 So what I want to show, if I can get this thing to show anything, is experiments I've been playing with when it comes to using spectral color modeling for ink simulation. And I'm not going to go into the details of how this works. Or maybe I should do it, because I've built my computer. Light is not RGB, as we see it. There's a, from like 380 nanometers up to around 800 nanometers, energy is distributed. And the way that we actually get a color, I'm not going to get an image here now it seems. The way that you get, okay, the way that we get a color, when I'm looking at these green chairs, is that we have energy in this entirety of the visual spectra, and the chair absorbs most of it, and only gives me back what will, in my retina, be perceived as green. So it absorbs all the other energy. And this is how subtractive color mixing works. And to simulate pigments or inks, it's not enough to simulate like RGB or opposite like cyan, magenta, yellow. You actually have to go all the way of simulating the full spectrum. It is possible to even create a couple of paints which are gray, and you mix them, and you get a blue color or a reddish color. And the actual appearance of a paint or ink is not really relevant for how things end up being. So I'll load up an image in GIMP, and I'll reduce the size of it slightly just to make things go faster. So we'll have an image here which has quite a few colors in it. And I'm going to invoke this gaggle operation which is called the ink simulator. I showed a slide of this in my starting opening talk. What it does first now is to kind of rig up a very bad simulation of a cyan, magenta, yellow, and black printing process. But I can edit this specification of the colors interactively. So if I remove these inks and say we only have cyan, magenta, well, we don't really would be able to achieve a really good result. But if we change our mind a little bit and say we have black and we have magenta, then what this thing does is that it figures out with which different combinations of ink levels between 0 and 100%. When you run it through this spectral simulation engine, it would get closest to colors on the retina that we had in the original image. And I can do other fun things as well. So I can say that my substrate is not white, but it's red. So we have a red background, but printing with black and magenta and red, that's kind of crappy. And so we instead are going to print with yellow. That doesn't really help, but I can say that the yellow is actually opaque. So it's a paint instead of an ink. And then we can add red on top of that. So this is a simulation of how we can do something even more reasonable, not print with red ink on top of that. So we can have a black background and then have a yellow opaque ink and a red one on top of it. So in this way, I can specify a set of different inks and generate a separation. So now we had yellow and red ink. If we just say separate, for now we just hack it up and chose it in a red and green channel. And then this operation also has another mode for proofing a given thing if you feel it in RGB image. This is kind of a work in progress, but I'm trying to go down the route of how close can I get to physics in simulating this. And yeah, thank you.