Originally posted by: FrankSchwab
I can't think of many real-world tasks where having a 64-bit integer is going to speed anything up.
Having twice the number of registers? Sure, that'll speed things up.
Having a 64-bit instruction pointer? Sure, that'll help a class of current applictions that need more than 4GB of memory.
Having 64-bit integers? That's a tough one. Perhaps it's the reason that Microsoft's Visual Studio still uses 32 bits for its "long" datatype - too many compatibility issues with changing the size of long for too few benefits.
One common application for 64 bit integers is counters.
For example databases usually count everything they count in 64 bits, so that they don't wrap over a 2 or 4 billion. This is one of the reasons why databases see good speedup going to 64 bit on the same AMD64 chip.
Operating systems also have a lot of 64 values in them. With modern hardware you have to count everything in 64 bits, whether it's number of packets, highres time, number of interrupts etc etc
Computation on all these values is greatly sped up if you have native 64 bit integers and don't have to assemble them from two 32 values and manually sorting out the carry and sign bits.