Ah, interesting. I've only experienced it on the A7iii, which has it fixed more or less completely (none of the green blobs or any errors showing up in the article you linked).
https://petapixel.com/2018/06/08/sonys-star-eater-problem-ha...
It still exists on A7 III: https://blog.kasson.com/a7iii/a7iii-long-exposure-spatial-fi...
Those are noise power density graphs that show that signal went through a low-pass filter. On Sony cameras without 'star-eater' and cameras from other manufacturers it is flat.
But I admit that this issue is most noticeable on S-series (low resolution/high sensitivity) cameras, as their lower resolution makes it easier to resolve stars down to 3x3 pixel size. Normal and R-series cameras have such high resolution that it is hard to notice and not much of a problem in amateur astrophotography.
I have not seen any proper tests of latest A7S III as it was only released last month.
Re:The Sony “Star Eater” (2017)
by sudosysgen on Tuesday September 01, 2020 @06:35PM
(#24347925)
I don't really think much of anyone has optics sharp enough to resolve a star to one pixel with perfect contrast on an a7R. Maybe the 135 2.8 GM stopped down to f/4 in the middle could do it?
Many people do not see that issue or claim that it was 'fixed', because they do not have optics sharp enough to resolve stars to 1x1 or 2x2 pixels.