Hello Alice,
in general, the total time of processing depends mainly on quantity and quality of data inputs, as well as their mutual overlap, and your hardware performance.
However, the reconstruction in normal detail is usually faster than in high detail, that is right. Keeping the cache also saves a lot of time.
See an example here: https://support.capturingreality.com/hc/en-us/articles/115001483771-Huge-scene-reconstruction