How this calculator works
Text input avoids the precision limit of ordinary numeric fields. Decimal digits are stored as an integer with a decimal scale, so very long numbers and small fractions retain their exact value. Normalisation puts one non-zero digit before the point. Optional significant-figure rounding acts on the digits directly; a carry can increase the exponent. The scientific output keeps requested trailing zeros in rounded mode, while the plain decimal shows the value without redundant zeros.
Formula
x = a × 10^n, with 1 ≤ |a| < 10 for x ≠ 0. If an integer coefficient has d digits and decimal scale s, n = d - s - 1. Round the coefficient to the chosen significant-digit count before normalising.
Worked example
The 30-digit integer 123456789012345678901234567890 becomes 1.2345678901234567890123456789 × 10^29 without losing its value. Rounding 0.009995 to three significant figures carries into 1.00 × 10^-2, equal to 0.01.
Advertisement
Frequently asked questions
Can I paste e notation?
Yes. For example, -4.20e-3 is accepted and represents -0.0042 exactly. Exponents are limited to -1000 through 1000.
Are trailing zeros measurement uncertainty?
Exact mode preserves the numeric value and removes redundant zeros. Rounded mode shows the requested significant-digit count, but it does not infer measurement uncertainty from the input.
What happens to zero?
Both results show zero. There is no unique normalised exponent for zero, so the tool explains that exception.