@lupyuen Yes, I have known about this for about 20 years. I recall one benchmark that to outperform the vector-insert-in-the-middle for a random element, the collection in the benchmark had to be >50,000 entries.
Then to complicate further, it is language sensitive, since for objects, OO languages (all?) need to make the vector a vector-of-pointers-to-objects, compared to an array-of-structs (or struct-of-arrays) in C.
CleverLibre Social is an inclusive social instance for open discussion, learning, and community.
All cultures welcome.
Hate speech and harassment strictly forbidden.
@lupyuen Yes, I have known about this for about 20 years. I recall one benchmark that to outperform the vector-insert-in-the-middle for a random element, the collection in the benchmark had to be >50,000 entries.
Then to complicate further, it is language sensitive, since for objects, OO languages (all?) need to make the vector a vector-of-pointers-to-objects, compared to an array-of-structs (or struct-of-arrays) in C.