For David Millier

I can't tell is the answer because I don't know where the false detail is.

I can see clear traces of jaggies and rope artefacts afflicting certain branches and twigs and I don't consider them beneficial. But this is a minor problem in this image.

I can't tell you whether there is beneficial aliasing going on because I don't have a comparison image that would let me see what it looked like without aliasing. So, all I'm left with is the assumption that there must be aliasing in there that is helpful. But I have no evidence of this, just the assumption.It could all be detail with no aliasing at all, who can say.

That problem aside, this thread was predicated on the conjecture that sometimes aliasing is helpful and sometimes it is harmful. And what we are looking for is an alternative to aliasing that has the benefits without the drawbacks.

If it were possible to find some image processing that compensated for the presence of an artefact killing AA filter by adding in "a hint of texture" above Nyquist that fooled the eye into seeing detail, that would be good, no? One hypothesis is that film-like grain could do this, roughening up plastic surfaces and simulating texture where there isn't any. A kind of dithering-like effect.

--
Galleries and website: http://www.whisperingcat.co.uk/
Flickr: http://www.flickr.com/photos/davidmillier/