Complete guide
Rounding calculator for decimal places, significant figures and increments
Enter a number, choose decimal places, significant figures or an increment, and select the rounding rule. Calculator24 shows the scaled value, adjustment, absolute error and nearby boundaries.
Decimal places
A positive count preserves digits to the right of the decimal point. Zero rounds to a whole number; a negative count rounds to tens, hundreds or larger places.
Significant figures
Significant figures begin at the first nonzero digit, so their decimal position changes with the magnitude of the number.
Rounding to an increment
An increment such as 0.05 restricts results to its multiples. This is useful for measurement steps and cash increments where applicable.
Nearest ties to even
NIST describes retaining an even last digit when discarded digits are exactly five followed only by zeros. This avoids always pushing exact ties upward.
2.5→23.5→4Directed methods
Floor moves toward negative infinity, ceiling toward positive infinity, and truncation toward zero. Negative inputs therefore distinguish these methods clearly.
Precision limitations
Binary floating point cannot represent every decimal fraction exactly. For regulated reporting, tax, money or uncertainty propagation, apply the exact rule and decimal precision required by the governing standard.
Rounding methods at ±2.5
| Method | +2.5 | −2.5 |
|---|---|---|
| Nearest, ties to even | 2 | −2 |
| Nearest, away from zero | 3 | −3 |
| Floor | 2 | −3 |
| Ceiling | 3 | −2 |
| Toward zero | 2 | −2 |
Frequently asked questions
What does two decimal places mean?
Keep two digits to the right of the decimal point.
Can I round to hundreds?
Yes. Choose decimal places and enter −2.
What are significant figures?
Digits counted from the first nonzero digit, regardless of decimal location.
What is bankers rounding?
A common name for rounding exact halfway ties so the retained digit is even.
How do negative numbers round?
The result depends on whether the method targets nearest, an infinity, or zero.
Is binary floating point exact for decimals?
Not always; use a required decimal arithmetic standard for regulated work.
What to keep in mind
JavaScript numbers use binary floating point. The engine detects practical half ties with a scale-aware tolerance, but decimal quantities requiring exact legal or accounting rules should use a decimal system designed for that domain.
Results display up to 8 decimal places; exported numbers preserve calculation precision. This is a calculation summary, not an official certificate.