How to keep the precision of double variable in C++?

Question

It is not about printing. It is about storing the value.

int x = 1234;
double y = 0.3456;

double z = x + y;

Currently z contains 1234.35. I want z to contain 1234.3456

What can be done to achieve this?

`z` contains an accurate value. You aren't printing it with sufficient precision. — user207421, Aug 04 '16 at 09:47
if you're still having trouble, show the complete program that is producing the undesirable output — M.M, Aug 04 '16 at 09:49
@TheIndependentAquarius I agree. You've presented no evidence. — user207421, Aug 04 '16 at 11:44

score 5 · Accepted Answer · answered Aug 04 '16 at 09:45

5

It does contain 1234.3456. Check for yourself:

#include <cstdio>

int main()
{
    int x = 1234;
    double y = 0.3456;
    double z = x + y;

    printf("%.8f\n", z);
}

Output:

1234.34560000

answered Aug 04 '16 at 09:45

rustyx

80,671
25
200
267

In other words, yes it is about printing (or formatting output) after all. – Rudy Velthuis Aug 04 '16 at 12:25

Franck · Answer 2 · 2016-08-04T10:48:55.593

2

The double accuracy contains a bit less than 16 decimal digits. The printed value of z is less precise than the stored one. If you want more than 16 digits, you should consider specific libraries like mpfr

edited Aug 04 '16 at 10:48

answered Aug 04 '16 at 09:49

Franck

1,635
1
8
11

15.9 decimal digits actually. – user207421 Aug 04 '16 at 09:57
Depends on if you need the round-trip decimal to binary to decimal work. http://www.exploringbinary.com/number-of-digits-required-for-round-trip-conversions/ – Maxim Egorushkin Aug 04 '16 at 09:59
@EJP thanks, 15.9 is 53 / log_2(10). – Franck Aug 04 '16 at 10:51

How to keep the precision of double variable in C++?

2 Answers2