C# Overriding the GetHashCode method

Question

In this example, the poster has overridden the get hash code method. I understand that this has been done in order to provide a better hash value for the returned object, to reduce the number of collisions, and therefore reduce the number of occasions it will be necessary to call Equals().

What i would like to know, is how this algorithm been calculated:

return 17 + 31 * CurrentState.GetHashCode() + 31 * Command.GetHashCode();

Is there a particular reason that the numbers in question were chosen? Could i have simply picked my own numbers to put into it?

• Marc Gravell: Just for info, the MS C# compiler (for anon-types) uses a seed of -1134271262, and a multiplier of -1521134295. Just sayin'
• DeepSpace101: @MarcGravell: Do you have the source for that?
• Marc Gravell: @DeepSpace101 ILDASM ;p
• DeepSpace101: @MarcGravell: I see the same -1521134295 multiplier in ILDASM but the seed appears to be a composite seed (it's a random int32 multiplied by the same multiplier). So the seed's not only NOT prime (contrary to popular belief), it's also random based on the anon-type member names/type/count. Weird. Any ideas why?
• Marc Gravell: @DeepSpace none. Interesting.

Answer 1

Generally you should choose primes. This helps you to avoid getting the same hash-value for different input parameters.

Answer 2

Prime numbers are usually used in hashcode computation to minimize the collisions. If you search for hashcode and prime numbers on this iste, you will find some detailed explanations on this (note that it is note language specific):

• What is a sensible prime for hashcode calculation ?
• Why does Java's hashCode() in String use 31 as a multiplier ?

Answer 3

You typically want to use prime numbers (as is done above) because it reduces the chance of collisions (two instances yielding same result). For more info, see: http://computinglife.wordpress.com/2008/11/20/why-do-hash-functions-use-prime-numbers/