# Any spreadsheet provide more then double precision?

Excel is not designed for highly accurate calculations. Its design follows the IEEE 754 specification for floating point math, which calls for only the 15 most significant digits to be stored. If the integer portion of a number is 15 places or longer, the mantissa or fractional part will be dropped. Other errors limit the accuracy of Excel calculations further.

Is there a spreadsheet app which has greater precision than Excel?

There was a web site for PrecisionCalc, which offered an add-on, xlPrecision, for Excel to increase data precision, but I cannot reach it at this time. CNET offers a free (trial) download of xlPrecision, though.

Caveats:

• I've not tried it, and cannot vouch for it. The executable from CNET can be unzipped, the CAB file with installer extracted, and EXE and DLL files can be checked at VirusTotal.com. A scan of the CNET file itself showed 65 out of 67 AV engines found no issue.
• Development appears to have stopped a while ago; the description claims compatibility with Windows 98/Me/NT/2000/XP/2003/Vista/Server 2008/7 and MS Excel up to Office v. 2010.
• Gold! PrecisionCalc / xlPrecision is alive and kicking at time of writing. Working in latest Excel 365, appears to be flawless both in high precision decimal / large numbers math. The only downside is that you need to use the formulas of xlPrecision to do your math, i.e. the native math operators like `+`, `-`, `*`, `/` and any functions working with numbers aren't influenced, you have to write things like `=xlpADD(1234567890,0.123456789)` or `=xlpADD(A1, B2)` where you normally would have done `=1234567890 + 0.123456789` and `=A1 + B2`, respectively. Jul 20, 2021 at 17:14

If you need the precision but don't mind moving away from a spreadsheet approach then the Python npmath library allows you to perform real and complex floating point arithmetic with arbitrary precision. It and python are cross platform, free, gratis & open source.

Example (computes 50 digits of pi by numerically evaluating the Gaussian integral with mpmath):

``````>>> from mpmath import mp
>>> mp.dps = 50
>>> print(mp.quad(lambda x: mp.exp(-x**2), [-mp.inf, mp.inf]) ** 2)
3.1415926535897932384626433832795028841971693993751
``````

You could also look for software based on GMP such as Maple.