I was reading that the maximum F number of a human eye, if treated the same way as a camera lens, would have an F number of 2.1, but basically since the eye cannot be treated the same way as a camera lens, the actual maximum aperture of what we see is about 3.2 (to put it shortly).

The glass of a lens absorbs a portion of the light that passes through and lowers the amount of light the camera receives. For example, my f/1.8 lens was rated a T number of 1.9 since it transmits as much light as a theoretically perfect f/1.9 lens would. I would assume that since the eye is filled with liquid not air like a camera lens, this could have a significant effect on its T Number. I was searching for a value of the T stop of a human eye, but I couldn't find anything on it. Does anyone know what it is or how it could be calculated?

  • 1
    Fascinating question! I see the wikipedia page on the Eye Lens has some information like the index of refraction of the lens. But I don't know enough about it to know whether it contains more useful information to your question. I look forward to an answer! – user1118321 Oct 2 '16 at 1:36
  • 1
    Another interesting question is how much compensation for chromatic and other aberrations the optic chiasma brain supply. It would be nice to use simpler (read: cheaper) lenses, with aberrations handled in software. – DrMoishe Pippik Oct 2 '16 at 2:32

Your eye can be treated exactly like a camera lens, and its maximum f/# really is around f/2.2 or so. The cornea has an index of refraction of around 1.33, so the fresnel reflections are around 2% or so. The reflections off of other surfaces are very small because of minimal index breaks.

T# = F# / T_avg over some bandwidth, so we have f/2.1 / 0.98. The T/# is around 2.15.

| improve this answer | |
  • 1
    According to Wikipedia only half of the photons impacting the cornea actually get to the retina. This would result in a full stop of difference between F-stop and T-stop. See here: en.wikipedia.org/wiki/Absolute_threshold (references: Fundamentals of Sensation and Perception, Michael Levine. Oxford University Press (3rd Edition). London, 2000.; Visual Perception, Tom Cornsweet. Harcourt Publishing (1970). Chapters 2 and 4.; Hecht, Selig; Shlaer, Simon; Pirenne, Maurice Henri (1942-07-20), "Energy, Quanta, and Vision", doi:10.1085/jgp.25.6.819) – FarO Oct 3 '16 at 8:53
  • @OlafM That's part of the story. What's the effective QE of the retina,and what's the fill factor? (same questions for the camera's pixels in its sensor) – Carl Witthoft Oct 3 '16 at 11:23
  • I assume what the OP meant is that the brain is able to compensate (somewhat) for the limitations of the eye through further processing of the raw images. – jpaugh Oct 4 '16 at 14:07

The transmission is meaningless unless you know the sensitivity of the sensor. The amount of light that can be detected by a digital sensor or time is based on light and time. Camera sensors can build up an image of a period of time (shutter speed). The human eye is instantaneous and cannot build up an image over time. The rods which sense light levels are sensitive to 1 photon per rods. The cones which sense color are much less sensitive. Although the lens remains the same, the level of light that can be detected varies. The fluid of the eye (the vitreous body - not the outer aqueous layer varies over a person's life. It gets impurities and floaters with age which effect the light transmission.

| improve this answer | |
  • 1
    That's just plain wrong concerning the human receptor and neurological processing of the image. – Carl Witthoft Oct 3 '16 at 11:24
  • 1
    What does the human eye is instantaneous mean? – Olivier Oct 3 '16 at 18:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.