I didn't benchmark the difference between cache usage with precalculating the hash code in a constructor vs calculation in the hash code itself but string conca

I didn't benchmark the difference between cache usage with precalculating the hash code in a constructor vs calculation in the hash code itself but string concatenation then taking the hash code of that was the real performance killer.