How Do You Find Altitude Of A Triangle

You’re staring at a triangle on a page — maybe it’s a homework problem, maybe it’s a roof truss you’re trying to size for a shed — and the question asks for the altitude. The word sounds technical. The diagram looks simple enough. But the moment you try to do it, the numbers don’t line up the way you expect.

Been there. Worth adding: the altitude of a triangle is one of those concepts that seems obvious until you actually need to calculate it for a scalene triangle with messy side lengths. Then it gets sticky fast.

What Is an Altitude of a Triangle

At its core, an altitude is just a fancy name for height. But it’s a very specific kind of height.

Draw a line from one vertex (corner) straight down to the opposite side — or the extension* of that opposite side — so that the line meets the base at a perfect 90-degree angle. In real terms, that segment? And that’s the altitude. The side it hits is the base. Every triangle has three altitudes, one from each vertex, and they all intersect at a single point called the orthocenter.

Here’s where people trip up: the altitude isn’t always inside* the triangle. In an acute triangle, sure, all three altitudes sit neatly inside. Day to day, in a right triangle, two of the altitudes are just the legs themselves — easy. But in an obtuse triangle? That said, two of the altitudes fall completely outside the shape. You have to extend the base line out past the vertex to drop that perpendicular. If you don’t realize that, you’ll measure the wrong segment every time.

Altitude vs. Median vs. Angle Bisector

Three lines. Also, three different jobs. The median connects a vertex to the midpoint* of the opposite side. The angle bisector splits the angle in half. Plus, it only cares about the right angle. The altitude? Don’t assume they’re the same. They coincide only in special cases — like an equilateral triangle, where every line from a vertex does all three jobs at once. They’re not.

Why Finding the Altitude Actually Matters

You might wonder why anyone bothers with this outside of a geometry classroom. Short answer: area.

The standard area formula — one-half base times height — requires* the altitude. Now, you need the perpendicular height. So if you only know side lengths, you can’t plug them directly into that formula. That’s the bridge between side-length data and area.

It shows up in structural engineering, too. Roof pitches, bridge trusses, the slope of a ramp — they all rely on that perpendicular drop. In coordinate geometry and computer graphics, altitude calculations determine collision detection, shading normals, and shortest-path distances. Surveyors use it when they can’t measure straight across a property but can measure angles and sides from accessible points.

Bottom line: if you can find the altitude, you access the area. And area unlocks material estimates, load calculations, and a whole lot of practical math.

How to Find the Altitude — Every Way That Works

There isn’t one single method. The right approach depends entirely on what you’re given. Let’s walk through the scenarios you’ll actually encounter.

Using the Area Formula (When You Already Know the Area)

This is the most direct route. If a problem hands you the area and the base length, rearrange the formula:

Altitude = (2 × Area) / Base

That’s it. Divide double the area by the base length. The units work out cleanly — if area is in square meters and base is in meters, altitude comes out in meters.

Example: A triangle has an area of 36 square centimeters and a base of 9 cm.
Altitude = (2 × 36) / 9 = 72 / 9 = 8 cm.

Done. But most real problems don’t give you the area upfront. They give you sides. That’s where it gets interesting.

Right Triangles — The Legs Are Already Altitudes

Right triangles are the gift that keeps giving. The two legs are altitudes relative to each other. If the legs are a and b, then the altitude to leg a is just b, and vice versa.

But what about the altitude to the hypotenuse*? That one isn’t a side. You’ll need the geometric mean relationship (sometimes called the right triangle altitude theorem):

Altitude to hypotenuse = (leg₁ × leg₂) / hypotenuse

Or, if you know the two segments the altitude creates on the hypotenuse (call them p and q), the altitude is the square root of p × q*.

Let’s say a right triangle has legs 6 and 8. Altitude to hypotenuse = (6 × 8) / 10 = 48 / 10 = 4.Think about it: 8. Hypotenuse is 10 (classic 3-4-5 scaled by 2). Quick. Clean.

Equilateral and Isosceles Triangles — Symmetry Is Your Friend

Equilateral triangles are the easiest non-right case. Which means all angles 60°. On top of that, all sides equal s. That said, the altitude splits the triangle into two 30-60-90 right triangles. The short leg is s/2, the long leg (the altitude) is s√3/2*.

Altitude = (side × √3) / 2

For a generic isosceles triangle with equal sides a and base b, the altitude to the base bisects it. You get two right triangles with hypotenuse a and one leg b/2. Pythagoras does the rest:

Altitude to base = √(a² − (b/2)²)

If you need the altitude to one of the equal legs instead, use the area bridge: compute the area via the base altitude above, then rearrange Area = ½ × leg × altitude_to_leg*.

General Triangles (Scalene) — When Nothing Is Special

No right angles. No equal sides. You have three side lengths — a, b, c — and need an altitude. The standard path goes through Heron’s formula.

1. Semiperimeter: s = (a + b + c) / 2*
2. Area: A = √[s(s − a)(s − b)(s − c)]*
3. Altitude to side a: hₐ = 2A / a*

Repeat for h_b and h_c by swapping the base in step 3.

Example: Sides 13, 14, 15.
s = 21.
A = √[21(8)(7)(6)] = √7056 = 84.
Altitude to side 14: h = (2 × 84) / 14 = 12.
(Notice the 13-14-15 triangle splits into two Pythagorean triples: 5-12-13 and 9-12-15. The altitude is the shared leg.)

If You Have Two Sides and the Included Angle (SAS)

Skip Heron. Use the trigonometric area formula directly:

Area = ½ × a × b × sin(C)

Then altitude to side a is:

hₐ = b × sin(C)

Basically faster and avoids the square-root nesting of Heron. It’s also the native method in surveying and physics when angles come from theodolites or force vectors.

If You Have Coordinates — The Analytic Shortcut

Vertices at (x₁, y₁), (x₂, y₂), (x₃, y₃). You want the altitude from vertex 1 to side 2-3.On the flip side, 1. Length of base (side 2-3):
b = √[(x₃ − x₂)² + (y₃ − y₂)²]* 2. Area via Shoelace (determinant) formula:
A = ½ | x₁(y₂ − y₃) + x₂(y₃ − y₁) + x₃(y₁ − y₂) |* 3.

For more on this topic, read our article on what is 4/5 as a decimal or check out this table shows how many male and female.

Even faster: Use the point-to-line distance formula. The line through points 2 and 3 has equation:
(y₃ − y₂)x − (x₃ − x₂)y + (x₃y₂ − x₂y₃) = 0

Plug vertex 1 into the normalized distance formula:

h = | (y₃ − y₂)x₁ − (x₃ − x₂)y₁ + (x₃y₂ − x₂y₃) | / √[(y₃ − y₂)² + (x₃ − x₂)²]

One step. No intermediate area variable. This is the workhorse in CAD, GIS, and game engines.

In 3D or Vector Form — Cross Product Power

Given points A, B, C as position vectors. The altitude from A to line BC:

h = | (B − A) × (C − A) | / | B − C |

The numerator is the magnitude of the cross product (twice the triangle’s area). Now, the denominator is the base length. Works in any dimension ≥ 2, handles floating-point robustness better than coordinate algebra, and extends instantly to tetrahedron altitudes in 3D.

When Altitudes Misbehave: Obtuse Triangles

In an obtuse triangle, two altitudes fall outside* the triangle. The foot of the perpendicular lands on the extension* of the opposite side, not the segment itself.

The formulas above still work. But if you’re constructing the altitude physically (paper, CAD, layout), you must extend the base line. Area, Heron, coordinates, vectors — they all return positive lengths. The geometric mean relations in right triangles?

The geometric mean relations in right triangles? Worth adding: in an obtuse triangle, the altitude from the acute angles intersects the extensions* of the opposite sides, so the “segments of the base” become signed distances—one positive, one negative. They assume the foot lies on the hypotenuse. The relation $h^2 = |p \cdot q|$ still holds if you treat the segments as directed lengths, but the classic Euclidean construction (semicircle on the hypotenuse) no longer applies directly.

This distinction matters in computational geometry. A naive point-in-triangle test using barycentric coordinates or “same-side” checks will fail if it assumes all altitudes intersect the interior. dependable code checks the sign* of the dot products $(\mathbf{B} - \mathbf{A}) \cdot (\mathbf{C} - \mathbf{A})$ etc., to classify the triangle as acute, right, or obtuse before deciding which altitude foot lies on the segment versus the line.

The Orthocenter: Where the Three Lines Meet

The three altitudes (extended as lines) are always concurrent. Their intersection is the orthocenter $H$.

• Acute triangle: $H$ lies inside.
• Right triangle: $H$ is the vertex of the right angle.
• Obtuse triangle: $H$ lies outside, on the same side of the longest side as the obtuse angle.

The orthocenter has elegant vector and coordinate expressions. If the vertices are vectors $\mathbf{a}, \mathbf{b}, \mathbf{c}$ and the circumcenter is $\mathbf{O}$, then: $\mathbf{H} = \mathbf{a} + \mathbf{b} + \mathbf{c} - 2\mathbf{O}$ In trilinear coordinates (relative to side lengths $a, b, c$): $H = \sec A : \sec B : \sec C$. Here's the thing — in barycentrics: $\tan A : \tan B : \tan C$. These vanish or go infinite for right angles, correctly placing $H$ at the vertex.

The Euler Line connects the Orthocenter $H$, Centroid $G$, and Circumcenter $O$ with $HG = 2GO$. On the flip side, the Nine-Point Circle (center $N$, midpoint of $HO$) passes through the three altitude feet, the three midpoints, and the three midpoints from vertices to $H$. The altitudes are the radical axes of the nine-point circle with the three circles having the sides as diameters.

Pedal Triangles and the Orthic Triangle

The feet of the three altitudes form the orthic triangle (or pedal triangle of $H$). Worth adding: it is the inscribed triangle of minimum perimeter* (Fagnano’s problem). Its vertices are the points where the incircle of the orthic triangle touches the sides of the original triangle—meaning the altitudes of the original triangle are the angle bisectors* of the orthic triangle.

The sides of the orthic triangle are $a \cos A, b \cos B, c \cos C$. Its area is $2\Delta \cos A \cos B \cos C$. In an obtuse triangle, the orthic triangle has two vertices outside the original figure, and its “area” becomes a signed quantity.

Numerical Stability: A Practitioner’s Warning

When sides are nearly equal or angles approach $0^\circ$ or $180^\circ$, altitude calculations suffer catastrophic cancellation.

• Heron’s formula loses precision for needle-like triangles ($s \approx a$). Use Kahan’s variant: sort sides $a \ge b \ge c$, then $A = \frac{1}{4}\sqrt{(a+(b+c))(c-(a-b))(c+(a-b))(a+(b-c))}$.
• Sine formula ($h = b \sin C$) loses precision for tiny angles ($\sin \epsilon \approx \epsilon$). Use sinpi or sinl if available, or the versine: $\sin \theta = 2 \sin(\theta/2) \cos(\theta/2)$.
• Cross product in 3D is the gold standard. $|(\mathbf{B}-\mathbf{A}) \times (\mathbf{C}-\mathbf{A})|$ computes $2A$ directly without subtracting large, nearly equal numbers. Divide by $|\mathbf{B}-\mathbf{C}|$ last.

In floating-point code, always compute the altitude via the cross product magnitude divided by base length if coordinates or vectors are available. It is dimension-agnostic, branch-free, and the most numerically stable path.

Summary: Choose Your Weapon

Conclusion

The humble altitude is far more than a simple geometric construction—it is a bridge between algebraic computation and deep geometric structure. From its role in defining the orthocenter and the Euler line, to its connection with the nine-point circle and the elegant solution to Fagnana’s problem via the orthic triangle, the altitude reveals layers of mathematical richness. Yet, in practical computation, it demands care: naive implementations fail silently in edge cases, while thoughtful choices—like leveraging cross products in vector form or Kahan’s refinement of Heron’s formula—can preserve accuracy even in the face of numerical adversity.

For practitioners, the key takeaway is this: always match your method to your data type and precision requirements. Which means whether working with symbolic expressions, Cartesian coordinates, or high-dimensional vectors, understanding both the theoretical underpinnings and computational pitfalls of altitude calculation ensures dependable, reliable results. The altitude, properly computed, is not just a line—it is a foundation.