[slackercruster]

How can yougo past infinity?

[luftfluss]

Buzz Lightyear can.

[lytrytyr]

The lens needs room to let the AF mechanism go back and forth around the exact focus point.
So if the exact focus is at infinity, the lens will go beyond and out of focus,
signaling the mechanism to bring it back.

[dms]

some lenses (longer focal length, mirror, etc.) are more prone to changes with temperature (expansion and contraction) a Infinity setting is likely the postion at nominal temperature

[SteveM]

Buzz Lightyear can.

You beat me to it!

My guess would be that it would more difficult to manufacture a lens that bottoms out exactly on infinity.

[lytrytyr]

some lenses (longer focal length, mirror, etc.) are more prone to changes with temperature (expansion and contraction) a Infinity setting is likely the postion at nominal temperature

Actually, if the whole system, camera, lens body, and glass, all expanded uniformly at the same rate,
then a lens focused at infinity when cold should stay focused at infinity when hot.

Metal lenses and camera frames are close to this ideal, while most plastic lenses are not.
However, I believe that Sigma now use a new plastic material which does get closer.

[RobA_Oz]

I was perplexed about this with a MF mirror tele lens (so it isn't just about AF) to the point where I took it to a local optical repair shop for adjustment. The response I got was that "lots" of lenses do this, even some more expensive ones, and unless the user was particularly concerned about it, he always suggested they learn to live with it. Given that they're a successful agent for surveying and other professional optics, I took this to be advice worth considering, especially as they were advising me against their commercial interest.

The point about differential expansion is well made, too, and I guess that focussing beyond infinity is also more noticeable with longer FL lenses, where the visible effect of DoF is generally more critical. There are also manufacturing tolerances, that are bound to be looser with cheaper lenses (cheaper, that is, relative, say, to Leitz or CZ products, whose build as well as optical excellence you pay a premium for). It would be interesting to know the variation in such measures that exist for given lenses.

[lytrytyr]

My guess would be that it would more difficult to manufacture a lens that bottoms out exactly on infinity.

Most old-style MF lenses have a so-called "hard" infinity setting,
so you can focus to infinity just by turning the focus ring to the limit.
Of course, for maximum depth of field,
it's usually better to back off that to the hyperfocal setting.
I don't know how astrophotographers handle this.

[lytrytyr]

Seems like some Nikon D800 users are having problems with this "hard infinity,"
so another reason for focusing beyond infinity might be
to allow camera manufacturers to get away with being sloppy about the register distance:

D800/E and Carl Zeiss lenses - Photo.net Nikon Forum

[RobA_Oz]

Most old-style MF lenses have a so-called "hard" infinity setting,
so you can focus to infinity just by turning the focus ring to the limit.
Of course, for maximum depth of field,
it's usually better to back off that to the hyperfocal setting.
I don't know how astrophotographers handle this.

Given manufacturing tolerances, that "hard" setting would have to have been a final factory adjustment on each lens, which could be so, but my source (see my comment above) seems to have experience that runs counter to it.

[lytrytyr]

Given manufacturing tolerances, that "hard" setting would have to have been a final factory adjustment on each lens, which could be so, but my source (see my comment above) seems to have experience that runs counter to it.

The 7/13/12 11.10am post in the Nikon forum I quoted
seems to suggest that Zeiss at least
would even be willing to reset the hard infinity on lenses under guarantee,
to compensate for a defective Nikon body:

"If you really need a reliable infinity stop we can adjust all of your lenses to your individual Nikon D 800 body"


The 7/13/12 11.51am post emphasizes the utility of a hard stop:

"[T]he shot on the splash page of my website right now was . . . shot wide open or maybe at f2.8 at the dead of night using the hard stop alone - no focusing performed of any kind."

[wombat2go]

I refer to MF, not AF.
If a lens focusses past infinity on the stop, it means that the elements are too close to the sensor, caused by the register being wrong, or the helicoid being incorrectly meshed, or by the shims between the lens barrel and the K mount being incorrect.

I find a lens that goes past infinity to be uncomfortable because 1) you can't quickly rack to the stop and take a sharp shot and 2) because you can't use the distance numbers on the ring for a hyperfocal shot etc.

Also if a zoom lens is not correct at infinity on the stop, it might not focus sharply at any distance.

I make home machined lens mounts and adaptors etc and I spend a fair amount of time to get the registration as close as possible.
Adjustment usually needs to be less than 0.02 mm error ( ~ 1 thousandth inch), and I use shim rings made of cooking parchment if necessary.

However as far as I have tested, all of the collection of old -M and -A lenses here are still correct, so there doesn't seem to be any variation with age.
I have taken some off brand lenses apart for repair, and had to mess around on reassembly to get them right.

[dms]

I refer you to books on lenses for the effect of temperature. It is most pronounced with very long focal length and partcularly mirror lenses. About the latter Cooper says (in Nikon Lenses and Lens Systems) " To compensate for minute displacement of focus caused by temperature changes ... may be turned beyond the infinity marking."

[demp10]

If you add a filter to a long telephoto and/or a teleconverter, you are effectively changing the optical formula of the combined system. A hard infinity mark may not be "at infinity" on such system. Given that telephotos have very thin DOF this is much more pronounced than on shorter lenses.

Allowing past infinity is an insurance policy that can accommodate such accessories along with manufacturing tolerances and temperature variations.

[lytrytyr]

If you add a filter to a long telephoto . . . you are effectively changing the optical formula of the combined system.

A well-made filter should be plane,
so shouldn't make any geometric change
to rays coming in from infinity.