08-04-2015, 10:36 AM
Yeah, it doesn't make a lot of sense in most cases, but players are certainly going to experiment with having windows without mirrors on their hab modules:
![[Image: 8krEh.jpg]](http://i.snag.gy/8krEh.jpg)
...and when they do, we have to decide what light level to simulate inside.
Now in design mode, we don't have the light direction quite what it should be; it should be coming straight in from the side, but the harsh shadows make it difficult to work that way, so we've angled the light a bit here.
But once you actually build your colony, it's oriented perpendicular to the plane of the ecliptic, which means that the light will be coming in straight. So if you've got a torus, or a cylinder with inverted endcaps, those windows are seeing nothing but starlight.
But if you have a sphere or a cylinder with outward endcaps, then the windows do see some sunlight. Much of it would bounce right off, of course, given the low angle to the sun, but some of it would get in and bounce around inside (especially if the windows are diffuse, as I assume they would probably be).
This is where it gets really hand-wavy, though... how well lit is that? Just in terms of area, if you look at one of these spheres from the end (as when using them with a mirror), the windows are about 50% of the presented area. But from the side, the area presented is much smaller, more like 25%. So that's a 50% reduction or so.
Or you could look at it this way: with a mirror, 100% of the windows are receiving sunlight, but without it, only 50% of the windows are lit (and much of that is at an extremely low angle), so that also means 50% or less light getting inside. Waving frantically, I say we peg the amount of light you can get this way at 40%.
For a cylinder, the situation is much the same, except that if your endcaps aren't hemispheres, you're going to get proportionally less light as you flatten them. And of course we also need to take into account the length of the cylinder... even when you're using mirrors, it makes sense that the longer the cylinder, the less light (per unit area) you're getting from the same windows.
Finally, there's barbells. These ones are really complex. Each bell obviously faces away from the sun half the time, but when its facing the sun, whether it's shadowed or not depends on whether the number of bells is even or odd, and (if odd) how big the hub is. And of course even if you have a tiny hub and 3 bells, the sunlight is going to be going on and off at the same rate as your spin (e.g. every 15-30 seconds). If that doesn't drive your residents up a tree, I don't know what will!
(And in the mirrorless cylinder or sphere, though the light would be a constant amount, the angle of the light would cycle at the same rate... less annoying than cycles of light/darkness, but still pretty annoying I bet.)
...and when they do, we have to decide what light level to simulate inside.
Now in design mode, we don't have the light direction quite what it should be; it should be coming straight in from the side, but the harsh shadows make it difficult to work that way, so we've angled the light a bit here.
But once you actually build your colony, it's oriented perpendicular to the plane of the ecliptic, which means that the light will be coming in straight. So if you've got a torus, or a cylinder with inverted endcaps, those windows are seeing nothing but starlight.
But if you have a sphere or a cylinder with outward endcaps, then the windows do see some sunlight. Much of it would bounce right off, of course, given the low angle to the sun, but some of it would get in and bounce around inside (especially if the windows are diffuse, as I assume they would probably be).
This is where it gets really hand-wavy, though... how well lit is that? Just in terms of area, if you look at one of these spheres from the end (as when using them with a mirror), the windows are about 50% of the presented area. But from the side, the area presented is much smaller, more like 25%. So that's a 50% reduction or so.
Or you could look at it this way: with a mirror, 100% of the windows are receiving sunlight, but without it, only 50% of the windows are lit (and much of that is at an extremely low angle), so that also means 50% or less light getting inside. Waving frantically, I say we peg the amount of light you can get this way at 40%.
For a cylinder, the situation is much the same, except that if your endcaps aren't hemispheres, you're going to get proportionally less light as you flatten them. And of course we also need to take into account the length of the cylinder... even when you're using mirrors, it makes sense that the longer the cylinder, the less light (per unit area) you're getting from the same windows.
Finally, there's barbells. These ones are really complex. Each bell obviously faces away from the sun half the time, but when its facing the sun, whether it's shadowed or not depends on whether the number of bells is even or odd, and (if odd) how big the hub is. And of course even if you have a tiny hub and 3 bells, the sunlight is going to be going on and off at the same rate as your spin (e.g. every 15-30 seconds). If that doesn't drive your residents up a tree, I don't know what will!
(And in the mirrorless cylinder or sphere, though the light would be a constant amount, the angle of the light would cycle at the same rate... less annoying than cycles of light/darkness, but still pretty annoying I bet.)
Joe Strout
Lead Developer, High Frontier
, what about the O'Neill type cylinders that bring in light not via the endcaps but through mirrors that shine the light on the cylinder's window panes?
