But not related to computability per se, isn't there also an issue of physical cardinality? What I mean by this is that maybe we can't efficiently simulate a 3

But not related to computability per se, isn't there also an issue of physical cardinality? What I mean by this is that maybe we can't efficiently simulate a 3D human brain on any 2D computer, because there are vertical connections that are nearly instant in 3D that must map to a very very long route in a 2D "layers" simulation. My hunch is we will efficiently simulate human brains once (1) we have finished the Human Connectome Project (2) we know much more systems biology about what happens inside both neurons and glia and (3) we have solved the heat-transfer and interconnect and yield issues in building 3D computer chips