I think the point here is that Aidan's system has no moving parts and does not track the sun. If you accumulate the total energy produced by the array over a period of one year in, say, capacitors, then I am not certain at all that his tree configuration would not be better than a fixed array of cells.

Of course, the fact that the picture in his paper showed a bright white wall just besides his experimental setup removed credibility to the whole thing.

But this prevents no one from repeating the experiment with better control over the variables and who knows...