The Cuboid might be better, but because he used two completely different fans, this is not a valid comparison. The Cuboid uses a ducted fan, where the tall duct

The Cuboid might be better, but because he used two completely different fans, this is not a valid comparison. The Cuboid uses a ducted fan, where the tall duct will increase the flow rate by promoting expansion in the exhaust. The other fan is just a non ducted (or very poorly ducted) box fan. It looks like he also only does one trial for his measurements. Most of the efficiency CADR/power probably comes from using a better fan, which is the whole purpose of adding a long duct. The extra effectiveness might also come from the Cuboid fan being elevated off the floor. With the box fan standing on the floor and pointing horizontally, it has an additional solid boundary restricting intake air. The vertical pointing fan in the Cuboid does not have this restriction. A proper test would use the same fan in both designs. The non-Cuboid fan would be oriented upwards and mounted on stilts. Multiple trials should be done with the results averaged. He omitted a chart showing both particles counts vs. time of both approaches, which would be the most important figure to include.