Rounding SO small float values (0.00000000000001)

Question

Hello, everyone
Im trying to solve a problem my way (not good, i know, but just searching) and found another problem with rounding values.

int discount = Convert.ToInt32(Console.ReadLine());

Dictionary<string, int> coffee = new Dictionary<string, int>();

coffee.Add("Americano", 50);
coffee.Add("Latte", 70);
coffee.Add("Flat White", 60);
coffee.Add("Espresso", 60);
coffee.Add("Cappucino", 80);
coffee.Add("Mocha", 90);

string[] s = coffee.Keys.ToArray();
int[] p = coffee.Values.ToArray();

for (int i = 0; i < p.Length; i++)
{

Console.WriteLine(s[i] + ": " + p[i] * (0.01 * (100 - discount)));
}

And I have 2 questions about that:
1) I have a dictionary here which have <string, int> types. So, why the result with float? It should be integer.
2) when I put "5" in "discount" variable it calculate : 60 * 0.95 = 57.00000000000001. I found the reason, but I cant understand how to fix exactly this problem. If I use "Ceiling" method, it is rounding this to 58 from 57. So how to fix this small float value to closest integer? Exactly this. Cause I need all another rounding to bigger one integer (i mean 66,5 to 67, 47.5 to 48... etc)

Answer 1

You can use Math.Round(num) ;
Math.Round(57.001) returns 57
Math.Round(47.8) returns 48

Answer 2

I was trying that, but didnt get the way to solve this more procedurally.
If I put 57.000000000001 exactly it will not solve same problem with other values

Answer 3

for (int i = 0; i < p.Length; i++)
{
Console.WriteLine( s[i] + ": " + Math.Round( p[i] * (0.01 * (100 - discount))) );
}

This way am telling, is this not solved your problem

Answer 4

Its almost right, but it returns 66 (should return 67).

Ive found that better use integer values in this formula, then it returns right result.

But i cant understand, why it work like that. Results should be same.

value - value * discount / 100 -------- it returns right result......

Answer 5

There occurs a slite difference in fraction part value, when you do calculation with double values..
With integers, you always get the floored value.