I do hope you deduced that by 'memory space' I mean the size of the memory allocated to the random access data structure of the PoW algorithm.
Oops; I didn't realize you were talking about a tiny Cuckoo Cycle instance. I normally think of sizes in the dozens or hundreds of MB. Apologies for the misunderstanding.
In that case you are right that most memory accesses will be coelescing into the same row/page.
But when you start running multiple instances they may occupy different pages in the same bank
and conflict with each other.
The 2^-14 applies when most of your physical memory is used for Cuckoo Cycle,
as might be tried in an ASIC setup.
I had PM'ed you to collaborate on PoW algorithms and alerted you to my new post thinking that in the past you've always been amicable and a person to collaborate productively with.
I'm just keen to correct what I perceive as misrepresentation or false claims about my work.