I love this story because it demonstrates the difference between science and conspiracy mongering, and between experimental science and ungrounded theory.
There is a famous quote from Shakespeare that is often hurled a scientists:
There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy.
– Hamlet (1.5.167-8)
The retort from scientists should be, “that’s why we do experiments, so more things can be included in our understanding”. But the Bard was correct that Nature is always more complex than our best representations of Her, which is why “all models are wrong, but some are useful.”
In this case, some anomaly hunters were making noise that the pictures from the Moon landing just didn’t “look right.” This is not science – which requires that you take the data points that don’t appear to fit the existing theoretical framework and offer a revised theory that works better.
After talking with many engineers and NASA employees, Nvidia discovered that there were two key factors, both of which could be addressed using voxel global illumination. First, the moon’s surface is comprised of what are essentially thousands of tiny mirrors — moon dust if you will — that bounce light back at a viewer. Yet that didn’t account for the necessary level of brightness to light up Aldrin.
So Nvidia engineers began tinkering with different elements of the photograph until they discovered that it was not what was in the frame, but who was behind it. The famous shot was snapped by Neil Armstrong — who was off to the side of Aldrin in full view of the Sun — wearing a 85 percent reflective spacesuit that contained five layers of the highly reflective fabric Mylar blended with four layers of the flexible yet durable material Dacron on top of an additional two layers of heat resistant Kapton.
Using modern computer graphics, the lighting from the picture was recreated. The essential step was including not only sunlight reflecting off the lunar regolith, but also from the photographer’s spacesuit! Since we don’t live on the moon, and our intuition about lighting is based on Terran photography, we are surprised that this was important. Clearly, someone trying to fake a moon landing in 1969 would have to be unbelievably smart to think of this. Similar to the halting problem in computer science – which says that no method exists for predicting if a program will terminate early or not – sometimes the only way to know what will happen is to actually try it. This is the inherent limitation of theory uncoupled from empirical evidence. Trial and and error was also the way Aperture Science was able to discover that, in addition to reflecting 12% of incident light, moon dust is an excellent conductor of portals.