 So if there are so many advantages with simulations, are there any problems? Well, yes, there are. Actually, I wouldn't necessarily say that's limited to simulations, but many of the things we do both in this class and in experiments for that matter. Have you seen this image? René Maquitte, 1929, from an exhibition called The Treatury of Images. On one hand it's funny and it's deep depending on how you see it. You could argue that it's not a pipe, but it is a pipe. Well, Maquitte's argument here was he can't say that this is a pipe because you can't really stuff it. It's two-dimensional, not three-dimensional. It's just a picture of a pipe, and in your case it's electrons delivered over the internet to your screen. So the whole point is while we might think it's a pipe, this is kind of a model of a pipe. I can't turn it around, I can't look inside it, and I certainly can't stuff or smoke it. The reason I'm showing this is I want you to think of this as not being a protein. This is a model of a protein. This particular model happens to be based on a simulation, but it's not that the simulation aspect is particularly problematic here. If you're determining an experimental structure, there are no experimental structure methods that will give you XY and Z coordinates that are based on either your or somebody else's model interpretation. It doesn't mean that the models are bad, but don't confuse your model with reality. You have to be aware that you're looking at a model, in particular when we have computers that render beautiful things. They're even moving and they have amazing colors, but this is just a set of XY and Z coordinates from a simulation that I have visualized and rendered for you, and it could be incorrect. So what are the things to think about? Well, we can compare them with experiments, which is quite instructive.