Double randomly adds 0.000000000000001

Question

I have this code:

double timeTillTick = 15.0;
private void lower_Tick(object sender, EventArgs e)
{
    if (timeTillClear > 0)
    {
        timeTillClear -= 0.1;
        clearingIn10SecondsToolStripMenuItem.Text = "Clearing in " + timeTillClear + " seconds.";
    }
    else
    {
        lower.Enabled = false;
    }
}

lower ticks once every 100 milliseconds. When it:

Gets to 8
Gets to 5
Gets to 1

it increases by 0.000000000000001. Why?

We can use `decimal` to keep floating points accurate(called fixed points) — Aniket Inge, Nov 06 '13 at 05:04
http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html — Andrew Savinykh, Nov 06 '13 at 05:07
@chipperyman573 `double` is floating point decimal, where as `decimal` is a fixed point decimal. — Aniket Inge, Nov 06 '13 at 05:12
@Aniket `decimal` is not fixed-point. `double` is floating-point binary (base-2), `decimal` is floating-point decimal (base-10) — Cory Nelson, Nov 06 '13 at 05:20
@CoryNelson oh yeah you're right, buts its still deemed good enough. (Just looked at MSDN and its confirmed), I was wrong. My bad. — Aniket Inge, Nov 06 '13 at 05:23

score 5 · Answer 1 · answered Nov 06 '13 at 05:09

5

Much like 1/3 cannot be represented exactly using decimal notation (0.333333...), 0.1 cannot be represented exactly as floating point number, which is internally using binary notation (aka IEEE-754). That's why you get that inherent error.

answered Nov 06 '13 at 05:09

mvp

111,019
13
122
148

Double randomly adds 0.000000000000001

1 Answers1