How far away is infinity? (optically speaking, of course)

[.... _]

I have always considered lens infinity to be the point at which the light rays are as close to parallel that any closer makes no discernable difference in the final image output. How close that is is dependant on the variables as discussed in the zeiss document on DOF.
Tricky to nail it down but if you are testing a lens then pick something at least twice as far away as you think you need to which roughly speaking will be twice as far as your lens distance scale is telling you is infinity and you should be OK.

[Dralowid]

I wonder how can a rangefinder work at infinity?

Surely with a baseline of no more than a few inches there would be no angle of view between the two 'views' ie they would be parallel and the images would never match?

And yet they seem to when set up right.

Then I think of the rangefinders you could see on heavy warships of the late '30s (ie before radar ranging) and I think....hhhmmmm...long range gunnery using range finders and mechanical mathematical devices (avoiding the use of the word computer) must have been truly impressive.

But does that have anything to do with infinty??????

Michael

[ferider]

Depending how accurate your camera registration and rangefinder adjustment are, how sharp you want the outcome to be, how much resolution the lens has, how much you care and compensate for focus shift, etc.

Traditionally the rule of thumb for RFs was 1000 x focal length. With modern, sharper lenses, you might want to increase that to 4000 x focal length or similar and as suggested above.

Roland.

[Sparrow]

Anyway, there is very little worth photographing that far away

[sevo]

Originally Posted by Rob-F

The focusing mechanism doesn't enter into it. I'm assuming that the lens is being focused perfectly. DOF does not enter into it.

As to aperture: the effect of a small aperture would be only to mask that difference, via DOF.

Something that does need to enter into it would be the size of the blur circle.

In absolute terms, photographic infinity is the lens-to-focal plane distance where rays entering the lens in perfect parallel coincide in one point on the focal plane. For any "practical value", we're talking of some sort of hyperfocal distance - but there, the above do matter.

[Rob-F]

An interesting, wide variety of associations to this topic! Everything from Butterflies to black holes! Infinite universe, parallel universes. Rangefinder calibration. Lenses that focus past infinity. 42. (That last should be in AU's, BTW, not meters.)

Originally Posted by ferider

Traditionally the rule of thumb for RFs was 1000 x focal length. With modern, sharper lenses, you might want to increase that to 4000 x focal length or similar and as suggested above.
Roland.

This is along the lines of what I was looking for.

Originally Posted by sevo

In absolute terms, photographic infinity is the lens-to-focal plane distance where rays entering the lens in perfect parallel coincide in one point on the focal plane. For any "practical value", we're talking of some sort of hyperfocal distance - but there, the above do matter.

This is true, of course. I'll have to concede that hyperfocal distance does have to be considered. Let's try it this way, though: I wonder how many focal lengths, for optimum results, the subject would need to be for a lens that is optimized for infinity, rather than at some compromise middle distance?

[fstops]

This is an interesting topic.

I don't know the answer to OP but in my case I focus at infinity if I don't want the foreground to be in sharp focus. I focus at hyperfocal distance of a particular f stop if I want focus to be sharp from foreground to infinity.

If this is an incorrect method then please let me know.

[.... _]

Just to throw a spanner into the works, the following has a formula for calculating rangefinder focussing accuracy. I'm not sure I understand it cos 'e' is a new one to me and I have no idea where it comes from.

http://r-d1.info/camera-use/rangefinder-accuracy/

Basically I think its saying you better not go too wide otherwise your focussing will be off. I guess you can calculate for your cameras base length.

[semilog]

Originally Posted by jarski

this was carefully calculated before. answer was 42.

What was the question, again?

[.... _]

Originally Posted by semilog

What was the question, again?

The point was that if you are focussing on infinity to check your rangefinder accuracy at infinity, then you better be sure you are not shooting wide open otherwise you may be fooled into thinking your rangefinder needs adjusting when it doesn't. Just sayin....

[mdarnton]

It depends on your purpose. For adjusting the focus of my 85/1.5 on an M4, two miles was far enough. By that I mean that if I used any closer distance, I couldn't close the two images at farther distances, but aligning at two miles, more distance didn't matter.

If you are relying on or asking about "depth of field". . . there really is no such thing--the specific distance you focus on will always be noticibly better focused than closer or beyond. "Depth of focus" really means "how low are your standards, and what will you tolerate?"

[sparrow6224]

Ansel Adams, as others, accepted the formula of 1000 x focal length as the photographic distance called infinity. We need to know this number because sometimes we need to focus at 1/3 of that distance for best depth of field. For a 28mm lens, 28m, for a 50mm, 50m, and for 100mm lens, 100m. It makes sense. My source for the Ansel Adams is the first of his great series of three books, The Camera, The Negative, The Print. He mentions it in discussing focal length and depth of field in The Camera.

[Rob-F]

Vince, that is some great information. I even have those books, and it never occurred to me to look there. That is just what I was looking for.

Thanks!

[sparrow6224]

Rob -- Good Ruskin quote. I like Ruskin. His reputation was ruined by his marriage -- a very strange and amusing story that if you're interested in him you should look up.

And glad to help.

[Roger Hicks]

Originally Posted by sparrow6224

Rob -- Good Ruskin quote. I like Ruskin. . . . .

Interesting. I really loathe the fellow. He had an exceptionally and unjustifiably high opinion of himself and was the most repulsive and arrogaat reactionary in most matters artistic: think of his reaction to Whistler's Nocturne in Black and Gold: The Falling Rocket.

Also, amusing though it is today, that quote is nothing like as witty in the context of his time. In that context, before widespread affluence, it was the vapourings of a nasty little bourgeois snob.

The story about the marriage has been queried, incidentally. I suspect it has been embroidered by those who dislike him as much as I do, but are/were less scrupulous about how they blacken people's names.

Cheers,

R.

[Sparrow]

... and he often gets accused of burning many of JWM Turners less salubrious work while working as the old man's executor ... which I like to think the little ****e was capable of

[Roger Hicks]

Originally Posted by Sparrow

... and he often gets accused of burning many of JWM Turners less salubrious work while working as the old man's executor ... which I like to think the little ****e was capable of

He may have been capable of it, but I am less convinced that he actually did it; possibly because he wasn't capable of much, except nasty attacks on far greater artists than himself. As far as I recall, at least some of the 'destroyed' pictures have been rediscovered, despite Ruskin's saying he had destroyed them. "All mouth and no trousers" might have been coined for Ruskin.

Cheers,

R.

[citizen99]

Originally Posted by jarski

this was carefully calculated before. answer was 42.

Originally Posted by semilog

What was the question, again?

Hmm, tricky. I'll have to go away and think about it ...

[Highway 61]

Originally Posted by photomoof

Most of my long Nikkors focussed well past infinity, so one could get infinity sharp.

Yes, this was a very clever Nikon feature, to compense for the dilatation of the lens barrels due to temperature variations, which long lenses are very sensitive to.

[jarski]

Originally Posted by citizen99

Hmm, tricky. I'll have to go away and think about it ...

it is "Answer to the Ultimate Question of Life, The Universe, and Everything" from the supercomputer, Deep Thought, specially built for this purpose. Source.

[Pherdinand]

Originally Posted by mdarnton

...If you are relying on or asking about "depth of field". . . there really is no such thing--the specific distance you focus on will always be noticibly better focused than closer or beyond. "Depth of focus" really means "how low are your standards, and what will you tolerate?"

sorry, but that's wrong.
Ideally, you would be right, but, only in case of infinitely good resolution of your sensor, (sub)atomically thin emulsion/sensor and picometer flatness of film/sensor.
Which is obviously not the case.
In practice, it is like drawing two lines: An exponential and a straight line asymptotic to it. In principle, ideally, they only touch in the infinity. In practice, they touch where your pencil thickness is more than the distance between them.
Or where your pixel size is bigger than the distance, in our digital world.

I am surprised that noone gave a **** on my previous reply/replies.
I see three tendencies here in replies:

1. The one who goes very practical and says "2000x focal length is enough". This is a bit-just a bit- too vague and too simplified since it does depend ultimately on magnification of final image, on sensor resolution, flatness, etc. Of course if you pick a number large enough, it will fit all. But that's not 2000xfocal length i am afraid.

2. The completely theoretical answers like the ones with absolute sharpness only at focus distance. This is impractical, plus, it is also not what you see in nature since we live in a 3D world, plus, optics are not perfect, plus, sensors are not perfect, etc.
3. The ones who say 42

If i disregard my own replies, which are the best so far , i agree most with the 3rd types

[Roger Hicks]

Originally Posted by Pherdinand

. . . If i disregard my own replies, which are the best so far , i agree most with the 3rd types

Assuming perfect competition, should we hang economists with light inextensible string?

Note for those who have studied neither economics nor physics: certain assumptions are often made in order to simplify calculations, or as one might say philosophically, to render the conclusions at best partially valid.

Cheers,

R.

[funkydog]

Originally Posted by mdarnton

It depends on your purpose. For adjusting the focus of my 85/1.5 on an M4, two miles was far enough.

That works for me too.

[kraj8995]

This question isn't as puzzling as it appears - potential is, by definition, measured as the work done in bringing a charge from infinity up to a particular point .. so a 'normal' calculation for the mid-point potential is with respect to infinity.a5 envelope

[sparrow6224]

Who knew RFF was red hot center of Ruskin antipathy? Not I, surely. I should not have said anything, having hardly read him. I like the idea of massively influential critics but they do so often prove unpleasant personally. Roger speaks as if from extensive and repugnant experience -- and puts his distaste so forcefully it is difficult not to be influenced by it. Me, I have only read him on how to sketch. Rather thick going too. And, of course he was hyper bourgeois: he was a Victorian.