Store a long decimal in C

Question

I am trying to do a physics problem and need to store a value around 5 * 10^-11; After trying float, long double and a few others none of them see to be long enough. Is there a data type that will allow me to do so? Thanks

long double I = 0;
I = 0.01902*pow(0.00318,3)/12;
printf("%Lf\n",I);

Output is 0.000000

Take a look at *arbitrary precision numbers*, e.g. in [this thread](http://stackoverflow.com/questions/1218149/arbitrary-precision-arithmetic-explanation) or [GNU MP](https://gmplib.org/). — Jens, Apr 04 '16 at 00:42
What exactly makes you conclude that `float` or `long double` are not "long enough"? — Kurt Revis, Apr 04 '16 at 00:44
IIRC, that value should be encoded correctly in an IEEE754 single-precision floating point number. — 3442, Apr 04 '16 at 00:44
Show us your code. You're probably screwing up your output formatting or something so the result just looks like it's zero. — user2357112, Apr 04 '16 at 00:47
@Jonski: According to your edit, your problem is due to the precision you're printing the number in. Read about `printf()` format specifiers. — 3442, Apr 04 '16 at 00:52
@KemyLand Thanks man, just needed to put this as my print statement printf("%.15Lf\n",I); — Jonski Goldstein, Apr 04 '16 at 00:55
http://stackoverflow.com/questions/16839658/printf-width-specifier-to-maintain-precision-of-floating-point-value — Déjà vu, Apr 04 '16 at 00:59
What compiler are you using? Some of the Windows compilers mess up `long double` — M.M, Apr 04 '16 at 02:38
Note that you are performing the calculation in `double` precision and then assigning the result to `long double`. To do the calculation in `long double` precision, put the `L` suffix on your constants, and call `powl` instead of `pow`. — M.M, Apr 04 '16 at 02:43

score 5 · Accepted Answer · edited Apr 04 '16 at 02:01

long double I = 0;
I = 0.01902*pow(0.00318,3)/12;

At this moment, I's value is approximately 5.096953e-11. Then...

printf("%Lf\n", I);

The sole format specifier in this printf() call is %Lf. This indicates that the argument is a long double (L), and that it should be printed as a floating-point number (f). Finally, as the precision (number of digits printed after the period) is not explicitly given, it is assumed to be 6. This means that up to 6 digits will be printed after the period.

There are several ways to fix this. Two of them would be...

printf(".15Lf\n", I);

This will set the precision to be 15. As such, 15 digits will be printed after the period. And...

printf("%Le\n", I);

This will print the number in scientific notation, that is, 5.096953e-11. It too can be configured to print more digits if you want them.

Store a long decimal in C

1 Answers1