Alright, I made some progress.
This sheet metal will do just fine for testing. It's a piece of the bus ceiling and is the same gauge thickness as the roof. White on one side, black on the other.
Six equal sized pieces cut. 12"x20".
All pieces cleaned and mounted on a frame built of scrap wood. I turned one of the pieces over so we can determine the temperature difference between plain black and white paint.
I picked up 5 US gallons of Henry's Solarflex
Here we have the magic beads. I was really disappointed in the labelling. It up-talks the product a bunch, but contains no useful information about the contents. Specifically weight or volume. I had to dump it all into bowls then measure it back into the bag to find out that it contains 4.25 cups (just over 1 litre). The package says the contents are properly measured for 1 US gallon (3.79 litres) of paint.
Since I only need a small amount of paint for this test run, I mixed 0.5 quarts (473 millilitres) of paint with a bit over 0.5 cups (125 millilitres) of Hytech beads; a ratio of 3.79:1.
Now, the elastomeric paint alone is obviously going to be thinner than elastomeric paint + ceramic beads. As such, I would expect the straight paint to be less effective (less material, thinner thermal-bridge). I wanted to have a test sample that would be about the same thickness as the paint + beads so I whipped up a batch of paint + sawdust using the same ratios.
Here it is all painted up! From bottom to top: 2 coats elastomeric paint + sawdust; 2 coats elastomeric paint + beads; 2 coats elastomeric paint; 4 coats elastomeric paint; black painted side up; off-white painted side up. I'll post some test results at midday when it's good and sunny.