The Rayleigh criterion gives a limiting resolution assuming monochromatic light. The quoted 16 mas is a demonstrated resolution for a broadband filter. So, I wa

The Rayleigh criterion gives a limiting resolution assuming monochromatic light. The quoted 16 mas is a demonstrated resolution for a broadband filter. So, I was ballparking things and they have achieved an awesome result for a relatively short wavelength! That resolution is just achieved for the inner 1 arcsec of the image, though.