A perfect 32″ telescope is diffraction-limited to 0.1725 arc-seconds of separation between objects, which translates into one ‘effective’ pixel of resolution per 0.1725 arc-seconds. The pixels in our current setup are 0.28 arc-seconds wide. Atmospheric seeing limits terrestrial telescopes to between 5 arc-seconds and 0.5 arc-seconds of resolution on average, which we’re probably on the wrong end of given our altitude – but this changes with weather systems (in particular the jetstream) and from millisecond to millisecond. It’s hard to distinguish atmospheric seeing from the in-telescope thermal variation in large telescopes, so it would be interesting to investigate that case as well.