First, maps can differ in the correspondence they have to a territory. Put another way, some maps are better than other maps. The correspondence a map has is a function of a reader's ability to receive its representation of the territory. (Note to self: Let's make a point of increasing out ability to receive representational forms, even the ones that we don't feel very comfortable with.)
Some maps are better for me, and others are better for you. We in our different lives living according to different maps sorting the information received. We rub alone one another in civil ways we appreciate inside and outside our communities.
Some maps are not good for anything more than a message to us that these thoughts were so wrong they ended in ruinous whorls of pain, sorted into concerns and worries after the fact. Seeing those ruins can take a generation or more to settle out into the maps we keep in our hearts and minds to save us the pain we know they can cause any of us.
Some maps are better, still. The representation of memory accesses in C is not one of them, at least from what I've read. To be frank, C as a language I have a lot of unexplored-ness about. Well, the compiler too. Just what does the compiler know, apparently more than it should some times as it aggressively does predictive branching, sigh.
Some maps are better, as one can see comparing the efficiency of operations between Roman numerals and Arabic numerals. In the land of ciphering, the Adding to Subtracting: The operations with Arabic numerals are mapped more efficiently from first operands to the last. It is not the case that one works and the other does not, only one is far more efficient than the other in processing quantitative narratives. And so the correspondence may initially be the same and result in the same end, but the journey via one representational system compared to another is different.