Exact answers where exact answers exist, and the honest domain warnings
A power is repeated multiplication until the exponent stops being a whole number, at which point it becomes a definition: a fraction exponent is a root, a negative exponent is a reciprocal, and zero to the power zero is undefined, which this calculator says out loud instead of guessing. Enter a base and an exponent (whole, decimal, or a fraction p/q) and the page gives the exact form wherever one exists: whole-number powers of integers are computed exactly, however large; fraction powers of integers are simplified into an exact radical by factorising the base; everything else carries its decimal and a plain statement of its domain, including the complex answer a negative base with an even root requires.
Disclaimer: This calculator gives estimates for general information only. It is not financial, medical or other professional advice. Exact results are given for the integer and rational cases described; other values are decimal approximations.
What this calculator does
Three exponent ideas share one page because they are one idea at three ages. The main tool evaluates: whole powers exactly by integer arithmetic (2^64 prints all twenty digits, not a rounded shadow of them), negative powers as exact reciprocals, and fractional powers exactly as simplified radicals wherever the base factorises kindly, since 8^(2/3) = 4 deserves to be shown as 4 and 12^(3/2) as 24√3. The rules tool then works symbolically, applying the product, quotient and power rules to a general base x: the algebra move, isolated from arithmetic so its shape can be learned. Throughout, the domain is treated as part of the answer: 0^0 undefined, zero to a negative power undefined, and a negative base under an even root answered in the complex numbers and labelled as such.
How to use it
- Enter fractions as fractions. 2/3 as p/q is exact; 0.667 is a different, approximate question, and the tool knows the difference and answers accordingly, decimalising only what you decimalised.
- Negative exponents are reciprocals, not negatives. 2^(−3) = 1/8 = 0.125: the sign moves the power across the fraction bar, a distinction the results spell out because it is the most-lost mark in the topic.
- Growth is the context. Powers explode: 2^10 ≈ a thousand, 2^20 ≈ a million, 2^30 ≈ a billion, and the exact-integer mode exists so that pattern can be met at full size rather than in scientific notation’s shorthand.
- The rules tool takes exponents, not values. a and b may be fractions or negatives; the result exponent is computed exactly as a fraction, showing why x^(1/2) × x^(1/3) = x^(5/6) and not anything involving roots multiplied blindly.
The method
- Whole exponents: exact integer power by repeated multiplication in arbitrary precision when base and exponent are integers of manageable size (results up to about 300 digits); otherwise floating point with the value in standard form.
- Fraction exponent p/q (lowest terms): factorise the base; each prime exponent e becomes e·p, split as (e·p div q) outside the root and (e·p mod q) inside it, exactly the radical simplification of the sister calculator; a negative exponent takes the reciprocal of the positive result.
- Domain: 0^0 undefined; 0 to a negative power undefined; negative base with denominator-even fraction exponent has no real value and the principal complex value is given; negative base with whole or odd-denominator exponents is real and signed by the power’s parity.
- Rules: product adds exponents, quotient subtracts, power of a power multiplies, each computed in exact fraction arithmetic and displayed as a fraction where it is not whole.
Worked examples
8^(2/3) = 4. Read the fraction as root then power: the cube root of 8 is 2, squared is 4. The factorisation route agrees: 8 = 2³, the exponent triples to 6, split by 3 gives 2² outside and nothing inside. A calculator that prints 3.9999 has understood less than the method, which is why the exact path exists.
12^(3/2) = 24√3 ≈ 41.57. Read it as root then power: √12 = 2√3, and cubed that is 8 × 3√3 = 24√3. The factorisation route agrees: 12 = 2² × 3, exponents tripled give 2⁶ × 3³, and splitting by 2 puts 2³ outside with nothing left, and 3¹ outside with 3¹ still inside, so 2³ × 3 × √3 = 24√3. Two methods, one value, and the decimal 41.57 to locate it.
Reading your results
Read exact results as the answer and decimals as its size: 24√3 is what the value IS, 41.57 is roughly where it lives. An exact whole-number power in full digits is a small marvel worth pausing on: 2^64 ends …616 and has twenty digits, the grains-of-rice-on-a-chessboard number, and printing it whole is the difference between knowing and being told. When the page reports a value as undefined, that is a mathematical verdict with reasons given, not a gap in the tool: some expressions have no value, and a calculator honest about them is worth more than one that invents 1 for 0^0 because a limit once suggested it.
Limitations
- Very large exact powers are capped. Integer results are exact to about 300 digits; beyond that the page switches to floating point and says so, rather than timing out in pursuit of a ten-thousand-digit trophy.
- Irrational exponents are decimal only. 2^π and π^2 have no exact finite form to print; the decimal is the answer and the page does not dress it as more.
- Complex results are principal values. A negative base to a fraction power has several complex values arranged by angle; the principal one is reported and labelled, as on the radical page.
- The rules tool is single-base algebra. Mixed bases (2^a × 3^a) factor differently (they combine as (6)^a, a different rule’s territory) and sums of powers do not simplify at all, which the tool respects by not offering them.
Frequently asked questions
Why is 0^0 undefined?
Because the two patterns that would define it collide: powers of zero are 0 (0^3 = 0, 0^1 = 0) and zeroth powers are 1 (5^0 = 1, 2^0 = 1), and 0^0 sits at the crossing with both claims and no consistent referee. Some fields adopt 1 as a convention for their own formulas (combinatorics often does), but as a value of the power function it is undefined, and this calculator follows the analysis convention and tells you so rather than quietly picking a side.
What does a negative exponent mean?
A reciprocal: x^(−n) = 1/(x^n). It is a definition chosen, beautifully, to keep the product rule working: x^3 × x^(−3) must equal x^0 = 1, and 1/(x^3) is the unique value that achieves it. What it never means is a negative result: 2^(−3) is the positive number 0.125, and the sign in the exponent is an instruction about position (denominator), not about the sign of the answer.
What is a fractional exponent?
A root in uniform: x^(1/q) is the q-th root of x, and x^(p/q) is that root raised to p (equivalently the p-th power, rooted). The worked examples show 8^(2/3) computed both ways to the same 4. The fraction must be in lowest terms for the domain rules: (−8)^(2/6) is the cube root in disguise and is real, while (−8)^(1/3)… the same: the reduction matters, and this page reduces before it judges.
How big is 2^64?
18,446,744,073,709,551,616, the number from the chessboard legend where the inventor asked for rice doubling per square. The exact mode prints it whole: twenty digits ending in 616. Its lesson generalises: every additional factor of 2 in the exponent roughly triples the digit count of the result per ten powers (2^10 ≈ 10^3), which is why exactness has a cap and logarithms exist.
Related calculators
Method: exact integer powers in arbitrary precision within the stated cap, fractional powers simplified by prime factorisation into exact radicals, explicit domain handling for zero and negative bases, and the three exponent rules applied in exact fraction arithmetic. Last reviewed: 15 October 2026.