> Or, maybe I'm just having trouble visualizing a smooth surface geometry that also has few normals back? I naively assumed the "smooth" planes were enabled by

> Or, maybe I'm just having trouble visualizing a smooth surface geometry that also has few normals back? I naively assumed the "smooth" planes were enabled by better absorption materials, rather than geometry. Maybe a mix? You're not wrong. From a head-on perspective, especially from below, you'll find few if any curves. Newer stealth planes look curvier because photographs are often taken from above where the canopy is prominent, and where the top of the body and wings have some curvature. But look more closely. Even from above, the curves you see are usually on the trailing portion of surfaces or facing laterally; the leading edge of the wings on a B-2, F-22, or F-35 are actually flat and triangular, not at all like a typical diagram of an airfoil. This is especially true of the F-22 and F-35--if you look very closely, they're far more angular from more perspectives than the older B-2. The B-2 looks curvier from below, but the flying wing design isn't a coincidence; and I believe the B-2 also relies more heavily on radar absorbing skin than later aircraft, which rely more on simple geometry much like the F-117. Moreover, beneath the skin of these planes it's widely believed that the framing uses sawtooth and other similar polygonal patterns you'd expect, a mitigation for radar that passes through the skin. And I would think that part of the engineering of these aircraft leverages radar transparent skin in some areas, not just absorbant skin, similar to a nose cone holding a radar. Here's a really amazing slide presentation on the basics of stealth design: https://understandingairplanes.com/Stealth-Airplane-Design.pdf https://understandingairplanes.com/Stealth-Airplane-Design.p... (I first came across that document via an HN post several years ago.)