35mm lens with R3A 40mm frame lines

[tom_f77]

Darrel, I've been experimenting with the CV 28/3.5 without an auxiliary finder on my Rollei 35RF - 40 mm frame lines like your R3A. I use it for people when I want to include their outdoor environment. I frame with the 40 mm frame lines to get the people situated where I want them in the picture & I don't care about how precise I am in just how much of their environment isw included. I'm also thinking that I can get in closer without cutting out what's around them. My first roll isn't developed yet, so I'll let you know how it goes.

Not relevant to the R3A finder, but the finder on your 35RF is the same as the one on my R2, and the whole view is a pretty close match for the 28/1.9 - I know we've been talking about differences in true focal length in another thread but I think that this will be equally true for your 28/3.5.

Of course you don't get the parallax correction for focused distance, but if you're interested in what background you're including this is actually a good thing; use the 40mm frameline to position your subjects and the whole finder will still show the correct field at infinity.

You're onto a winner.

Tom

 

[Huck Finn]

Thanks for the feedback, Tom. I hate auxiliary finders.

The 40 mm frame lines do adjust when I focus the lens, so is this part of my picture framed by the 40 lines parallax corrected. That's been my assumption & I assumed that what's outside the lines is not corrected. Am I wrong?

 

[tom_f77]

Thanks for the feedback, Tom. I hate auxiliary finders.

The 40 mm frame lines do adjust when I focus the lens, so is this part of my picture framed by the 40 lines parallax corrected. That's been my assumption & I assumed that what's outside the lines is not corrected. Am I wrong?

No - I think you are right. But the paralax correction shows what's in the frame at the focussed distance - say if your subject is a flat wall.

But if you've focussed on say a person infront of a distant background then the corrected frames will show where the person will appear in the picture but the part of the background included will be unchanged from what the frames would have shown when you were focussed at infinity (ie the whole uncorrected finder, in your case).

This sounds a bit complicated for a minor point, but it does make a difference if you are trying to frame a close up subject carefully against a distant background!

Tom

 

[Doug]

No - I think you are right. But the paralax correction shows what's in the frame at the focussed distance - say if your subject is a flat wall.

Tom, I think your point here is an important one that is often overlooked. The parallax paradox! The correction in the viewfinder is only correct for objects at the focused distance... that's the distance at which the line of sight from the lens crosses the light of sight through the viewfinder frame. In a 3D scene, the relative positions of objects at other distances is still different from those two lines of sight. Trying to line up someone's head with the mountain beyond won't turn out quite right. It would line up perfectly with an SLR...

A trick: Just before the exposure, move the camera up and to the left a little so that the lens is now where the main viewfinder window was. That should get the parallax spot on, but you need to be careful not to twist or rotate the camera so as to retain the original composition.

 

[Huck Finn]

Hmmm . . . I thought that it was everything in the frame lines that are parallax corrected. At least that's what I see when I focus; the whole frame lines move. I assumed that everything outside the frame lines are not parallax corrected. I'm a confused slow learner. :bang:

 

[Roman]

May I butt in here & try to explain it?

Huck,
imagine you are focussing at something at infinity (eg. a mountain range) - there is no parallax correction going on in the finder, since at that distance the few centimeters between your lens and viewfinder are irrelevant.
Now you focus at something close-by - say, a person's head 2m in front of you - here, those few cm of difference in viewpoints will be a problem, so, the parallax correcting rangefinder moves the framelines a bit (to the right and downwards).
Now place that persons's head in front of the mountain range; focus on the person - the framelines will show where in the picture the person's head will be placed; but since the parallax correction is now set for 2m distance, yet the mountains are still at infinity (and the taking lens did not move...), the framelines will NOT show where the mountain range is placed in the frame - you'll have to focus on infinty again to see that (or guess from qhat you see outside of the moving framelines).
Did I make it clearer, ore even less understandable?

Roman

 

[Huck Finn]

Roman, I think I understand. I assume that if part of the mountains are still in the area of the frame, they will be corrected.

More importantly, if I use a 28 mm lens with 50 mm frame lines, can I use the 50 mm frame lines to parallax correct the area in the middle of the picture & the rest of the viewfinder to get a general idea of the picture for composition? (I'm assuming that the full coverage of the viewfinder is about 28 mm as in the R2.)

Thanks for your help.

Huck

 

[unsharp]

Now I'm learning! Or...

Now I'm learning! Or...

Can I jump in and ask?

Roman, as the frame moves to the right and down towards the lens to paralax correct the near focus, why does it not stay there always 😕 ?
I mean, the nearer the frame is to the lens, the better corrected it will be at all distances?? Or am I way out here? 🙂

Anders

 

[Doug]

Hmmm . . . I thought that it was everything in the frame lines that are parallax corrected. At least that's what I see when I focus; the whole frame lines move. I assumed that everything outside the frame lines are not parallax corrected.

Oh my, I hope we're not increasing the confusion; this is a visual subject, yet we don't have visual aids that might show what's happening more clearly. In a quick Google search I don't find a nice graphic explanation for our situation.

Parallax refers to the difference between two views of a scene due to a difference in viewpoint. The camera rangefinder itself uses parallax to triangulate the distance to the desired focus point. That's not a problem. 🙂 What can be a problem is the couple of inches between the lens location and the VF location causing unexpected results on film.

Due to that difference in position, the lens and VF inevitably see the world slightly differently, and the difference is proportional to the distance to the subject. At long distances, that two inches or so is miniscule in comparison, so parallax error is so small it can be ignored. The closer you get to your focus point, the more significant it becomes. With an accessory viewfinder on top of the camera, the distance between it and the lens is greater than the camera's VF... more parallax error. The same problem exists for TLRs too, as the viewing lens is in a different location than the taking lens. View cameras and SLRs and digital LCDs are immune because the viewfinder "sees" through the taking lens, same as the film/sensor does.

Imagine you're lining up to shoot a pic of a flag hung on a wall 6m away, and at that distance your lens will just capture the whole flag but nothing else. Focused to infinity, the VF framelines are looking parallel to the lens but from a position 5cm up and left of the lens. If you frame the flag exactly in the VF, the film will record the flag missing its upper left 5cm, and show 5cm extra at the lower right. Parallax error!

As you focus the camera to the correct distance of 6m, the VF framelines shift down and to the right to show the area that the film sees. The view is still from a point 5cm away from the lens, but the framelines are just guiding your eye to look a little down and to the right so as to see the framing of the lens at that distance. Now both the VF and the film are looking at the whole flag... from a very slightly different angle. So there's still parallax, but one of the effects has been compensated for.

Roman, I think I understand. I assume that if part of the mountains are still in the area of the frame, they will be corrected.

We've been talking about a single flat subject ... adding in other objects at varying distances throws a monkey wrench into the parallax compensation. Remember the correction is only correct for that one object at the focus distance. The lens and VF are still seeing the scene from very slightly different viewpoints.

You can get a feel for this by holding your index finger pointing upwards at arm's length in front of your nose. Close your left eye and line-up your fingertip with some object across the room. Now, keeping your finger in the same place, close your right eye and open the left. Your fingertip isn't lined up any more, and that's parallax. Things at different distances need different compensation for parallax, more for closer objects. If you have objects at several distances, you cannot compensate for all at one time.

If you have a person sitting 2m away, say, in front of the flag, it's not possible to compensate for parallax for both person and flag at the same time, and the the way the person lines up in front of the flag will always be a bit different in the view of the VF and the slightly differing view of the lens. If you move the camera a bit this way and that so the viewfinder shows one stripe of the flag exactly centered on the person's head, the fact the lens is 5cm lower and to the right means the picture will show the stripe slightly displaced from the alignment you saw in the VF. If you are focused on the flag, the framing will be correct for the flag; if focused on the person, then the framing will be correct for the person, and slightly off for the flag.

More importantly, if I use a 28 mm lens with 50 mm frame lines, can I use the 50 mm frame lines to parallax correct the area in the middle of the picture & the rest of the viewfinder to get a general idea of the picture for composition? (I'm assuming that the full coverage of the viewfinder is about 28 mm as in the R2.)

Right, you can! If I understand you right... You can imagine some wider area centered around the 50mm frames, shifting in the VF as the framelines move for focusing, but I'd say this imagined larger area is going to be rather approximate, but it could help you avoid lopping off feet or heads... at the distance focused upon. (Not that for this purpose the focusing itself is important, but that the framelines move to compensate for parallax error at that particular distance.) This is what I end up doing with the 25mm Snapshot Skopar on the CLE, but I just try to imagine a modest additional area staying a set amount around, and following, the 28mm framelines.

 

[Roman]

Can I jump in and ask?

Roman, as the frame moves to the right and down towards the lens to paralax correct the near focus, why does it not stay there always 😕 ?
I mean, the nearer the frame is to the lens, the better corrected it will be at all distances?? Or am I way out here? 🙂

Anders

Good question...
Doug???

Roman

 

[tom_f77]

Good question...
Doug???

Roman

Doug is probably resting his typing fingers after that earlier effort. Until he recovers I suggest you do some experiments by tipping your head on one side, pretending one eye is the finder and the other the lens, looking at objects at various distances with each eye in turn and trying to relate your observations to what the finder frames do to try to tell you what "the other eye" is seeing.

I'm going to stop doing this now as my colleagues are looking at me funny.

Tom

 

[Doug]

LOL, Tom, good answer! I hope I answered Anders's question in my long diatribe...

Maybe I could emphasize that parallax compensation only addresses the framing error (and only at the chosen distance), and that nothing can really "correct" parallax because the viewfinder and lens are still in slightly different locations... inevitably they see the scene from slightly different viewpoints. If the lens is aligned to exactly straight-on to an object that has a bit of depth, say a box, then it sees only the front surface of the box, not any other side. But you are seeing the box from a slightly different position, through the viewfinder, some 5cm up and left of the lens. From this slightly different point of view, you are seeing the front surface of the box too, but from a slightly different angle that also allows you to see a sliver of the box's left side and top surfaces.

The camera's parallax-compensating framelines adjust in an attempt to show you correct framing at the distance you set, but still the world looks a little different to the lens! Due to its position, the lens sees that small bug crawling up the right side of the box that is not visible in the viewfinder... And the box is aligned a little differently in relation to its background.

 

[unsharp]

Thanks Doug

Thanks Doug

Thanks Doug,

I understand the Parallax phenomena from your excelent explanation, I just dont understand why the frames move to a closer position when you near focus, since they would be better there, closer to the lens, at any and all distances to minimize the parallax, or rather framing, error.

I'm a slow learner... 🙂

But then I thought of this theory:

All this so called "parallax compensation" is only to keep the lens from blocking the frames (or rather keep the frames away from the lens, and hence the true view)? Nothing else.

At closer distances the user is asumed to accept some viewfinder blocking in order to get good framing, so at close distances the frames are at their default position: as close to the lens axis as possible.

At longer distances the frames still don't cover all you get, so you might just as well move it away from the lens. What's within the frames will still be on the negative.

If this is correct then let's say that "parallax compensation" seams to be a flattering term to use... 😀 😀

But then again, I'm probably wrong. I have not read any of Leica's description of this feature and saying that Leica is not accurate describing the technical functions of their cameras? My life may be in danger.... 😀

Anders

 

[Doug]

I understand the Parallax phenomena from your excelent explanation, I just dont understand why the frames move to a closer position when you near focus, since they would be better there, closer to the lens, at any and all distances to minimize the parallax, or rather framing, error. ... 🙂

But then I thought of this theory:

All this so called "parallax compensation" is only to keep the lens from blocking the frames (or rather keep the frames away from the lens, and hence the true view)? Nothing else.

At closer distances the user is asumed to accept some viewfinder blocking in order to get good framing, so at close distances the frames are at their default position: as close to the lens axis as possible.

At longer distances the frames still don't cover all you get, so you might just as well move it away from the lens. What's within the frames will still be on the negative.

If this is correct then let's say that "parallax compensation" seams to be a flattering term to use... 😀 😀

But then again, I'm probably wrong. I have not read any of Leica's description of this feature and saying that Leica is not accurate describing the technical functions of their cameras? My life may be in danger.... 😀

Hi Anders -- Sorry, there is still misunderstanding here. You know how at very close distances, when you look at something close with both eyes, your eyes must cross somewhat. For distant objects your eyes can both look parallel to each other at the mountain... It's like that with the RF camera too, one eye being the VF window and the other eye is the lens: At close distances the camera's "eyes" must cross so they can look at the same object... When the VF frames move down/right away from the infinity position, it's changing the direction that "eye" is looking. This "cross-eyed" look gives similar framing for both VF and lens.

To directly answer your first question above... The VF eyepiece isn't moving back and forth, changing the amount of parallax error; that would be a fundamentally different thing. It's just the framelines changing the direction that you're looking through the center of the frame. For closer objects, the framelines are directing your view away from being parallel with the lens view, to a slight angle rightward and downward. "No" for the viewfinder blocking theory, and on the distant framing error... I hope I've clarified it a little more?

 

[Huck Finn]

Doug, your explanations have been superb!! I did the exercise with the findger & I really get it. Thank you so much; my understanding is improved.

How effectively do any of you think Zeiss/Cosina has been in designing a viewfinder that will parallax compensate for a 28 mm lens that focuses down to 0.5 meters (19.5 inches). they have told me that the camera effectively compensates for parallax at this distance, but after this explanation, I'm not sure. If you look at pictures of the ZI, it shows that the viewfinder window is located to the extreme edge of the camera as well, increasing the distance from the lens & therefore the parallax error.

Huck

 

[unsharp]

I see the light!

I see the light!

Thanks Doug, everyone, I think I finally got it! I well understand parallax phenomena, thats not the issue, I just could not buy anyone's explanation on how a rangefinder camera is compensating this exactly.

Short and easy:

The eye’s position is fixed against the viewfinder. The distance from the eye to the frame is creating a tunnel, and as the frame moves this tunnel will direct the eye’s angle of view!

The frame alone can not do this, as it is flat without depht. It is this relevant distance and the relative position of the eyepiece and the frame doing it. It’s like navigation, when you are entering a shallow port directed by aligned marks.

It is devine. God is into rangefinders. If he was not into rangefinders he would have designed our eyes to move towards the nose as we near focus 😀 😀

Anders

 

[Doug]

LOL, Anders; that's right! 😀 And that's why I'm now being more careful to call this "parallax compensation" and not "parallax correction." Nothing can correct the parallax except moving the camera a little so that the lens goes to where the viewfinder window was... but then the camera's parallax compensation leads to the framing being off!

I remember the "Paramender" product made for Mamiya TLR cameras, where the parallax error is entirely vertical. The Paramender fits between the camera and the tripod, and its lever raises the camera exactly the distance between the viewing and taking lenses. Then, with the taking lens in the same position as the viewing lens when you composed the picture, this perfectly corrects the parallax error...


The best the camera can normally do for us is to help show approximate framing for objects at the chosen distance. As RF camera users, I think we just need to learn to live with the parallax issue. If in our photography, parallax errors become troublesome, then the solution is an SLR or view camera.

 

[aizan]

or a digicam 😉