What are the actual ranges of floating point and double data types in C#?

Question

I'm learning C# and trying to get a logical visual representation of the actual range of data types in C#.

I have moved through the integers and am now up to float and double data types.

1. 8 bits (1 byte), sbyte, -128 to 127.
2. 8 bits (1 byte), byte, 0 to 255.
3. 16 bits (2 bytes), short, -32,768 to 32,767.
4. 16 bits (2 bytes), ushort, 0 to 65535.
5. 32 bits (4 bytes), int, -2,147,483,648 to 2,147,483,647.
6. 32 bits (4 bytes), uint, 0 to 4,294,967,295.
7. 64 bits (8 bytes), long, -9,223,372,036,854,775,808 to 9,223,372,036,854,775,807
8. 64 bits (8 bytes), ulong, 0 to 18,446,744,073,709,551,615.

Here are the references to float and double data types sizes at msdn:

Float: http://msdn.microsoft.com/en-us/library/b1e65aza(v=vs.110).aspx

Double: http://msdn.microsoft.com/en-us/library/678hzkk9.aspx

So, trying to keep with the convention of specifiying the actual range of numbers as in the numbered list above, what do these two ranges actually represent?

Answer 1

The ranges are actually –infinity to +infinity.

The largest finite float is 340282346638528859811704183484516925440. This is 2 ¹²⁸ –2 ^128–24 .

The largest finite double is 179769313486231570814527423731704356798070567525844996598917476803157260780028538760589558632766878171540458953514382464234321326889464182768467546703537516986049910576551282076245490090389328944075868508455133942304583236903222948165808559332123348274797826204144723168738177180919299881250404026184124858368. This is 2 ¹⁰²⁴ –2 ^1024–53 .

Answer 2

The ranges are represented in "exponential format" for conciseness. For example, +1.7e+308 means 17 followed by 307 zeros:

1,700,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000

So the exponential format is preferred for such large numbers. And the same goes for extremely small numbers.

Also, take a look at this reading by Jon Skeet.