Cubic Critical Points

What a critical point means

For a differentiable function, a critical point in this calculator means an x-value at which the first derivative is zero. The tangent line of the graph is horizontal there. A horizontal tangent is a local feature of the slope, not automatically a maximum or minimum. A curve may arrive at a flat point and continue in the same direction, or it may change from increasing to decreasing or the reverse. The page therefore returns stationary x-values and leaves the later interpretation explicit.

The word point can be misleading because this tool reports x-values rather than complete coordinate pairs. To obtain the corresponding y-value, substitute each returned x into the original cubic, including the constant d. That extra substitution is outside the handler because the requested output is the derivative's real roots. Keeping x-location and function height separate makes it clear what the calculator has established and what still requires another step.

• A reported x-value makes f-prime equal to zero.
• The derivative condition identifies a horizontal tangent.
• A stationary x-value is not automatically a maximum or minimum.
• The original cubic must be evaluated separately for a full coordinate.

The four coefficient fields

Enter ca as the coefficient of x^3, cb as the coefficient of x^2, cc as the coefficient of x, and cd as the constant. The labels are intentionally tied to powers so that a sign or position is not lost when a polynomial is copied from a worksheet. The default values describe f(x) = x^3 - 3x^2 + 5. The constant is not a fourth derivative coefficient; it is part of the original function only.

Each field must arrive at the pure handler as a finite JavaScript number between -1,000,000,000 and 1,000,000,000. The browser form advertises the same broad range, while the handler repeats the check for direct calls. Numeric strings, missing values, NaN, positive infinity, negative infinity, and out-of-range values are rejected. A valid number still needs a meaningful interpretation in the surrounding mathematics; the generic bound is not a statement that every cubic in that range is useful.

• ca multiplies x cubed and must be nonzero.
• cb multiplies x squared; cc multiplies x.
• cd shifts the original cubic vertically.
• All four inputs are finite and bounded numbers.

Why the leading coefficient cannot be zero

The formula is specifically for a genuine cubic. If ca is zero, the x^3 term disappears and the original expression is no longer cubic. The derivative would then have a different degree and a different interpretation. A zero leading coefficient is therefore rejected rather than silently downgraded to a quadratic. This protects the page from returning an answer that appears to belong to the requested model while actually solving another one.

A very small nonzero ca is allowed if it is within the field bounds and the resulting arithmetic stays finite. Small coefficients can make the derivative roots sensitive to the other coefficients because the quadratic coefficient A = 3ca becomes small. That sensitivity is a property of the entered polynomial, not a reason for the handler to replace the value with zero. Record the coefficients and retain enough precision when a small leading term matters.

• ca = 0 is a model mismatch, not a cubic critical-point case.
• The handler rejects zero before solving the derivative equation.
• Nonzero small coefficients remain valid but may be sensitive.
• Do not use this page to silently convert a lower-degree polynomial.

Differentiate the cubic term by term

The power rule gives the derivative f-prime(x) = 3ca x^2 + 2cb x + cc. The derivative of ca x^3 is 3ca x^2, the derivative of cb x^2 is 2cb x, the derivative of cc x is cc, and the derivative of the constant cd is zero. The calculator constructs A = 3ca, B = 2cb, and C = cc, then solves A x^2 + B x + C = 0. These names describe the derivative equation rather than replacing the original coefficient fields.

The constant disappears because adding the same vertical amount at every x does not change the slope. If two cubics have identical ca, cb, and cc but different cd, their graphs are vertical translations. Their y-values differ, but their derivative graphs coincide, so they have the same stationary x-values. This is a useful structural check for both hand calculations and software output.

• Derivative: f-prime = 3ca x^2 + 2cb x + cc.
• A = 3ca, B = 2cb, and C = cc in the solver.
• The constant derivative is zero.
• Only the x-location is solved by this page.

The discriminant chooses the real branch

For A x^2 + B x + C = 0, the discriminant is D = B^2 - 4AC. The sign of D determines how many real derivative roots exist. A positive D gives two distinct real roots, a zero D gives one repeated real root, and a negative D gives no real roots. The calculator checks D before taking its square root, so a negative discriminant becomes a clear explanatory error rather than a NaN result.

Because A, B, and C are built from the cubic coefficients, the page's displayed discriminant is a property of f-prime. It is not the discriminant of the original cubic equation f(x) = 0. A user looking for x-intercepts needs a different cubic-root problem. Here the discriminant answers a narrower question: can the slope equation equal zero at real x-values?

• D > 0 means two distinct real stationary x-values.
• D = 0 means one repeated stationary x-value.
• D < 0 means no real stationary x-value.
• The discriminant belongs to the quadratic derivative.

When there are no real critical points

A negative D means the quadratic derivative never reaches zero on the real axis. Since the leading derivative coefficient A is nonzero, the derivative keeps one sign throughout the real line. The original cubic is therefore strictly increasing or strictly decreasing under the ideal algebraic model, although the direction depends on the coefficient signs. The handler reports this condition with an error stating that there are no real critical points and does not manufacture complex roots.

No real critical point does not mean that the cubic has no roots, no inflection point, or no interesting behavior. A cubic can cross the x-axis while its slope remains positive, and every cubic has an inflection structure in the broad calculus sense. Those are different questions. The page only tests where the first derivative is zero, so use a root solver or a second-derivative analysis when another feature is needed.

• Negative D is a valid mathematical branch, not a failed calculation.
• The derivative has no real zero in this branch.
• A cubic can still cross the horizontal axis.
• Complex derivative roots are intentionally not displayed.

Two distinct stationary x-values

When D is positive, the derivative parabola crosses the x-axis twice. The two crossing locations are the two stationary x-values of the original cubic. The handler calculates both and sorts them from smaller to larger before returning them as Critical point 1 (smaller) and Critical point 2 (larger). Sorting makes the output stable even when the quadratic formula's plus and minus branches would naturally arrive in the opposite order.

The locations alone do not state whether the first is a local maximum or the second is a local minimum. For a cubic with a positive leading coefficient and two distinct critical points, the usual shape often rises, falls, and rises, but the precise classification should be checked from f-double-prime or from the sign of f-prime on intervals. The calculator's note points to that next analysis without pretending to perform it.

• Positive D produces two real derivative roots.
• Returned roots are sorted in ascending order.
• The labels describe x-order, not maximum/minimum type.
• Classification requires an additional derivative check.

The repeated-root branch

When D equals zero, the derivative quadratic touches the x-axis at one repeated root. The calculator returns Critical point (double) and the discriminant value zero. This is one stationary x-value, but the derivative does not change sign through a simple crossing. For the original cubic, the horizontal tangent is commonly associated with a stationary inflection: the curve can flatten while continuing from one side to the other.

The repeated label refers to the algebraic multiplicity of the derivative root, not to two separate physical locations. Do not count it twice when reporting stationary x-values. As always, classification should use the second derivative and the surrounding sign pattern if the distinction affects the conclusion. The page keeps this branch separate so that one repeated answer is not mistaken for two nearly equal numerical answers.

• D = 0 returns one location with multiplicity two.
• The output does not duplicate the same x-value.
• A repeated derivative root often signals a stationary inflection.
• Use local derivative signs for a definitive classification.

The default example step by step

The default inputs are ca = 1, cb = -3, cc = 0, and cd = 5. The original polynomial is f(x) = x^3 - 3x^2 + 5. Differentiating gives f-prime(x) = 3x^2 - 6x. In the solver notation, A = 3, B = -6, and C = 0. The discriminant is D = (-6)^2 - 4(3)(0) = 36, which is positive, so two real stationary x-values are expected.

The stable quadratic calculation gives x = 0 and x = 2 after sorting. Substitution into the derivative confirms f-prime(0) = 0 and f-prime(2) = 12 - 12 = 0. If full points are wanted, the original cubic gives f(0) = 5 and f(2) = 8 - 12 + 5 = 1, so the coordinates are (0, 5) and (2, 1). The page does not include those y-values because cd is not needed for the derivative roots.

• A = 3, B = -6, C = 0.
• D = 36, so two real roots are expected.
• The stationary x-values are 0 and 2.
• The original cubic gives heights 5 and 1 at those x-values.

Why changing the constant does not move them

Keep ca = 1, cb = -3, and cc = 0, but change cd from 5 to -100. The original curve moves down by 105 units, yet the derivative remains 3x^2 - 6x. The reported critical x-values remain 0 and 2. This is not a shortcut or an ignored input bug; it follows from the derivative of a constant being zero.

The constant still matters if the question changes from location to height. At x = 0, changing cd changes f(0) directly. At x = 2, it changes f(2) directly as well. If a report includes both a stationary location and a function value, retain the exact cd used for that report. A result that combines x-values from one constant with heights from another is internally inconsistent even though the derivative locations happen to match.

• Changing cd translates the graph vertically.
• The derivative and critical x-values stay unchanged.
• Function heights at those x-values do change.
• Keep cd in any separate coordinate calculation.

A numerically stable quadratic calculation

The familiar formula x = (-B plus or minus sqrt(D)) / (2A) is mathematically correct, but one branch can lose significant digits when -B and the square root have similar magnitudes. The handler uses q = -0.5 times (B plus a sign-aware square root), then obtains one root as q/A and the other as C/q. This arrangement reduces subtractive cancellation in many cases while preserving the same quadratic roots.

The output still uses ordinary finite JavaScript numbers. Stability is a safeguard, not an exact-arithmetic guarantee. Very large or badly scaled coefficients can make the last displayed digits sensitive to floating-point representation even when every input is within its declared range. The handler checks the discriminant and output values for finiteness, and the result formatter limits the display to six decimal places. Preserve raw inputs when more reproducibility is needed.

• q reduces cancellation in one quadratic branch.
• The second root is recovered from C/q.
• The displayed precision is six decimal places.
• Finite bounds do not turn floating-point arithmetic into exact arithmetic.

Signs, shape, and classification

The derivative sign tells whether the cubic is increasing or decreasing on an interval. A simple derivative root can mark a sign change, while a repeated derivative root can leave the sign unchanged. The first and second derivatives work together: f-double-prime(x) = 6ca x + 2cb provides a quick local classification when evaluated at a distinct critical x. A positive second derivative indicates local concavity up, and a negative value indicates local concavity down, subject to the usual calculus interpretation.

This page deliberately does not add classification labels because doing so would require returning function values, evaluating a second derivative, and explaining edge cases such as a zero second derivative. Those are useful extensions but not part of the current result contract. If a decision depends on a local maximum, minimum, or inflection, use the returned x-values as inputs to a separately checked step and show that method in the surrounding work.

• f-prime sign describes increase and decrease.
• f-double-prime can support local classification.
• A repeated root needs a sign-pattern check.
• The current outputs intentionally stop at stationary x-values.

Validation and failure behavior

Validation occurs inside the handler before any result is returned. A non-number, nonfinite number, or value outside the inclusive coefficient range produces a clear error naming the relevant field. A zero ca produces a cubic-specific nonzero error. A negative discriminant produces the no-real-critical-points message. This fail-closed behavior is preferable to clipping a coefficient or returning an empty list that could be mistaken for a successful calculation.

The browser also applies numeric field bounds, but a direct JavaScript caller cannot rely on HTML attributes. The pure engine repeats type, finiteness, and range checks so tests and other interfaces receive the same contract. If an error appears after copying a value from a spreadsheet, inspect whether the transfer produced a string, an unsupported notation, or a value outside the page's generic safety bound. Correct the source rather than expecting the handler to guess.

• Invalid types and nonfinite values are rejected.
• ca = 0 is rejected as not genuinely cubic.
• Negative D returns an explicit real-domain error.
• The handler does not coerce or silently clamp inputs.

How to report a result responsibly

A reproducible report should include the four coefficients, the derivative equation, the discriminant, and the returned branch. If two points are returned, state that they are ordered by x. If the repeated branch is returned, state its multiplicity. If no real points exist, preserve that result and explain that the derivative has no real zero rather than calling the calculation unsuccessful. These details let another reader recreate the branch decision without relying on a screenshot.

If the polynomial represents a measured or modeled quantity, add the units and describe what x means outside the calculator. The page knows only numbers; it cannot know whether x is time, distance, price, or an abstract variable. Do not turn a stationary point into an operational recommendation, an optimized setting, or a forecast without a model for the original quantity, domain restrictions, uncertainty, and consequences.

• Record coefficients, derivative, D, and branch.
• Keep x-units and y-units in the surrounding report.
• Evaluate the original cubic separately for y-values.
• A stationary point is mathematical evidence, not an automatic recommendation.

Scope and useful next steps

This calculator is useful for calculus practice, checking a derivative equation, locating possible turning behavior, and comparing how the first three coefficients shape a cubic's slope. It is also a compact test of discriminant branches and finite numeric handling. The result is strongest when the polynomial is explicitly written and the coefficient order is checked before calculation.

The page does not graph the function, solve the original cubic for x-intercepts, determine a global maximum or minimum on a restricted interval, propagate measurement uncertainty, or optimize a real system. A complete analysis may require domain constraints, endpoint comparisons, the second derivative, a graph, or an independent algebra system. Use the returned values as one transparent step in that larger analysis, and keep the model's assumptions visible.

• Use it to solve f-prime(x) = 0 for a bounded real cubic.
• Check the returned locations by substitution into the derivative.
• Add domain, graph, and second-derivative analysis when needed.
• Do not treat a finite answer as proof of a global optimum.

Substitution and independent checks

The quickest arithmetic check is to substitute each returned x-value into 3ca x^2 + 2cb x + cc. A result close to zero supports the displayed critical location within the chosen tolerance. For the default polynomial, substituting 0 gives zero and substituting 2 gives 12 - 12 = 0. If the displayed value has been rounded to six places, use the underlying calculation or a tolerance rather than demanding that a rounded decimal reproduce an exact zero digit for digit.

A second check is to calculate the vertex of the derivative parabola. Its x-coordinate is -B/(2A), where A = 3ca and B = 2cb. When D = 0, that vertex lies on the horizontal axis and matches the repeated critical point. When D is positive, it lies between the two returned roots. This geometric relationship can catch a coefficient-order mistake even when a copied quadratic formula appears syntactically correct.

A third check is to compare the sign of the derivative on intervals separated by the roots. Choose a test value below the smaller root, between the roots, and above the larger root, then evaluate f-prime. A sign change through a simple root indicates a change in increase or decrease; no sign change through a repeated root supports the stationary-inflection interpretation. These checks add interpretation without changing what the calculator itself reports.

• Substitute returned x-values into the derivative equation.
• Use the derivative-parabola vertex as a discriminant check.
• Test derivative signs on intervals around the roots.
• Use a tolerance that matches displayed rounding and input precision.