Multilayer assembly layers
In a first approximation convergent assembly every layer of a specific size needs only the height of one single cell.
See: Branching factor and Level throughput balancing
Contents
Stacking & keeping speeds constant => Downstream bottleneck at upper layers
Stacking more than one layer without also slowing down these layers relative to the bigger chambers above
leads to the the chambers above becoming a bottleneck.
Quantitatively stacking sub-layers to the thickness of super-layers leads to overproduction
by a factor of the branching factor. This is undesired.
Stacking & dropping speeds => Potentially less losses
Stacking is helpful though in case slowing down is desired in order to reduce friction and other losses.
Especially at the lowest levels of convergent assembly molecular mills might benefit from that because of
inefficiencies in piezochemistry and chemomechanical conversion possibly far exceeding friction losses.
Assembly-motions vs transport-motions
While the assembly-motions in multilayer assembly layer can be dropped down the
transport-motions still need to stay fast.
This is due to the macroscale slowness bottleneck.
But with higher branching factors the ratio of assembly-motions to transport-motions (in terms of distance)
massively increases. (TODO: quantify this)
