1. Normalize the input
Coefficients are read in descending powers. Common factors do not change the zeros, so the equation may be scaled before solving.
Get all roots counted with multiplicity, exact rational values when detected, method steps, real/complex counts, and a numerical substitution check.
3 roots found
A zero or root is a value r for which P(r)=0. The corresponding factor is (x−r). This calculator combines exact algebra with numerical approximation so it can cover linear through degree-10 polynomials.
Coefficients are read in descending powers. Common factors do not change the zeros, so the equation may be scaled before solving.
When coefficients can be represented as manageable integers, the Rational Root Theorem supplies candidates ±p/q. Exact hits are removed by synthetic division.
A remaining linear factor is solved directly. A quadratic factor uses x = (−b ± √(b²−4ac))/(2a), including complex results.
An Aberth-style simultaneous complex iteration finds any remaining roots. Newton polishing improves each result, and substitution measures the scaled residual.
P(x) = an(x − r1)(x − r2) ··· (x − rn)
The Fundamental Theorem of Algebra guarantees n complex roots for degree n, counted with multiplicity. If all coefficients are real, every non-real root a+bi is accompanied by its conjugate a−bi.
For x³−6x²+11x−6, candidate 1 gives zero. Synthetic division leaves x²−5x+6, which factors as (x−2)(x−3). The roots are 1, 2, and 3.
x³+1=(x+1)(x²−x+1). The real root is −1; the quadratic gives 0.5 ± 0.8660254i.
x⁴−4x³+6x²−4x+1=(x−1)⁴. The value 1 is one distinct zero with multiplicity 4.
The roots of x⁴+1 are approximately ±0.7071068 ± 0.7071068i. All four are non-real and occur in conjugate pairs.
All work runs in this page. Coefficients are never uploaded or stored. Rational-root tests use exact integer arithmetic when the entered fractions can be normalized safely; other roots are numerical approximations shown to up to 10 significant digits.
The calculator accepts degree 1–10, limits each coefficient to 14 digits and an absolute value of 1012, and rejects coefficient scaling that would make the monic polynomial numerically unsafe. The residual is scaled against the coefficient magnitudes, but a small residual does not guarantee many accurate digits for a severely ill-conditioned polynomial or tightly clustered roots. Last reviewed: 4 August 2026.
A degree-n polynomial has exactly n complex roots when multiplicity is counted. A repeated value occupies more than one of those n root positions.
Yes. With real coefficients, non-real roots occur in conjugate pairs: if a+bi is a root, then a−bi is also a root.
A root has multiplicity m when its factor appears m times. For example, (x−2)³ has root 2 with multiplicity 3.
Exact rational roots are kept as reduced fractions when detected. Roots from a final quadratic or a higher-degree numerical solve are decimal approximations, including complex values.
There is no general solution by radicals for arbitrary degree-five and higher polynomials. Numerical root finding also gives a practical common method for irreducible cubic, quartic, and higher-degree factors.
Enter zero for its coefficient. For x³−4x+1, enter 1, 0, −4, 1 from the highest power to the constant.
The calculator substitutes each displayed numerical root into the original polynomial. It scales the remaining error by the sizes of the coefficients and powers of the root; smaller is better.