Here's a comparison of bad and good seeing (captured with a D = 90 mm telescope with a solar Hα filter):
https://app.astrobin.com/u/GreatAttractor?i=246828#gallery
Technically speaking, in the first half of the vid the incoming wavefronts are distorted (the Fried parameter r₀ is smaller than the telescope's diameter D) and do not focus to a clean, tight Airy pattern ([1]). In the second half, r₀ is above D and we're left with just some rubber-membrane distortion (the wavefronts are tilted this way and that, but remain mostly planar), which can be corrected in software.