Absolutely. So in which criterion was lack of 64-bit guaranteed data type based on, in C89? At the time of publishing that standard (1989), one could argue problems didn't need more than that integer size, or it may have been unusable with processors of that time (I think MIPS32 and MIPS64 were popular in the early 90's, none of which could handle more than 64 bits number). And as I'm reading pooya's reply, we haven't exceeded that on the newest processors neither.