elkcub
Silicon Valley, California

Hi tenex,This theory sounds interesting but I don't understand it yet. Wouldn't a defocus gradient due to the eye's own curvature of field still apply when gazing straight ahead through binoculars? (And wouldn't it then be curved-field models that exaggerate that gradient in a potentially odd way?) Or am I falling into the old camera-analogy trap, whereas field curvature here is a property of the whole eye-binocular system, so FFs correct for both?
My reference to a "cue conflict situation" is based in part on the following understanding of how the eye takes advantage of field curvature.
"Field curvature is a measure of defocus for off-axis objects and implies that the best image is formed not on the paraxial image plane but on a parabolic surface called the Petzval image surface (Smith & Atchison 1997). In real eyes, the retina, which may be approximated as a sphere with a radius between 11 and 13 mm, constitutes a curved image plane that in most cases compensates for field curvature." (from: Image Formation in the Living Human Eye," by Pablo Artal, 2015)
To answer your question, I don't know whether field-flattener design includes corrections for the eye's cornea and lens curvatures, or where the resulting "flat image" is formed relative to the eye's Petzval surface. It would seem that they project an extended paraxial image tangent to the curved retinal surface, which would then produce cue conflicts that somehow result in the spatial perceptions I mentioned. I could be completely wrong. It's just a conjecture.
Ed
Last edited: