Jump to content

Why a flat chip ?


adrian bastin

Recommended Posts

Because any existing M lens focuses at a given distance, not a range of distances. How would

your existing M camera work if you decided to replace the flat filmback with a dished one?

<p>

The idea you're grasping for isn't a dished sensor. It's a flat sensor, in which the sensor pits

are not vertical but are splayed. But that idea won't work because the angle of the pits would

have to differ for each lens you use. A microlens above each pit seems to work...

Link to comment
Share on other sites

Ah eyes, well the retina is, for all intents and purposes, flat at the microscopic level of rods and cones. Even the foveal pit (the precise center of the macula at which the highest density of cones resides) is flat-ish at it's bottom. The only part of the human eye capable of seeing 20/20 (or better) is the fovea. And it's only about 1.8mm across. Even the crappiest lenses can manage to project a flat plane over such a small area. Still, it is not the eye that "sees", rather the brain compensates for any imperfections secondary to a curved receptor surface (the fundus). And, yes, the cornea is the original aspheric refractive surface.

What does this all mean wrt photgraphy? Not much, the two systems are quite different.

 

Ron (an optometrist by profession)

Link to comment
Share on other sites

If the sensor were curved, none of the existing lenses would work on the M8. While this would make for interesting reading on these hallowed pages, it doesn't make much sense. Rectilinear lenses are designed to focus a subject plane parallel to the film in a plane at the film. A few cameras in history have been designed in this way, primarily to eliminate spherical aberation. The now retired Schmidt Cassagraine telescope at Mt. Palomar is the most famous example. This incidently permits a very wide field and low numerical aperture because the lens is so simple. Of course, Cassagraine telescopes don't have chromatic aberation.
Link to comment
Share on other sites

I posted this before. Focus on a word in a paragraph longer than five lines. Can you recognize a word more than one line or word more than one removed horizontally? Peripheral vision is built in the eye/brain combination for survival, but not for detail. The Dept. of Motor Vehicles (to get a drivers license), and a former employer (to determine if I'm fit for field work), for our Int'l members, showed I had greater than 180 degree field of vision, given my forward set of eyes in an Asian facial structure.
Link to comment
Share on other sites

The reason why chips are flat is because the processes used to put microelectronics on them have been developed to work best on flat surfaces. The masks used to selectively expose and etch the surfaces are flat so they can be laid up almost in contact with the silicon chips during processing. I suppose it'd be possible to make masks & chips that are spherical but it'd be really out the ordinary & cost cubic dollars.

 

It's a little bit like the joke about why manhole covers are round - because manholes are round. AFAIK, the round shape is the only simple shape that doesn't allow a cover to fit thru itself.

 

As someone pointed out, a better solution would be to modify the sensor cell well angle to better accept the photons that don't come in at right angles to the surface, but this would be difficult. Microlenses are not perfect, but by offsetting the periperal lenses they could accept more light from the center axis. Maybe the peripheral cells could be made larger to the same effect but this could lead to greater noise.

 

Sigh. As an engineer I'm often unhappy with the universe and its limitations, but I try to work around them and take advantage of those weird effects that break in my favor. I'm continually surprised at how well well we've been able to crutch things and bend the rules.

 

Cheers,

Link to comment
Share on other sites

Adrian said, "So howabout a gently bent chip and a dedicated lens ?"

 

Can you imagine if the number of current lenses were doubled to accommodate a special digital sensor. Considering the bandwidth dedicated to discussing the endless permutations of the 50mm Summicron, Photo.Net would need a new server ;-)

Link to comment
Share on other sites

If the light sensor were curved as a retina and the lenses were no longer flat field but deliberately curved field, the sensor output would have to be "flattened" just as the brain does in a way for our vision. It could never be anywhere close to a 24 x 36 frame, but a curved one, i.e. not a 2D data map.

 

How would we get the sensor's input onto flat piece of paper or a flat screen?

 

Or would we store our prints as curved dioramas as well and get concave computer screen with exactly one sweet spot to view our photos from? That would be neat but expensive: in material, workmanship, storage space, etc etc.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...