Now I don't disagree that an array of cells that track the sun would produce better results, but it would be fair to assume that this would cost energy.
Keeping that in mind, do you think it would still provide better results?
"tracking systems can increase viable output by up to 100%...For large systems, the energy gained by using tracking systems outweighs the added complexity (trackers can increase efficiency by 30% or more)"
The new tree design can make 20%-50% more electricity. Wouldn't know till they do some legitimate tests but it seems like the sun trackers fare better.
Why compare the designs, when you can combine them? I'm sure that the tree design can be improved a lot more, by just slightly turning the "leaves" so that they face the sunlight more directly.
I am willing to wager the suggested "combined" design would produce a very similar results anyway. The tree design is likely not necessary if the panels are moving.
I'm not an expert here, but it's pointless to make tracking sun panels that costs a lot more if the outcome wasn't that great, I bet tracking sunlight is still more efficient even after spending energy to move the panels, it makes no sense otherwise. The real benefit of this is that you get a cheap motionless panel that performs a lot better than the flat version, this is specially important in places like Africa where you need to be as efficient and cost effective as possible.
SolarFlower.org - the open source solar collector - has a non-electronic sun tracking system that I think might be more efficient than PV-powered tracking. See http://www.solarflower.org/faq.htm.
Tracking panels may prove to be a better design for maximum power output.
However, there are other constraints to engineer for. Cost and durability (through simplicity of parts) come out ahead in a design with no moving parts.
At the end of the day, it's better to have knowledge of both designs to know what's most appropriate for your needs.
The real benefit of a tracking system is that it makes it possible to use a concentrator to focus more sunlight on a smaller PV area. The panel cost is mostly in the PVs, so increasing collecting area by using mirrors is more cost effective than just using more PVs. However, this setup only works if it's pointing right at the Sun, so it requires a tracker.
Keeping that in mind, do you think it would still provide better results?