This page is the exercise gym for parametric differentiation. The parent note built the two formulas; here we drive them through every kind of situation a problem can present — good slopes, vertical tangents, degenerate points, negative signs, a real-world word problem, and an exam twist.
Recall The two tools we keep reusing (from the parent)
First derivative — the speed ratio (vertical speed divided by horizontal speed):
dxdy=x˙y˙(x˙=0)
Second derivative — quotient rule plus one extrax˙ underneath:
dx2d2y=x˙3y¨x˙−y˙x¨
Here a dot means "rate of change with respect to the parameter t": x˙=dx/dt, x¨=d2x/dt2, and likewise for y. If any of that feels new, re-read the parent — this page uses it, it doesn't re-derive it. The engine underneath is the Chain rule and the Quotient rule. Throughout this page, we call these two displayed equations the first-derivative formula and the second-derivative formula.
Every parametric-differentiation problem falls into one of these cells. Our worked examples below are labelled with the cell they hit, so by the end no cell is empty.
#
Cell (case class)
What makes it tricky
Example
A
Plain "nice" point, x˙=0
none — warm-up
Ex 1
B
x˙=0 → vertical tangent
slope undefined, not an error
Ex 2
C
y˙=0 → horizontal tangent
slope exactly 0
Ex 3
D
Sign of d2y/dx2 → concavity
up vs down, and sign of x˙ matters
Ex 4
E
Degenerate point: x˙=y˙=0 (cusp)
0/0, need a limit
Ex 5
F
Real-world word problem (projectile)
units, choose the right t
Ex 6
G
Exam twist: find t giving a required slope
solve backwards
Ex 7
H
Limiting behaviour as t→ special value
asymptotic slope
Ex 8
Each cell shows a different way the same two formulas behave. Watch especially the sign of x˙ in cells B, D, H — it is the quiet troublemaker.
The figure below shows the unit circle in lavender. The red dashed lines are the vertical tangents at the leftmost and rightmost points, with red arrows pointing straight up to stress that the motion there is purely vertical; the mint dotted lines mark the horizontal tangents (top and bottom) for contrast. As you read Example 2, find the two red points (±1,0) on the picture.
Figure 1 — Unit circle x=cost,y=sint. Red dashed = vertical tangents where x˙=0 (points (±1,0)); mint dotted = horizontal tangents where y˙=0 (points (0,±1)).
The next figure shows the curve x=t2,y=t3. The lavender branch is traced for t≤0, the coral branch for t≥0; both meet at the slate dot at the origin, forming a sharp beak (cusp). The mint arrow shows the common horizontal tangent direction. As you read Example 5, notice that although the tangent is horizontal, the two branches leave the origin on the same line — that is what makes it a cusp rather than a smooth minimum.
Figure 2 — The cusp of x=t2,y=t3 at the origin. Both branches (t≤0 lavender, t≥0 coral) meet with a horizontal limiting tangent (mint), yet the point is not smooth.
The final figure shows one arch of the cycloid in lavender, with the cusp at the origin. Three short tangent segments are drawn near the cusp — at t=0.6 (mint), t=0.3 (butter) and t=0.12 (coral). Watch how they get steeper as t shrinks toward 0: this is the visual meaning of the slope running off to +∞ in Example 8.
Figure 3 — Cycloid x=t−sint,y=1−cost. Tangent segments at t=0.6,0.3,0.12 steepen toward vertical as the cusp at the origin is approached.
The flowchart below is a decision tree you can walk for any parametric point. Start at the top with your x(t),y(t). First ask is x˙ zero? If yes, you have a vertical tangent (Cell B). If no, ask is y˙ zero? — yes gives a horizontal tangent of slope 0 (Cell C). If neither is zero, ask are they both zero at once? — yes means a cusp needing a limit (Cell E, and its limiting-slope cousin Cell H); no means an ordinary point where you simply compute y˙/x˙ (Cell A). From that ordinary slope you can then branch to checking concavity (Cell D), interpreting real-world velocities (Cell F), or solving backwards for a required slope (Cell G). In words: zeros first, ordinary slope second, then everything else.