Should the same model using Quadtree with smaller cell count be always faster than HPC?

Not necessarily. By default, running model on a mesh (Quadtree) rather than a grid (HPC) is always slower, on average 20%. Quadtree really comes into an effect once there is at least three levels of cell size and judicious refinement - around 80% of cell count reduction. As Quadtree development is ongoing task this might further improve in the future.