Skip to content
Back to exploration
Calculus / Chinese

Why the cutoff N in a limit proof is not unique

Charles队长 · Bilibili · 1:13

Open original
READ & KEEP

The explanation, unpacked.

Reviewed learning material · Video analysis · English
Read the full overview

This video visually demonstrates the epsilon-N definition of sequence limits using Geometer's Sketchpad. It shows that for the same limit problem, different inequality scaling techniques yield different valid values for N.

Use the learning inspector for key ideas and moments, or open the reading tabs for the complete notes.

Chapters

0:00Introduction and Visualization0:17Method 1: Basic Algebraic Derivation0:32Method 2: Scaling Technique A0:48Method 3: Scaling Technique B1:04Conclusion

Learning script

Generated from the video's visuals and explanation; not verbatim speech.

A limit proof needs an effective cutoff, not a unique answer. Once all later terms are within ε of the limit, the requirement is met. The animation compares three ways of obtaining such a cutoff for the same sequence.

For n>√3 the error is 9/(n²−3). Solving the error inequality directly gives n>√(9/ε+39/ε+3). Choose an integer cutoff guaranteeing this condition.

For n≥3n\ge 3, n²−3≥n3\ge n²/2, so the error is bounded by 18/n18/n². Requiring n>√(18/ε18/ε), together with the stated range restriction, gives another valid cutoff.

For n≥3n\ge 3, the sharper bound n²−3≥2n3\ge 2n²/3 gives error at most 13.5/n13.5/n². Different bounds can therefore yield different integer cutoffs.

Different valid values of N do not conflict. Each must guarantee the error condition for every n>Nn>N, and any larger cutoff works too. Existence and control of all later terms matter, rather than uniqueness or minimality.

Knowledge cards

01

Non-uniqueness of N in Limit Definition

When proving sequence limits, the integer N satisfying the condition for a given ε>0\varepsilon > 0 is not unique. Any N large enough works. The video derives three distinct values (7, 10, 9) for the same example.

lim⁡n→∞an=A  ⟺  ∀ε>0,∃N∈N+,s.t. ∀n>N,∣an−A∣<ε\lim_{n\to\infty} a_n = A \iff \forall \varepsilon > 0, \exists N \in \mathbb{N}^+, \text{s.t. } \forall n > N, |a_n - A| < \varepsilon
02

Basic Solution Method

Directly solving the absolute value inequality algebraically to find the range of n. This is rigorous but can be computationally intensive compared to estimation methods. Use this equivalence only for n>√3.

∣f(n)−3∣=∣9n2−3∣<ε⇒n>9ε+3|f(n) - 3| = \left| \frac{9}{n^2 - 3} \right| < \varepsilon \Rightarrow n > \sqrt{\frac{9}{\varepsilon} + 3}
03

Scaling Method (Denominator Reduction)

Simplifies calculation by reducing the denominator to increase the fraction's value (e.g., replacing n2−3n^2 - 3 with n2/2n^2/2). This typically yields a larger, simpler N. This bound holds for n≥3n\ge 3; the cutoff must also satisfy that restriction.

n2−3⩾n22⇒∣f(n)−3∣⩽18n2<ε⇒n>18εn^2 - 3 \geqslant \frac{n^2}{2} \Rightarrow |f(n) - 3| \leqslant \frac{18}{n^2} < \varepsilon \Rightarrow n > \sqrt{\frac{18}{\varepsilon}}
04

Alternative Scaling Strategy

A compromise scaling approach where n2−3n^2 - 3 is bounded below by 2n2/32n^2/3. This results in a coefficient of 13.5, further illustrating the flexibility in finding N. This bound holds for n≥3n\ge 3; the cutoff must also satisfy that restriction.

n2−3⩾2n23⇒∣f(n)−3∣⩽13.5n2<ε⇒n>13.5εn^2 - 3 \geqslant \frac{2n^2}{3} \Rightarrow |f(n) - 3| \leqslant \frac{13.5}{n^2} < \varepsilon \Rightarrow n > \sqrt{\frac{13.5}{\varepsilon}}

Explore the knowledge in this video

Open video knowledge graph →

  • Limits ExplanationAt 0:00
    Why this connection?

    The same sequence-limit example admits cutoffs N=7N=7,10,9. The epsilon-N definition requires a sufficiently large cutoff rather than a unique smallest one. The supporting estimates retain their domains: direct equivalence for n>√3 and the displayed denominator bounds for n≥3n\ge 3.

Questions this video answers

Meet the concept

↗
Find a method

↗
Understand why

↗
Understand why

↗
Find a method

↗
Meet the concept

↗