I have an expression which I am evaluating both in Java and JavaScript.
Java
System.out.println(582344008L * 719476260);
Output: 418982688909250080
Javascript
document.write(582344008 * 719476260)
Output: 418982688909250050
Why is there a difference in both values?
Javascript numbers are all IEEE 754 doubles, so it looks like there is some floating point error going on here. see Difference between floats and ints in Javascript? for some more details
582344008.0 * 719476260.0 in Java is indeed 4.1898268890925005E17, which is the same as the result reported for JavaScript.As @Charlie Wallace says
582344008 * 719476260 is greater than Number.MAX_SAFE_INTEGER
What is JavaScript's highest integer value that a number can go to without losing precision?
As **@baseballlover723 and @RealSkeptic say it's a floating point error in precision/different storage type size rounding error.
See javascript types:
There are 7 basic types in JavaScript.
number for numbers of any kind: integer or floating-point. Integer/Decimal values up to 16 digits of precision. JavaScript numbers are all floating point, stored according to the IEEE 754 standard. That standard has several formats. JavaScript uses binary64 or double precision. As the former name indicates, numbers are stored in a binary format, in 64 bits.string for strings. A string may have one or more characters, there’s no separate single-character type.boolean for true/false.null for unknown values – a standalone type that has a single value null.undefined for unassigned values – a standalone type that has a single value undefined.object for more complex data structures.symbol for unique identifiers.The typeof operator allows us to see which type is stored in a variable.
typeof x or typeof(x)."string".null returns "object" – this is an error in the language, it’s not actually an object. byte for 8-bit signed integer
short for 16-bit signed integer
int for 32-bit signed integer
long for 64-bit signed integer
char for two bytes in Unicode
float for *decimal* values up to 7 digits of precision
double for *decimal* values up to 16 digits of precision (64 bits)
boolean for true/false
Decimal values with a fractional component is called floating point. They can be expressed in either standard or scientific notation.
A quick way to test is use an online java compiler (Compile and Execute Java Online (JDK 1.8.0))
public class HelloWorld{
public static void main(String []args){
System.out.println("Long Output: "+582344008L * 719476260L);
System.out.println("Float Output: "+582344008.0 * 719476260.0);
}
}
The command:
$javac HelloWorld.java
$java -Xmx128M -Xms16M HelloWorld
The Output:
Long Output: 418982688909250080
Float Output: 4.1898268890925005E17
My desktop calculator goes to this:
4.189826889E17
and for JavaScript online compiler:
//JavaScript-C24.2.0 (SpiderMonkey)
print("Output: "+582344008 * 719476260)
The Output:
Output: 418982688909250050
doublein Java, notlong.