsilon values. The warning explains this. Sample points are included only as checks. They do not replace the proof.
Exporting Work
Use the CSV option for spreadsheet review. Use the document export for a compact proof record. The exported data includes the main inputs, delta, formulas, and sample checks. Keep the local radius and assumptions with the proof, because they explain where the bound is valid.
Good Input Habits
Use realistic numbers and units. Avoid zero epsilon values. Check the denominator warning for rational forms. Increase sample counts when reviewing tight tolerances. Compare the final delta with field precision. If the needed delta is too small, improve measurement methods or relax the output requirement.