I swap from looking at left and center together, to right and center together, so that she never makes a full circuit and is just swinging her leg back and forth
This is a very unoptimized GIF. Did you know GIFs are limited to at most 256 colors (or 255+transparency) but the palette can be redefined (good export software automatically optimizes it at least according to the first frame)? As such, GIFs that are largely grayscale or sepia can do without dithering and still have smooth colors. (In this one, I would assign 248 colors for the black to white gradient plus 4 shades of red and blue. Or almost halve the file size by limiting the palette to the roughly 16 actually used colors.)
I have been nerdier on Reddit before, and I’m happy to bring that over here. (You won’t see my Reddit nerdposts though, they have been overwritten by PowerDeleteSuite.)
This one has higher resolution and no unhelpful dithering. Also the color palette, while reducible to 64 or fewer colors instead of 256, is actually somewhat optimized (there is no green or yellow).
Don’t worry about optimizing your GIFs, that is a rant about a problem that needs to be solved by someone else. Just accept the higher compatibility but worse quality/bitrate ratio and move on.
Yeah, this is actually much more interesting than the original.
It’s impressive how just a casual glance at the overlapping colours gives you an immediate and definite sense of direction.
And depending on the display size / viewing distance you can easily lose the direction of the left one when looking at the right (and vice versa), so they all start turning in the same direction. That’s because you only have sharp and good color vision at a rather small area (the fovea) everything outside is blurry and has less colour information, so your brain loses that trigger.
The outstretched foot is what I'm looking at, the angle is highly unnatural. I suspect it was done to be intentionally obscure, but the outstretched foot, to me, from this horrible gif rendering, appears larger and smaller based on the clockwise rotation (indicating closer/further from the pov).
But the images depth appears to be compressed, you get this effect in photography by standing very far back and using a high zoom lens. The difference between near/far is smaller by comparison to the distance to the focal point, giving the illusion that things are more 2D. There's still hints you can look at, but the effect makes the differences between near and far, much smaller and harder to pick out.