lens filter

[false_Aesthetic]

Hi,


I'm wondering if there's any down-side to putting a filter on behind the lens instead of in front of it.

I'm not so interested in whether or not it'd be a PITA. I'm more interested in the uhm optical quality of the image.

Specifically, I'm wondering if putting a filter behind the lens will cut down on some of the flare I'm getting when shooting with the sun so low and in the frame. I realize I'm not going to get rid of all the flare, but I'd like to minimize it.

Thanks

T

[BrianShaw]

I read somewhere but, of course, can't remember where, that there is a focus shift if filters are used on the rear element. I don't quite understand why that would be, nor do I particularly believe it (unless someone can show some proof that is better than opinion/speculation).

Calumet once, and maybe still does, sell a gel filter holder that mounts to the back of the lensboard. It had an odd name - something like "Xenophon". I use gel filters when I use filters on LF but they go in front of the lens, in a filter drawer that is part of my lens hood.

A lens hood, by the way, is perhaps a better way of reducing flare... unless the sun is setting and you are shooting to the West!

[Ole Tjugen]

Filters have an optical thickness that is different from the physical thickness (if that makes sense).

So if you focus without a filter and then put a filter on the back of the lens, focus will shift a little bit. That's why lenses made for rear-mounted filters (like fisheyes) come with a neutral glass filter for use when no filter is needed - the filter is a part of the optical system!

You can get away with a thin gel filter, and the change is less with longer focal lengths.

[BrianShaw]

Ahhhh.

[Emmanuel BIGLER]

that there is a focus shift if filters are used on the rear element.

There are two effects of inserting a plane & parallel piece of glass behind a lens that focuses an image at a finite distance (the usual photographic situation ;) )

The first effect is a simple focus shift, the amount of which can be found in all basic textbooks in optics, the displacement of the image is equal to t.(1-1/n) when "t" is the tickness of the glass an "n" its refractive index. Textbooks usually take n=1.5 which yields a focus shift equal to one-third of the glass thickness. Gelatin filters being thin will induce a very small focus shift. Note that when the glass is inserted in front of the lens, the image is shifted as well, but shifted from infinity to infinity plus or minus something small ;-) so the image, on film (or silicon ;) ), after passing through the lens, is (in principle) not at all affected.

So this focus shift with a filter behind the lens could be problematic for a rangefinder camera, but for a view camera where you focus, I should say : you re-focus all the time on the ground glass, no problem. For reflex SLR camera, provided that you can re-focus, what you see on the ground glass (GG) is what you get on film as well ! ( provided that the filter is located just behind the lens in front of the reflex mirror so that both optical paths in the reflex camera are identical between viewing/GG and film plane.)
So in practice the only drawback I see of putting a filter behind the lens is that you have you take the lens off to attach/detach the filter.

The second effect is more subtle and is visible only in certain situations like microscope lenses, could be visible for an extreme wide-angle view camera lens.
Even if you correct for the focus shift, the filter introduces a certain amount of spherical aberration if it is inserted between the lens and the image, again in a situation where the image focuses somewhere near the focal plane.
I have no idea if this subtle effect is noticeable in real life for photographic lenses.
If the filter is placed in front of the lens and the object located at infinity, as strange as it may sound, if the piece of glass is really flat and parallel, no spherical aberration is introduced even if incident rays fall at a high angle ! the only image degradation is : flare in any case and of course image quality degradation if the glass is not of good optical quality.


The additional spherical aberratiion induced by a piece of flat glass inserted in a converging/diverging beam is taken into account for high-power microscope lenses used for biology where you focus through a standardized 0.17 mm piece of covering glas.

But for photographic use, using a filter behind the lens it is worth the experiment ! I'm quite sure that it could reduce some flare.

[BrianShaw]

Ahhhhh!

[JW Dewdney]

The focus shift you get from doing that is WELL WITHIN the margin of focus error anyway - so, practically speaking there is no difference. But it would be good practice to refocus with the filter ON - in general. Otherwise, to answer your question- YES, it would yield a better result (again - at least in theory) than being exposed to the source of illumination - where you could pick up flare, etc... If it's buried inside a really good lens hood setup - it would be just about the same result, front or back.

[GPS]

The focus shift you get from doing that is WELL WITHIN the margin of focus error anyway - so, practically speaking there is no difference.
...

Oh, oh, slow it down! It's surely not true for the short focal lengths (wide angle lenses) if by the "focus error" you mean the depth of focus...

[JW Dewdney]

By 'margin of error of focus' - I mean (the distance you actually focus on vs. the distance that you INTENDED to focus on) x (the location of the film vs. the manufacturer's ideal intention) x (whatever other mitigating factors come into play - i.e, wind, vibration, focus shift on stop-down, etc etc..). It MAY be more noticeable on a wide angle (i.e. -result in a larger 'circle of confusion')

[Bob Salomon]

Along with the mentioned focus shift of 1/3rd the thickness of the filter there is also a very good possibility of image degradation from any marks or scuffs on the filter. In addition, filters that rotate; polarizers, grads, etc. are impossible to use conveniently behind the lens.

The only time a filter should be placed in or behind the lens is when that filter is designed to be part of the optical system. An example would be the corrector plates that screw in to the back of the Apo Sironar Digital HR lenses when they are used for film.

[GPS]

JW - the "focus shift" is the change in the distance at which the lens is focused on the film plane, not on the subject distance.

[JW Dewdney]

JW - the "focus shift" is the change in the distance at which the lens is focused on the film plane, not on the subject distance.

Yes, that IS true.. and that is what I understand to be the meaning of focus shift. The way I'm understanding things is, just so we can get on the same page, is that the index of refraction of the glass, being different from the IR of glass, deflects the incoming ray ever so slightly - and, two points, traced back to the same origin in the subject - may have slightly different landing points on the film as a result of the filter being in it's path between lens and film.

That is what I mean when I refer to focus shift.

However - the fact that

a) most people can't focus to save their lives (myself included)
b) most cameras are FAR from rigid
c) most tripods are FAR from rigid
d) film position can be up to 1.5-2mm (or more, depending on other factors) out of registration from the groundglass
e) when setting focus, if one is focussing 'hyperfocally' - there is a fairly broad range of choice upon which one may focus - especially when one is beyond the f/22 mark.
f) any number of other factors

means that this 'focus shift', which would more than likely be on the order of .1mm, though Bob tells us 1/3 the thickness of the glass, which I'm not sure is quite so - as it really depends on the maximum angle of entry for a given lens (so - longer lenses would be more immune to this).

It just seems a picky point when most of us aren't really seeing the big picture in terms of sharpness.

[Ole Tjugen]

... though Bob tells us 1/3 the thickness of the glass, which I'm not sure is quite so - as it really depends on the maximum angle of entry for a given lens (so - longer lenses would be more immune to this)...


No, it doesn't. Bob is absolutely correct. It depends only on the thickness and the refractive index of the glass.

[JW Dewdney]

No, it doesn't. Bob is absolutely correct. It depends only on the thickness and the refractive index of the glass.

I don't think bob said anything ABOUT the IR of the glass, did he?

I'd have to contest that, Ole. So - you're saying that if you had a lens of infinitely long focal length, with all rays coming in perfectly parallel, and perpendicular to the lens board plane - that you would need to refocus to compensate for the filter thickness? The 1/3 the thickness thing is a simple 'rule of thumb' derived from using a 50mm lens on a 35mm camera, under 'standard conditions' however calculated. Probably from a John Hedgecoe 'take great pictures' book somewhere.

But yeah- for most 'general situations' it may well be about a third. But I would place a healthy wager on the fact that it varies with focal length. That's why I was suggesting to focus with filter in place. But my disclaimer was that - it wouldn't make a whole lot of difference in the resulting image in practice - for most people.

(sorry if i seem in a pissy mood tonight - html's getting me down)

[Ole Tjugen]

The "optical thickness" of glass with a refractive index of 1.5 is 2/3 of the "physical thickness. That's one way of stating the definition of "refractive index".

See Emmanuel Biglers post a bit further up.

(sorry if i'm in a pissy mood - bad science is getting ME down:))

[JW Dewdney]

That's a poor way of thinking about it, Ole (speaking of bad science! - heck - bad RHETORIC!). You're thinking like a republican, now. I understand the desire to think about systems simplistically - we all do it - with many things. It might work most of the time - but it's not going to lead to an accurate result. To show you just how much I care - I made a drawing - just for the purpose of illustrating this (small) point.

I hope this makes some sense to you. I have crudely represented our beloved lens, naked, if you will, in one scenario and then, with filter, in the second (see figs. 1 & 2, respectively). We see how the filter (made thick for the purpose of illustrating the principle) has distorted the paths of the light rays passing through the lens on the way to the film. ?1 and ?2 represent the resulting linear error as a result of NOT changing focus. Clearly, we can see here, that ?1 is much greater than ?2, as a result of having passed closer to the center of the optical axis (by the way - optical axis is incorrectly identified on my drawing - X1 should be 'lens plane'). So- clearly, a lens whose design admitted a narrower cone of 'light' would clearly be less affected by this 'focus shift' phenomenon. I hope that makes sense.

Note: Refract = bend. The light is BENT. That means that the angle changes. But only while passing through the 'filter'. The ray continues again, at it's original angle after leaving the filter.

[GPS]

...
I'd have to contest that, Ole. So - you're saying that if you had a lens of infinitely long focal length, with all rays coming in perfectly parallel, and perpendicular to the lens board plane - that you would need to refocus to compensate for the filter thickness? ...

Of course you would need to refocus. The rays hitting the film from the rear of the lens focused at infinity are not parallel.

[JW Dewdney]

I don't want to get into a messy debate here, GPS - but - well, yes, they would be parallel - it's a lens with infinitely long focal length. It's got ZERO CURVATURE - right? It's not a PRACTICAL application obviously - but it illustrates my point. If you have a close look at the diagram I've prepared, I think you will see what I'm driving at.

[GPS]

Sorry JW, I missunderstood you - was thinking that you're speaking about a lens foucused at infinity.

[Ole Tjugen]

...If you have a close look at the diagram I've prepared, I think you will see what I'm driving at.

Thanks for illustrating what I was saying. :)

[JW Dewdney]

I'm just trying to make the case that longer focal length lenses would require LESS focus correction than shorter ones, not only RELATIVELY (according to focal length) - but in fact ABSOLUTELY. So - while I cannot tell you definitively what the rate of correction is for increasing focal length - it'd be ROUGHLY inversely proportional (i.e. where the correction for a 150mm lens might be 0.1mm - a 300mm might be something like 0.05mm (or JUST slightly more).

[JW Dewdney]

Thanks for illustrating what I was saying. :)

And, what, pray tell, Ole, was it that you were saying then??

I NEVER said there was no focus shift. Am I that poor a communicator? Come on.

[JW Dewdney]

Thanks for illustrating what I was saying. :)

What the drawing illustrates is that the thickness of the filter, while it is a factor, isn't the only one. And, IF (a big if) the focus shift correction is one third the filter thickness for ONE very special particular situation... it sure ISN'T that for all other situations.

[Ole Tjugen]

So- clearly, a lens whose design admitted a narrower cone of 'light' would clearly be less affected by this 'focus shift' phenomenon. I hope that makes sense...

That's where you go wrong. The focus shift is exactly the same in any case, but with a narrower angle the "Depth of Focus" is greater so it has less practical consequence. Just as depth of field increases when you stop down - which is essentially what you're doing.

Yes, when exiting the glass the light rays continue in exactly the same angle as before. But they're shifted back a little, as shown in your illustration. And that is the focus shift.

[JW Dewdney]

That's just a riot of mixed concepts, there Ole...! You've just introduced about three new models on TOP of the one we were just exploring (i.e. - depth of focus, depth of field, aperture, etc..). Actually - using a smaller aperture does NOTHING to change the angle of view in terms of making it smaller. If anything at all - it would make it wider...

the argument I am trying to make - and that I believe I've provided a FAIRLY successful, if crude, geometric proof for, is that focus shift cannot be accurately calculated, willy nilly, by simply using the credo 1/3 the filter thickness. I mean - if you want to be anal about it, just refocus. Of course, there is an equivalent focus shift for lenses in FRONT of the lens too - it's just that, at a repro ratio of 1:20 it would be about 1/20th of the 0.5mm or so required for the other situation.

Okay - that's it. I've blown WAAAYYY too much time on this petty issue for the night. I'm quite ashamed of myself! I'm leaving now...so you can have the thread to say what you want at my expense now...! Good night..!


That's where you go wrong. The focus shift is exactly the same in any case, but with a narrower angle the "Depth of Focus" is greater so it has less practical consequence. Just as depth of field increases when you stop down - which is essentially what you're doing.

Yes, when exiting the glass the light rays continue in exactly the same angle as before. But they're shifted back a little, as shown in your illustration. And that is the focus shift.

[GPS]

Jw, the 1/3 rule comes as a rule for glass of n = 1,5 refractive index and paraxial rays.

[JW Dewdney]

Jw, the 1/3 rule comes as a rule for glass of n = 1,5 refractive index and paraxial rays.

Can you tell me where you got that from? I'd be interested to see this..!

[GPS]

Can you tell me where you got that from? I'd be interested to see this..!

My good friend - Sidney F. Ray in Applied Photographic Optics, from Focal Press 1988. Otherwise any optics school book worth of its name.

[JW Dewdney]

You should ask him about that.... what his source was.. I'd love to hear. But anyway - in the meantime, yours truly has been doing a bit of nosing around in his books... and I THINK I found the source of the whole 1/3 thing... I'm referring, of course, to the BIBLE (Photography: Theory and Practice, by L.P. Clerc). But I found a FORMULA which appears to have been the source of this nonsense. I found this formula:

D=L/(3m∧2 + Lm/f)

where D is the distance of the 'focus shift' , L is the thickness of the filter, m is the reproduction ratio and f is the focal length....

Though I'm trying to figure out if it's the right one - it may be for the filter in FRONT of the lens... I'll have to get back to you on that one.

[Ole Tjugen]

typical glass has a refractive index of 1.5, which means that light travels at 1 / 1.5 = 0.67 times the speed in air or vacuum.

That's the source of the "whole 1/3 thing".

If the filter is in front of the lens it's the same thing, but it doesn't matter if distant objects are seemingly shifted by 1/3 of the thickness of the filter. Unless you're shooting macro, of course - in that case, it does matter.

[JW Dewdney]

Well - if ya wanna get technical, Ole - 2/3 the speed of light is... guess how much? The speed of light. But I understand what you mean. That's how the optical model works.

[JW Dewdney]

Well - if ya wanna get technical, Ole - 2/3 the speed of light is... guess how much? The speed of light, according to einstein's predictions. But I understand what you mean. That's how the optical model works.

[Helen Bach]

If I got my sums right, these two graphs show the focus shift for a behind-the-lens filter that has a refractive index of 1.5. The vertical scale on the left is the distance from the axis to the point at which the ray enters the filter divided by the distance along the axis from the front of the filter to the original focal plane.

The calculation applies for points on the axis, but it can be interpreted for other points - that is why I made the second, extended graph. I'll draw a little diagram to explain that tomorrow. Enough for tonight.

The tick marks on the first graph represent the relative distance from the axis as if it was an f-stop, if you see what I mean. It is supposed to illustrate that both angle of view and relative aperture have a part to play, though the former is likely to be more important than the latter.

Best,
Helen

http://gallery.photo.net/photo/5786985-md.jpg

http://gallery.photo.net/photo/5787026-md.jpg

[Ole Tjugen]

Well - if ya wanna get technical, Ole - 2/3 the speed of light is... guess how much? The speed of light, according to einstein's predictions. But I understand what you mean. That's how the optical model works.

The constant c is the speed of light in vacuum. The speed of light in other mediums is slower, and in glass with a RI of 1.5 it's 2/3 of the speed in vacuum.

[JW Dewdney]

If I got my sums right, these two graphs show the focus shift for a behind-the-lens filter that has a refractive index of 1.5. The vertical scale on the left is the distance from the axis to the point at which the ray enters the filter divided by the distance along the axis from the front of the filter to the original focal plane.

Wow.. where did you find THOSE??

[JW Dewdney]

I was just musing on the car chase scenario - the one where two cars are travelling at the speed of light relative to a fixed point. The rear car shines it's lights on the car in front. The light from it's beams is travelling at at C RELATIVE to both cars. So what is the velocity of the light beams of the rear car relative to the fixed point that we referred to..? The answer = also the speed of light.




The constant c is the speed of light in vacuum. The speed of light in other mediums is slower, and in glass with a RI of 1.5 it's 2/3 of the speed in vacuum.

[Struan Gray]

Johnathon, you need to go back and read Emmanuel's post. Helen's graph includes the corrections for larger angles (where sin(theta) is significantly different from theta), which is exactly the spherical aberration that Emmanuel mentioned.

Only massless objects can travel at light speed (in a vacuum), and they tend not to have headlights.

[JW Dewdney]

Having looked closely at both - I don't think they are related - at all. Emmanuel was suggesting that there was a spherical aberration in the optical assembly due to the thickness of the glass. That is very different from the focus shift varying with axial angle. Unless I'm mistaken. Perhaps Helen can correct me on this if I'm wrong...(!) What I'D read on the subject last night with respect to something resembling spherical aberration - is that a filter placed BEHIND the lens will introduce a barrel distortion, whereas a filter placed in FRONT - will give a pincushion distortion. The effect is pretty subtle, apparently, but it exists..! Interesting to know.


Johnathon, you need to go back and read Emmanuel's post. Helen's graph includes the corrections for larger angles (where sin(theta) is significantly different from theta), which is exactly the spherical aberration that Emmanuel mentioned.

Only massless objects can travel at light speed (in a vacuum), and they tend not to have headlights.

[Struan Gray]

Having looked closely at both - I don't think they are related - at all. ....

Perhaps Helen can correct me on this if I'm wrong...(!)

Helen's nicer than I am. I'll let her do it.

[false_Aesthetic]

Wow!

I can't believe my post generated 4 pages of responses and responses to responses and responses to the responses that were responding to the first response and responses to those responses responding to the responses that were responding to the responses for the response to my question.



So, if I understand correctly, putting a filter in front or behind the lens WILL change the focus...but not that much (esp if I'm focused at infinity and stopped down to F16 and beyond) It *should* also cut down on the flare I'm getting when shooting into the sun.


If I'm understanding this correctly, lemme know. If I'm not.... well, let me know too.


Thanks

T

[Remigius]

I was thinking about putting filters behind the lens recently, namely strong nd filters (#2 and/or #4). I plan to use them with long exposure times, but i assume if they are in front of the lens, stray light might have a noticeable effect, whereas behind the lens potential stray light would be absorbed by the bellows interior. However, I would not be able to focus WITH the filter for obvious reasons (but I wouldn't care for focus shifts either). Has anybody used such nd filters? if yes behind or in front of the lens?

[BrianShaw]

T... it really is incredible, isn't it? I get this kind of response all the time at work when asking what seems to be relatively simple questions. Thank God for geniuses!!!

One item that I got out of the discussion that you didn't mention in your summary: if you originally focus with the rear-mounted filter already in place, there is no "focus shift". :)

[Emmanuel BIGLER]

Well, ladies and gentlemen, for sure optics is sometimes a bit difficult to understand but we can manage this in a pleasant way.

We have in fact all ingredients to give a complete answer to the original question, but may be we should re-formulate again and again differently.

I should have mentioned that the image shift equal to t.(1-1/n) is the paraxial image shift. It is hard to understand that this paraxial image shift is independant from any focal length of the system.. actually, classical textbooks need a whole chapter to explain what a paraxial image is.

In addition to this paraxial image shift you have an effect similar to an additional spherical aberration. This is far from obvious and requires to be computed like Helen did.

In practice if you do not compensate for the paraxial shift, you are out of focus and this is a first source of blur. But if the filter thickness is small this effect might be small. Now if the angular aperture is high you get another effect as plotted by Helen, i.e. spherical aberration.

Bob Salomon is perfectly right and refers to the last model of 70mm Apo sironar digital which is designed to properly focus through a certain piece of protective glass that covers the silicon sensor array. If you want to use the lens on film, Linos-Rodenstock supplies a piece of glass equivalent to what is usually placed on the silicon sensor so that the results are perfect on film. The effect is similar to what happens in microscopy with the .17mm covering glass, Linos-Rodenstock people know their stuff and this is probably a good example for our discussion.
Note that you can use the 70mm Rodenstock lens on film witout the corrective glass. You will have to compensate for the focus shift, no problem, but you'll not compensate for the spherical aberration that has been introduced inside the lens to compensate what happens in the sensor window. So do not blame the manufacturer for this, you have to use the recommended glass in order to use the "digital" lens on film, or use instaed a lens designed for film ! But I am sure that film photographers will love this lens with medium format hand-held cameras (OFF-TOPIC here ;-)

For classical LF cameras, the additional spherical aberration can probably be neglected when the lens is stopped down to f/16-f/22, depending of the angle of view/focal length, and provided that you check your focus with the filter in, you can safely use it behind the lens. For a very critical application like with top-class digital imaging systems you should care for that.
Things become difficult if you want to use a visibly opaque infra red filter. Fo this reason, placing the IR filter in front of the lens is much better solution, since
1/ you focus without the opaque filter and you assume that your infra-red image will be located where the visible image is (not guaranteed ! this is another story !)
2/ you put the filter in
3/ since the object is at infinity the paraxial focus shift induced by the filter is nil, whereas behind the lens the focus shift could not be seen with the eye...
4/ the additional spherical aberration is nil.

[JW Dewdney]

Helen's nicer than I am. I'll let her do it.

Actually Struan - I think it would be really interesting to hear what you have to say about the information that's in the graph that Helen provided - and how it has anything to do with spherical aberration. Humor me!

edit: see my post to emmanuel, below. I think I've got us all on the same page now. But yes - I agree with you now - the effect is an 'effective' spherical aberration. This is exactly the point I was trying to make in my posts (please have a look, if in doubt).

thanks

[JW Dewdney]

So, if I understand correctly, putting a filter in front or behind the lens WILL change the focus...but not that much (esp if I'm focused at infinity and stopped down to F16 and beyond) It *should* also cut down on the flare I'm getting when shooting into the sun.



I think the reason there are so many posts is because there's a lot of miscommunicating going on here. What it sums up to is - you get 'focus shift' with a filter either in front of or behind the lens.

if in front of the lens; the focus shift is too miniscule to even be able to compensate for without a million-dollar laboratory setup.

if behind; too miniscule to worry about under most conditions - for most setups, especially perhaps when at hyperfocal distance. Focussing with a filter in place will alleviate this for all intents and purposes.

I'd probably stay away from focussing at infinity for all but tele landscape, aerial or astrophotography situations.

[JW Dewdney]

I was thinking about putting filters behind the lens recently, namely strong nd filters (#2 and/or #4). I plan to use them with long exposure times, but i assume if they are in front of the lens, stray light might have a noticeable effect, whereas behind the lens potential stray light would be absorbed by the bellows interior. However, I would not be able to focus WITH the filter for obvious reasons (but I wouldn't care for focus shifts either). Has anybody used such nd filters? if yes behind or in front of the lens?

You won't have any flare problems if you use a proper lens shade. A filter behind the lens will avoid this issue. But let's face it - who's going to take a lens off, put a filter on and put the lens back WITHOUT refocussing anyway? Moot point, kind of, right?

[JW Dewdney]

Well, ladies and gentlemen, for sure optics is sometimes a bit difficult to understand but we can manage this in a pleasant way.

We have in fact all ingredients to give a complete answer to the original question, but may be we should re-formulate again and again differently.

I should have mentioned that the image shift equal to t.(1-1/n) is the paraxial image shift. It is hard to understand that this paraxial image shift is independant from any focal length of the system.. actually, classical textbooks need a whole chapter to explain what a paraxial image is.

In addition to this paraxial image shift you have an effect similar to an additional spherical aberration. This is far from obvious and requires to be computed like Helen did.

In practice if you do not compensate for the paraxial shift, you are out of focus and this is a first source of blur. But if the filter thickness is small this effect might be small. Now if the angular aperture is high you get another effect as plotted by Helen, i.e. spherical aberration.

Bob Salomon is perfectly right and refers to the last model of 70mm Apo sironar digital which is designed to properly focus through a certain piece of protective glass that covers the silicon sensor array. If you want to use the lens on film, Linos-Rodenstock supplies a piece of glass equivalent to what is usually placed on the silicon sensor so that the results are perfect on film. The effect is similar to what happens in microscopy with the .17mm covering glass, Linos-Rodenstock people know their stuff and this is probably a good example for our discussion.
Note that you can use the 70mm Rodenstock lens on film witout the corrective glass. You will have to compensate for the focus shift, no problem, but you'll not compensate for the spherical aberration that has been introduced inside the lens to compensate what happens in the sensor window. So do not blame the manufacturer for this, you have to use the recommended glass in order to use the "digital" lens on film, or use instaed a lens designed for film ! But I am sure that film photographers will love this lens with medium format hand-held cameras (OFF-TOPIC here ;-)

For classical LF cameras, the additional spherical aberration can probably be neglected when the lens is stopped down to f/16-f/22, depending of the angle of view/focal length, and provided that you check your focus with the filter in, you can safely use it behind the lens. For a very critical application like with top-class digital imaging systems you should care for that.
Things become difficult if you want to use a visibly opaque infra red filter. Fo this reason, placing the IR filter in front of the lens is much better solution, since
1/ you focus without the opaque filter and you assume that your infra-red image will be located where the visible image is (not guaranteed ! this is another story !)
2/ you put the filter in
3/ since the object is at infinity the paraxial focus shift induced by the filter is nil, whereas behind the lens the focus shift could not be seen with the eye...
4/ the additional spherical aberration is nil.

Not to throw gasoline onto a dying fire or anything - but I think I found the link to any and all of the confusion in this thread. In your first post you are referring to a 'linear' or 'constant' focus shift referred to by textbooks - and then you refer to a secondary 'sperical aberration' seemingly independent of the focus shift. The point I want to make here is that these are, in fact the very same thing - as illustrated by Helen's graph. I was confused by people using the term 'sperical aberration' as it wasn't a property emerging from the parent optical system - but the filter itself. I suppose we could call it an 'effective spherical aberration' - but this was the very same effect that I was trying to describe in several of my posts on the matter - when I was trying to make the case for the 'focus shift' NOT, in fact, being 1/3 exactly - I was trying to suggest it varied with distance from the optical axis...!

So - anyway - hopefully we can all be in agreement now.

[Emmanuel BIGLER]

So - anyway - hopefully we can all be in agreement now.
Certainly.

About misunderstanding in technical things, we should remember what René Descartes used to say in his famous "Discourse on the Method of Rightly Conducting the Reason, and Seeking Truth in the Sciences"
1637.
http://www.literature.org/authors/descartes-rene/reason-discourse/chapter-01.html

and that the diversity of our opinions, consequently, does not arise from some being endowed with a larger share of reason than others, but solely from this, that we conduct our thoughts along different ways, and do not fix our attention on the same objects.

[Remigius]

You won't have any flare problems if you use a proper lens shade. A filter behind the lens will avoid this issue. But let's face it - who's going to take a lens off, put a filter on and put the lens back WITHOUT refocussing anyway? Moot point, kind of, right?
Well - I haven't actually tried that yet, but I assume with my Sinar F1, I can prepare everything, open the bellows, add the filter and close it again. If done carefully, the front and back standards should not move, and I don't have to remove the lens at all. Of course I would mount the filter holder beforehand.

[Remigius]

Addendum: Maybe my imagination is totally wrong (and, of course I'll just have to experiment), but I assume when having a filter that is nearly black (hence the image is nearly black as well), all light that finds a path to the lens besides the filter is potentially much stronger than the one that passes through it. The filter holder (lee in my case) doesn't close the opening completely, it always leaves a few slots for light to pass through. Although such light would have to be reflected on a few black surfaces, I imagine it could possibly be visible on the resulting image.

[false_Aesthetic]

You won't have any flare problems if you use a proper lens shade. A filter behind the lens will avoid this issue. But let's face it - who's going to take a lens off, put a filter on and put the lens back WITHOUT refocussing anyway? Moot point, kind of, right?

What kind've lens shade blocks out the sun when I'm pointing my lens at the sun? I'm also using a 3.0 ND (that's 10 stops) at dawn--I'll attach an image. Light is at a premium when I first focus, and there's def. not a way to focus with the filter on. Yeah, you're right, if I had a normal filter, I'd focus with it on already or again....


Just look at the picture and tell me if there's a lens shade that will help. If there is, tell me what it is and I'll go out today and buy it.


Anyways... Thanks for all your help.


T

[JW Dewdney]

T!!!

Didn't you ever read your John Hedgecoe books???? All photographs should be taken directly AWAY from the sun...! I won't even get into the 'rule of thirds'...!!!



What kind've lens shade blocks out the sun when I'm pointing my lens at the sun? I'm also using a 3.0 ND (that's 10 stops) at dawn--I'll attach an image. Light is at a premium when I first focus, and there's def. not a way to focus with the filter on. Yeah, you're right, if I had a normal filter, I'd focus with it on already or again....


Just look at the picture and tell me if there's a lens shade that will help. If there is, tell me what it is and I'll go out today and buy it.


Anyways... Thanks for all your help.


T

[Bob Salomon]

"Bob Salomon is perfectly right and refers to the last model of 70mm Apo sironar digital which is designed to properly focus through a certain piece of protective glass that covers the silicon sensor array."

No I am not. The new 70mm does not need a corrector plate when used with film.

All of the HR series do need the corrector plate with film as the cover glass over the digital sensor was computed into the optical formula when the HR lens series was designed. Remember, a piece of glass (like the cover glass of the sensor) changes the focus by 1/3rd the thickness of the glass.

[false_Aesthetic]

T!!!

Didn't you ever read your John Hedgecoe books???? All photographs should be taken directly AWAY from the sun...! I won't even get into the 'rule of thirds'...!!!




For the safety and well being of everyone (including myself) I'm going to assume you're joking.

[GPS]

T!!!

Didn't you ever read your John Hedgecoe books???? All photographs should be taken directly AWAY from the sun...! I won't even get into the 'rule of thirds'...!!!

Surely you're alluding to the third of the glass thickness, aren't you???

[BrianShaw]

For the safety and well being of everyone (including myself) I'm going to assume you're joking.

JW is probably joking... but I'm quite sure that Hedgecoe wasn't!

This "rule" gives me the willies - my Dad was a believer. All of my childhood pictures have me and my brothers squinting from staring into the sun. "The sun has to be coming from over my shoulder" always said my Dad. Looking at the pictures would lead the viewer to believe that we were born without eyes. It's quite creepy.

[JW Dewdney]

joking - of course! What sort of philistine do you take me for??

[false_Aesthetic]

JW is probably joking... but I'm quite sure that Hedgecoe wasn't!

This "rule" gives me the willies - my Dad was a believer. All of my childhood pictures have me and my brothers squinting from staring into the sun. "The sun has to be coming from over my shoulder" always said my Dad. Looking at the pictures would lead the viewer to believe that we were born without eyes. It's quite creepy.


Wow! If I ever start a series on people I'm gonna take their pictures that way. I think it could be BRILLIANT (WHOOO PUN!!!).

Again, thanks!

[Eric James]

joking - of course!

:rolleyes: This is the sarcastic smile - the internet equivalent of raising a big red flag over your head with bold white letters spelling out: "just kidding". I hate using the :rolleyes: , but I've offended more than one person by not making myself clear.

As for shooting into the sun, I think it's okay, provided that the orb is positioned according to the rule of thirds: :rolleyes:



Filters have an optical thickness that is different from the physical thickness (if that makes sense)...That's why lenses made for rear-mounted filters (like fisheyes) come with a neutral glass filter for use when no filter is needed - the filter is a part of the optical system!

Thank you for this, Ole.

[D. Bryant]

So, if I understand correctly, putting a filter in front or behind the lens WILL change the focus...but not that much (esp if I'm focused at infinity and stopped down to F16 and beyond) It *should* also cut down on the flare I'm getting when shooting into the sun.


If I'm understanding this correctly, lemme know. If I'm not.... well, let me know too.


Thanks

T

A filter behind the filter can or may change the focus. FWIW, I only use Wratten CC gelatin filters behind the lens with no ill effects.

Don Bryant