I found this post:
Why is memcpy slower than a reinterpret_cast when parsing binary data?
where somebody uses reinterpret_cast to convert binary data to an integer. However (I presume) the number they are converting is at the 0th element in the char* array.
How could I use the above for situations where the binary number I want to convert is offset N bytes from the beginning of the char array?
I want to convert the binary number in as fewer CPU cycles as possible, hence my interest in reinterpret_cast
and the above SO question.