Significant Figures Calculator

What significant figures are

Calculate, count and round significant figures with clear precision rules, exact arithmetic and step-by-step explanations.

Written digits convey measurement precision. For example, 12, 12.0 and 12.00 have equal numeric values but report different levels of detail. Significant-figure rules help avoid reporting more detail than the inputs support; they do not replace a full uncertainty analysis.

How to use

Choose Calculate for an expression, Count for a single written number, or Round for a number and a target of 1–15 significant figures. Results update as you type. Keep every intended zero in the input. Use a decimal point rather than a comma.

Calculate accepts +, -, *, /, ×, ÷ and parentheses. A single unary + or - may precede a numeric literal. Scientific notation accepts 1.200e3, 1.200E3 and 1.200 × 10^3; the last form is one scientific numeric value, not a measurement multiplied by an uncertain ten.

Examples switch to the appropriate mode without reloading. Show Steps opens the execution trace. Count and Round offer Show in scientific notation. Copy Result exports the visible input, raw result, final result, precision rule and any ambiguity note. Reset returns to Calculate with 12.30 × 4.5; Count’s example value is 0.004050 and the Round target is 3.

The default example, native steps disclosure and all educational content work without JavaScript. Changing values or modes and copying require JavaScript.

How to count significant figures

Leading zeros locate the decimal point. Zeros between nonzero digits are significant, as are trailing zeros after a decimal point. Scientific notation counts digits in the written coefficient; changing the exponent does not change that count.

Examples: 123 has 3 significant figures; 12.30 has 4; 0.004050 has 4; 0.00070 has 2. In Count mode, 0.004050 appears as 0.00[4050], with brackets as well as highlighting to identify significant digits. The middle zero and final zero inside those brackets both count.

Multiplication and division

For multiplication or division, use the fewest significant figures among the operands. The tool preserves the exact arithmetic value separately from that reporting constraint. Every entered literal is treated as precision-limited; there is no exact-constant switch in this version.

12.30 × 4.5 has an exact raw value of 55.35. Since 4.5 supplies only 2 significant figures, the displayed answer is 55. Likewise, 0.6238 × 6.6 reports 4.1 and 421.23 ÷ 486 reports 0.867.

Addition and subtraction

Addition and subtraction use the least precise decimal place, not the smallest total digit count. For 12.35 + 1.2, the tenths place governs. The raw sum 13.55 is therefore reported as 13.6.

2.334 + 0.31 reports 2.64 at the hundredths place. For 56.533 − 55.8752, the thousandths place governs, so the exact difference 0.6578 is reported as 0.658. Cancellation can reduce the number of significant figures even when the reporting place stays unchanged.

Mixed calculations and guard-values

Parentheses are evaluated first, then multiplication/division, then addition/subtraction. Operators at the same level proceed left to right. Each intermediate carries both an exact value and its reporting precision.

For (12.30 × 4.5) + 2.1, the product is 55.35 with a 2-significant-figure reporting limit, equivalent to the ones place. Addition retains 55.35 internally, adds 2.1 to get 57.45, then reports 57 at the ones place. Feeding the displayed intermediate 55 into the next step would discard useful arithmetic information.

For (3.40 + 1.2) × 2.00, the sum 4.60 is reportable as 4.6, or 2 significant figures. The product retains the exact 4.60 internally and reports 9.2. The trace distinguishes the guard-value used internally from each reported intermediate.

Internally, reduced BigInt fractions retain exact values, including non-terminating quotients. No finite guard-digit cutoff controls final rounding. Raw displays are exact for short terminating decimals; other values show a clearly marked approximate 24-significant-digit view. That view is never used for subsequent arithmetic.

Trailing zeros and ambiguous integers

NIST explains that plain integer notation such as 1200 does not uniquely identify whether its final zeros express precision. This tool applies a deterministic classroom convention: trailing zeros in an ordinary integer without a decimal point do not count. Thus 1200 counts as 2 here, with an ambiguity warning.

Write 1200. or 1.200e3 to explicitly indicate 4 significant figures; 1.2e3 indicates 2. Similarly, 100 counts as 1 under the convention, while 100. counts as 3. This convention is not an assertion about the actual accuracy of an unspecified measurement.

Scientific notation

A coefficient and power of ten can communicate precision clearly. 1.200e3 and 1.200 × 10^3 both mean a coefficient with 4 significant figures. The output uses readable superscript exponents: 1.200 × 10³. For 0.004050, the corresponding scientific form is 4.050 × 10⁻³.

Integer results ending in zeros use scientific notation to avoid ambiguous precision. A result of 1200 with a 4-digit target is shown as 1.200 × 10³. Ordinary output is otherwise preferred for exponents from −6 through 8. Displayed zeros are retained when they carry reporting precision. Superscript output is for reading; use e notation or ^ for typed exponents.

Round half to even

We use round half to even. When the discarded part is less than half a reporting unit, leave the retained value unchanged; when greater, move to the nearer value. At an exact halfway tie, choose the value whose final retained digit is even. Negative values follow the same symmetric rule.

At 2 significant figures, 2.541 becomes 2.5, 3.450 becomes 3.4, and 3.550 becomes 3.6. A nonzero digit beyond the five breaks the tie: 3.4501 becomes 3.5. The decision compares exact integer remainders, without relying on binary floating-point rounding.

Asking Round mode for more digits than the original number supplies adds formatting zeros; it does not improve a measurement’s accuracy.

Assumptions, zero and limits

Zero convention: 0 and 0.0 count as 1 significant figure; 0.00 counts as 2. For all-zero decimal literals, the fractional zeros set the count and the last written place sets decimal precision. This is an explicit display convention: zero has no nonzero leading digit from which to infer relative precision.

Addition/subtraction zero results retain the governing reporting place. A zero product or quotient uses the requested significant-figure count as fractional zeros, for example 0.00 for a 2-figure zero. Do not interpret this convention as a propagated uncertainty estimate. Division by zero is rejected.

Product limits: 512 expression characters, 128 tokens, 16 nested parentheses, 64 characters per literal, scientific exponents from −100 to 100, and at most 2,048 decimal digits in a reduced fraction’s numerator or denominator. Arithmetic beyond those limits is rejected. Round targets are integers from 1 through 15.

Functions, variables, assignments, units, implicit multiplication and general exponentiation are unsupported. Inputs are parsed as a small arithmetic language and rendered as escaped text. No code is executed from expressions.

Everything calculates in this tab. There is no calculator persistence, URL state, analytics or calculation request. Sources provide educational context and do not endorse BuzCalculator.

Worked examples

Multiply: 12.30 × 4.5 = 55.35 raw → 55 reported, limited by 4.5’s 2 figures.

Add: 12.35 + 1.2 = 13.55 raw → 13.6 reported, limited to tenths.

Count: 0.004050 → 4 significant figures, including its interior and final zeros.

Round: 12.3456 to 3 figures → 12.3. Round 0.004050 to 2 figures → 0.0040; the exact halfway tie retains the even zero.

Scientific: 1.200e3 ÷ 3.0 = 400 raw → 4.0 × 10², explicitly retaining 2 figures.

Significant figures FAQ

What are significant figures?

They are the digits used to convey a measured value’s reported precision. They do not by themselves establish accuracy or a complete uncertainty interval.

How do I count significant figures?

Ignore leading placeholders; count nonzero digits and interior zeros, then apply the stated trailing-zero convention. Count mode identifies the relevant digits in brackets.

Are leading zeros significant?

No. In 0.004050, the zeros before 4 only locate the decimal point.

Are trailing zeros significant?

After a decimal point they communicate precision. Trailing zeros in plain integers may be ambiguous, so this tool flags them.

How many significant figures does 1200 have?

Under this calculator’s ordinary-number convention: 2 significant figures. The notation itself is ambiguous. Use 1.200e3 or 1200. to explicitly state 4.

What is the rule for multiplication and division?

Use the operand or intermediate with the fewest significant figures. The raw numeric value is retained separately.

What is the rule for addition and subtraction?

Use the coarsest decimal place among the inputs or intermediate reporting limits. Total significant-figure counts do not govern this rule.

How do significant figures work in mixed calculations?

Follow precedence and carry the reporting constraint through each step. Show Steps explains which intermediate or value limits the next operation.

Does scientific notation change significant figures?

Changing only the exponent does not. Written coefficient digits establish the count; dropping or adding coefficient zeros can change the stated precision.

When should I round intermediate calculations?

Keep extra arithmetic precision until the final result. Here, exact fractions are retained throughout; rounded intermediates in the trace are reporting examples only.

Sources and methodology review

Methodology checked September 30, 2026. NIST’s SI guidance supports clear quantity notation and explains ambiguous integer zeros. Appendix B.7 describes half-even rounding. OpenStax provides classroom significant-figure and operation rules. Zero handling, parser bounds and the exact-fraction implementation are documented product choices.

• NIST Guide to SI, Chapter 7: Expressing values of quantities (opens in a new tab)
• NIST Guide to SI, Appendix B.7: Rounding numbers (opens in a new tab)
• OpenStax Chemistry 2e: Measurement Uncertainty, Accuracy, and Precision (opens in a new tab)
• OpenStax University Physics Volume 1: Significant Figures (opens in a new tab)