What is the distance formula?

The distance formula is d = √((x₂-x₁)² + (y₂-y₁)²). It calculates the straight-line (Euclidean) distance between two points in a plane. For 3D points, add (z₂-z₁)² inside the square root. Works in any dimension by summing squared differences.

Does point order matter?

No. Squaring eliminates sign, so (x₂-x₁)² = (x₁-x₂)². Distance from A to B equals distance from B to A. Just be consistent within the formula — same convention for both x and y differences.

How do I calculate distance in 3D?

Extend with z-coordinate: d = √((x₂-x₁)² + (y₂-y₁)² + (z₂-z₁)²). For (1,2,3) to (4,6,7): Δx=3, Δy=4, Δz=4. d = √(9+16+16) = √41 ≈ 6.40. Same principle works in any number of dimensions.

What's the difference between Euclidean and Manhattan distance?

Euclidean: straight-line distance, √(Δx²+Δy²). Manhattan: sum of absolute differences |Δx|+|Δy|. Same for axis-aligned movements; Manhattan ≥ Euclidean for diagonal. Manhattan used when restricted to grid (city streets, chess rook). Euclidean for unrestricted movement.

How is distance formula used in GPS?

Local positions: simple formula works. Long distances (continent-scale): must use spherical geometry (Haversine formula) accounting for Earth's curvature. GPS systems use 3D distance from multiple satellites to triangulate position (multilateration).

Can I use this formula on a map?

Yes, for short distances where map flatness is reasonable approximation. For coordinates in degrees of lat/long over long distances (>100 km), Earth's curvature matters significantly — use great-circle distance (Haversine formula) instead.

How is distance formula used in machine learning?

K-Nearest Neighbors classifies based on distances to neighbors. K-Means clustering groups points by closest center. Recommendation systems use similarity ∝ 1/distance. All compute distances in feature spaces (often high-dimensional). Choice of distance metric (Euclidean, Manhattan, cosine) affects results.

Distance Formula Calculator

Q: How is this related to the Pythagorean theorem?

The distance formula is derived directly from the Pythagorean theorem (a² + b² = c²). The horizontal difference (x₂-x₁) and vertical difference (y₂-y₁) form the legs of a right triangle; the distance is the hypotenuse c = √(a² + b²).

Find the straight-line distance between two points using the distance formula derived from the Pythagorean theorem. Enter the coordinates of both points to get the exact distance.

The distance formula gives the straight-line ("Euclidean") distance between two points in a coordinate plane. The formula d = √((x₂-x₁)² + (y₂-y₁)²) might look intimidating, but it's just the Pythagorean theorem applied to a triangle. The horizontal difference (x₂-x₁) and vertical difference (y₂-y₁) form the legs of a right triangle; the distance between the points is the hypotenuse.

This single formula underlies vast amounts of math, physics, computer graphics, navigation, and data analysis. GPS uses 3D distance to triangulate position. K-nearest-neighbors machine learning algorithms compute distances between points in high-dimensional feature spaces. Computer graphics calculate distances for collision detection, lighting, and rendering. Even basic geometric problems — finding the closest restaurant on a map, the length of a line segment, the closest point on a curve — all start with the distance formula.

The formula extends naturally to higher dimensions. In 3D: d = √((x₂-x₁)² + (y₂-y₁)² + (z₂-z₁)²). In n dimensions: d = √(Σ(pᵢ-qᵢ)²). Sometimes called Euclidean distance to distinguish from other distance measures (Manhattan, Chebyshev, etc.).

Common applications: geometry problems, GPS and navigation, computer graphics and game design, machine learning (KNN, clustering), physics (vector lengths), robotics (path planning), surveying (point-to-point measurements), and statistics (Euclidean norm of error vectors).

Inputs

Point 1: X

Point 1: Y

Point 2: X

Point 2: Y

Results

Distance

Horizontal Change (dx)

Vertical Change (dy)

Distance Squared

Last updated: May 29, 2026

Formula

**Distance formula (2D):** d = √((x₂ - x₁)² + (y₂ - y₁)²) Where (x₁, y₁) and (x₂, y₂) are the two points. **Distance formula (3D):** d = √((x₂ - x₁)² + (y₂ - y₁)² + (z₂ - z₁)²) **Distance formula (n-dimensional):** d = √(Σᵢ (pᵢ - qᵢ)²) **Worked example: 2D distance** Point A = (1, 2), Point B = (4, 6). Δx = 4 - 1 = 3 Δy = 6 - 2 = 4 d = √(3² + 4²) = √(9 + 16) = √25 = 5 Distance from (1,2) to (4,6) is exactly 5 units (forms 3-4-5 right triangle). **Special cases:** Same x-coordinate (vertical line): d = |y₂ - y₁| Same y-coordinate (horizontal line): d = |x₂ - x₁| Both reduce to absolute difference (no need for full formula). **Derivation (from Pythagorean theorem):** For right triangle with legs a, b, hypotenuse c: a² + b² = c² Applied to coordinate plane: - Leg 1 = horizontal difference = x₂ - x₁ - Leg 2 = vertical difference = y₂ - y₁ - Hypotenuse = distance d d² = (x₂-x₁)² + (y₂-y₁)² d = √((x₂-x₁)² + (y₂-y₁)²) **Common Pythagorean triples (distances are integers):** | Δx | Δy | Distance | |---|---|---| | 3 | 4 | 5 | | 5 | 12 | 13 | | 8 | 15 | 17 | | 7 | 24 | 25 | | 20 | 21 | 29 | | 9 | 40 | 41 | | 11 | 60 | 61 | | 6 | 8 | 10 | | 9 | 12 | 15 | Useful for "clean" distance problems with integer coordinates. **Manhattan vs Euclidean distance:** For points (1, 1) and (4, 5): - Euclidean: √(9 + 16) = 5. - Manhattan: |4-1| + |5-1| = 3 + 4 = 7. Manhattan = sum of absolute differences. Used when movement is restricted to grid (like Manhattan streets), in some ML algorithms. **Other distance metrics:** | Name | Formula | Use | |---|---|---| | Euclidean | √Σ(pᵢ-qᵢ)² | General geometric, physics | | Manhattan | Σ|pᵢ-qᵢ| | Grid paths, urban distance | | Chebyshev | max(|pᵢ-qᵢ|) | King moves on chessboard | | Minkowski | (Σ|pᵢ-qᵢ|^p)^(1/p) | General family (p=2 is Euclidean) | | Hamming | count of differences | Strings, binary | | Cosine | 1 - cos(θ) | Text similarity, embeddings | **Squared distance (for optimization):** d² = (x₂-x₁)² + (y₂-y₁)² When comparing distances, squared distance is often used (avoids square root, easier computation). The point with smallest distance is the same as the point with smallest squared distance. **Midpoint formula (related):** Midpoint M of segment from (x₁,y₁) to (x₂,y₂): M = ((x₁+x₂)/2, (y₁+y₂)/2) The midpoint is equidistant from both endpoints, at distance d/2 each. **Common applications:** | Application | Use of distance formula | |---|---| | GPS coordinates | Distance between locations | | Game development | Range checks, collision | | Computer graphics | Lighting (light-to-surface) | | Machine learning | KNN, K-means clustering | | Robotics | Path planning, sensors | | Physics | Vector magnitudes | | Surveying | Point-to-point | | Astronomy | Star/planet distances | | Sports | Field positions | **GPS distance:** For latitude-longitude coordinates, the simple formula doesn't quite work because Earth is curved. Use: - **Haversine formula**: great-circle distance on sphere. - **Vincenty formula**: more accurate for oblate Earth. For short distances (< 100 km), Euclidean approximation with degree-to-meter conversion gives ~1% accuracy. **Computer graphics:** - **Collision detection**: distance between objects vs sum of radii. - **Lighting**: brightness ∝ 1/d² (inverse square law). - **Bounding spheres**: object center + radius for quick culling. - **Path finding**: A* algorithm uses Euclidean or Manhattan as heuristic. **Machine learning:** - **K-Nearest Neighbors (KNN)**: classify by distance to neighbors. - **K-Means clustering**: assign points to nearest cluster center. - **Recommendation systems**: similarity = inverse distance. - **Anomaly detection**: distance from cluster centroid. **Computational complexity:** For each distance calculation: O(d) where d = number of dimensions. For all pairs in n points: O(n² × d). In high dimensions, "curse of dimensionality" makes distances less meaningful — most points become roughly equidistant. **Pythagoras in higher dimensions:** For 3D box with sides a, b, c: Space diagonal = √(a² + b² + c²) For 4D hypercube edge length 1: Hyper-diagonal = √4 = 2. **Software:** - **Python (NumPy)**: np.linalg.norm(p1 - p2) - **Python (SciPy)**: scipy.spatial.distance.euclidean(p1, p2) - **MATLAB**: norm(p1 - p2) - **JavaScript**: Math.sqrt(Math.pow(x2-x1, 2) + Math.pow(y2-y1, 2)) - **Excel**: =SQRT((B2-B1)^2 + (C2-C1)^2) **Common applications:** - **Construction**: rectangle diagonals, layouts. - **Real estate**: lot dimensions, building distances. - **Surveying**: point measurements with GPS or total station. - **Map reading**: distance "as the crow flies". - **Gaming**: player-to-enemy distance for AI behavior. - **Photography**: camera-to-subject distance. - **Navigation**: estimated travel time = distance/speed.

How to use this calculator

Enter x and y coordinates for first point.
Enter x and y coordinates for second point.
Calculator returns straight-line (Euclidean) distance.
Works in any units (meters, feet, etc.) as long as consistent.
For 3D: extend formula with z-coordinates (this calc is 2D).
For negative coordinates: use them directly (formula handles signs).

Worked examples

Map distance

**Scenario:** Two locations on a coordinate map. Point A at (10, 25), Point B at (40, 65). Distance? **Calculation:** Δx = 30, Δy = 40. d = √(900 + 1600) = √2500 = 50. **Result:** 50 units (whatever the map units are — km, miles, etc.). The 30-40-50 right triangle is a scaled 3-4-5 — always exact integer distance.

Robot arm reach

**Scenario:** Robot at origin (0, 0). Target at (12, 35) cm. Distance to target? **Calculation:** d = √(144 + 1225) = √1369 = 37 cm. **Result:** Target is 37 cm from robot base. If arm length is 40 cm, target is reachable. If 30 cm, not reachable. Distance check is the first step in any pick-and-place operation or motion planning.

Pixel distance in image

**Scenario:** Two pixels on screen: (100, 200) and (300, 400). Distance? **Calculation:** Δx = 200, Δy = 200. d = √(40000 + 40000) = √80000 ≈ 283 pixels. **Result:** ~283 pixels apart. Used in: collision detection (sprites), drag interaction, gesture recognition, image processing distance transforms.

When to use this calculator

**Use distance formula for:**

- **Geometric problems**: line segment lengths, triangle sides. - **GPS and navigation**: straight-line distances (with corrections for spherical Earth). - **Computer graphics**: object positioning, lighting. - **Game development**: player-enemy distances, ranges. - **Machine learning**: similarity measures, clustering. - **Robotics**: motion planning, sensor range. - **Surveying and mapping**: point-to-point distances. - **Physics**: vector magnitudes, force directions.

**Choosing distance metric:**

- **Euclidean**: when "as the crow flies" matters. - **Manhattan**: when constrained to grid movement. - **Chebyshev**: chess king moves, infinity norm. - **Cosine**: for direction similarity (ignores magnitude). - **Hamming**: for string/binary comparison.

**Higher dimensions:**

Formula extends naturally: - 2D: √(Δx² + Δy²). - 3D: √(Δx² + Δy² + Δz²). - n-D: √(Σ Δxᵢ²).

Used in ML for high-dimensional feature spaces.

**Avoiding square root (optimization):**

Comparing distances? Compare squared distances instead — avoids expensive square root operation. Particularly useful in: - Real-time game physics. - Sorting points by distance. - Anywhere only relative distances matter.

**Common pitfalls:**

- **Confusing signs**: Δx and Δy can be negative, but squared makes positive. - **Mixing units**: ensure all coordinates in same units. - **Curved surfaces**: don't use simple formula for Earth distances over long ranges. - **High dimensions**: curse of dimensionality affects intuition.

**GPS/lat-long conversion:**

For latitude (φ) and longitude (λ) on Earth (radius R): - Short distances (< 100 km): use linear approximation with degree-to-meter conversion. - Long distances: Haversine or Vincenty formula.

1° latitude ≈ 111 km always. 1° longitude varies: ~111 km at equator, decreasing toward poles by cos(latitude).

**Common applications:**

- **Real estate**: lot dimensions from corner GPS coordinates. - **Civil engineering**: surveying with total station + coordinates. - **Aerospace**: tracking aircraft positions. - **Astronomy**: angular separations of celestial objects. - **Photography**: focal distance to subject. - **Gaming**: AI proximity behaviors, weapon ranges. - **Visualization**: scatter plot distance interpretation. - **Audio**: 3D positional audio (distance-based volume).

**Software:**

- **Python (NumPy, SciPy)**: built-in distance functions. - **GIS systems**: ArcGIS, QGIS for geographic distances. - **Game engines**: Unity, Unreal have distance functions. - **CAD**: distance measurement tools. - **Spreadsheets**: manual formula or add-ins.

**Pitfalls:**

- **Forgetting to square the differences**: just adding (x₂-x₁) + (y₂-y₁) is wrong. - **Mixing point order**: doesn't matter for distance (squaring makes positive). - **Using simple formula for Earth**: doesn't account for curvature over long distances. - **Confusing distance with displacement**: distance is scalar; displacement is vector. - **High-dimensional intuition fails**: distances cluster oddly in many dimensions. - **Computer storage issues**: floating-point error for very large or very small coordinates.

Common mistakes to avoid

Forgetting to square the differences (just adding |Δx| + |Δy| gives Manhattan, not Euclidean).
Forgetting the square root at the end.
Mixing coordinate units.
Using flat distance for long Earth distances (need great-circle).
Confusing distance (scalar) with displacement (vector).
For high dimensions: assuming Euclidean intuition still works.
Calculating distance to wrong point (verify which is which).
Treating one coordinate system as another (e.g., lat-long as flat).

Frequently Asked Questions

Sources & further reading

SponsoredShop Top Deals on AmazonSupport CalcMountain — browse top-rated products at no extra cost to you.