Simply Sensor Size

[jaapv]

However, like most everybody else, I only use my RF with wide, normal and very mild tele lenses.

I do use my M6 with 200 and 280 mm lenses. And I am negotiating for a 400.On the Visoflex of course. Though, if it still counts a a RF then is a question 😉

Perspective of course depends only on where the camera is. However, with long tele lenses the issue becomes less critical. If you get something framed at (say) 100 yards with a 400mm lens, then walk forwards 50 yards and take a similarly framed shot with a 200mm lens, you'll be hard-pressed to notice much difference in the compression. Both shots will have a distinctly tele-type (!) perspective. Now if you walk in to 25 yards and shoot with a 100mm lens, you'll probably start to see the difference, if the thing hasn't flown away.

Sure you must walk quite a bit. The idea is to get your subject into the picture at the same magnification. However, with the subjects I take with long lenses, I am likely to get gored, trampled, stomped or eaten using a wideangle 😱 😀

 

[photogdave]

As the M8 will , given the lens mount, use 99.9 % CV,Zeiss or Leica lenses,ranging from excellent to exceptional, I would say that in this case, this is rather irrelevant.

By this time we were talking about DSLRs.

 

[rvaubel]

>>It even stays the same for a pinhole lens, that has no focal length at all. <<

I think a pinhole lens has the focal length of pinhole-to-film plane.

You might call a pinhole camera the ultimate zoom lens. Which goes to show you that perspective is totally determined by camera to subject distance. The "lens" just doesn't care what it's "focal length" is .

Even if LCT is wrong (and he is), I must admit for practical and artistic purposes we still act as if telephoto lenses compress and wide angles open up perspective. And more to the point, the M8 currently lacks any fast, wides to fill the gap.

Rex

 

[JohnL]

<snip>However, with the subjects I take with long lenses, I am likely to get gored, trampled, stomped or eaten using a wideangle 😱 😀

With that sweet little kitty I would surely approve use of the 400 !!! 🙂

 

[JohnL]

I do use my M6 with 200 and 280 mm lenses. And I am negotiating for a 400.On the Visoflex of course. Though, if it still counts a a RF then is a question 😉
<snip>

On the M8 your 400 will have the FOV of a 520 on 35mm 🙂
You are the first Visoflex user I have heard of for many a long year! I wonder how many there are on RFF?

 

[peterc]

"Full Frame" for digital cameras based on 35mm film cameras is APS-C or something in the crop range of 1.3 to 1.6.
This is dictated by physics.
Film doesn't really care what angle the llight is coming from. Digital sensors like the light to come straight on.
As you get to the edge of the 35mm frame the angle of light hitting the film/sensor plane becomes more acute. Lenses for film are designed to minimize the vignetting this can cause on a 35mm film frame. However, because sensors handle this vignetting less well than film, vignetting effects are more pronounced. Canon boosts the edges of the frame in software in the 5D to reduce the vignetting problem. However, with some "full frame" wide-angles vignetting is still quite pronounced in the 5D.
Lenses, particularly wide angles, would have to be redesigned to work acceptably on an FF rangefinder. A smaller sensor allows you to use current glass and get good results. It will force those who like wide lenses to buy another wider wide. The alternative is a larger sensor and buying pretty much a whole new set of lenses designed to properly cover the sensor.

Peter

 

[HenningW]

Hi LCT,
I think that you do have the compression factor involved in changing formats. I have been using 4/3rds format for several years and find myself drifting back to my favorite 90mm lenses, that I used with 35mm film. I noticed the same can hold true for 35mm compared to 6 X 7 MF where 90mm is a normal. The image compression gradient significance may have to do with our individual eye sight or mind's eye vision, the way we want the world to stack up. Changing formats might be a time to discover whether you want to "get it all in' or "have what you got stack up the way you imagine it should", when you make the picture.

Compression is a perspective phenomenon. It depends solely on relative position of the elements, and is not dependent on the optics in the camera nor the format.

Henning

 

[jaapv]

On the M8 your 400 will have the FOV of a 520 on 35mm 🙂
You are the first Visoflex user I have heard of for many a long year! I wonder how many there are on RFF?

Actually I started again after I decided to get the M8. The Visoflex3 will be rather nice for use on that camera and I am hoping for better quality than my Canon 10D. Not that that is a bad camera, not at all. I also got a VisoIII plus 200 and 280 for a steal. So I just now closed the deal for a Novoflex C 400, which is one of the very best 400 lenses ever built. It will have to be sent to the factory for shortening, but they still do that and they still sell the Visoflex adapters for the various lenses, which was a pleasant surprise. I have been shooting the 280 and 200 on my M6TTL for a few weeks now and it is rather nice to use, a lot better than I recall. The "soft" option for the mirror release is particularly useful. But then I had the Visoflex2 before, that did not have the quick-return mirror.
Anyone interested in a Canon 100-400L, mint,boxed ???😉

 

[photogdave]

A final word on distortion

A final word on distortion

Going back to my statement that wide angle lenses when cropped on a smaller sensor have more distortion, I was TOLD that I was incorrect, and given some interesting mathmatical formulas etc. to explain this.
Well, I failed math at school but I have been taking photos for a while. So I shot the same subject at the same distance. The first image is the 5D with the 28-135 IS @f4. The second image is the 30D with the 17-40 f4L @f4. No manipulation done. To my eye the wider lens on the smaller sensor clearly shows more distortion. I don't care if the physics are wrong but this is simply how it looks. End of story.

 

[HenningW]

After all is said and done, obviously any future M8 or 9er is worth a full frame sensor and should have it for the sake of the glorious glass that belongs in it.

Unfortunately the 'glorious glass that belongs on it' is not generally the current range of lenses that Leica makes. Especially the wideangle lenses are not suitable for a full frame sensor.

The 'glorious glass that belongs on it' will be a new series of digitally optimized lenses that will be sold at a 'glorious new price'. Or you could just use the Zeiss lenses.

Henning

 

[HenningW]

Going back to my statement that wide angle lenses when cropped on a smaller sensor have more distortion, I was TOLD that I was incorrect, and given some interesting mathmatical formulas etc. to explain this.
Well, I failed math at school but I have been taking photos for a while. So I shot the same subject at the same distance. The first image is the 5D with the 28-135 IS @f4. The second image is the 30D with the 17-40 f4L @f4. No manipulation done. To my eye the wider lens on the smaller sensor clearly shows more distortion. I don't care if the physics are wrong but this is simply how it looks. End of story.

Not quite. As you have focussed at a point that is fairly close, and obviously not at infinity, the focal lengths of the lenses are not necessarily what you think they are. From measuring the elements on the back rack I surmise that you are actually closer to the back rack in the second shot compared with the first. You do that to account for the wider angle of view that the second lens gives you, and that will increase the distortion evident in your pictures.

I also have the 5D and an APS sensor camera, the 20D. I don't have the 17-40, but the 16-35 lens which I do have has internal focussing which shortens the focal length as you focus closer. I don't have the 28-135 anymore, but the marked 28 focal length certainly wasn't 28mm; it was longer.

My point is this: if you have two lenses with focal lengths (in 35mm terms) of 27 and 29mm, you'll get quite different results and these will produce results such as you have shown.

Anecdotal: About 40 years ago I had to do a very large group shot (250 people) and the only equipment I had that would handle this was a 90mm lens on 4x5. As you might imagine, I had issues with distortion, and had to make sure that the thinnest people were in the outer zones of the photo. The group started getting antsy because they could see no sense in the way I rearranged people, and I didn't know how to explain to them politely that it was based on body mass. 😉

Now I take group pictures with a Noblex or Roundshot, and avoid any such distortion issues.

Henning

 

[photogdave]

All this goes back to my example of shooting a wedding with a 17mm lens to approximate 28mm full-frame, vs using a 28mm on a full-frame camera.
You wouldn't be focusing at infinity if you're doing a group shot in close quarters, and you wouldn't be measuring the alignment of your elements in relation to the imaging plane. You will have your camera, your lens and you will shoot.
From this experience I would rather use full-frame and a true 28mm lens rather than trying to approximate that focal length with a wider lens on a smaller-sensor camera.

 

[peterc]

From this experience I would rather use full-frame and a true 28mm lens rather than trying to approximate that focal length with a wider lens on a smaller-sensor camera.

The problem is that many current lenses would have to be redesigned to work with a so-called FF sensor ... they just wouldn't cut it as far as corner coverage goes.
The other option is to keep the current glass and add an 18mm designed to properly cover the APS-C sensor.

Peter

 

[JohnL]

This issue of whether current wide-angles work on a full-frame sensor is puzzling me somewhat. I have a 5D and also a Sigma 12-24 zoom lens, which IMHO performs extremely well on that camera. There is a little vignetting, true, and there is some loss of resolution towards the corners, but in both cases it's not noticeably more than you get on film. I don't see any real difference in performance over the entire range. If you stop down a touch, there is some improvement, but even wide open it's quite acceptable.

 

[nrb]

Unfortunately the 'glorious glass that belongs on it' is not generally the current range of lenses that Leica makes. Especially the wideangle lenses are not suitable for a full frame sensor.

The 'glorious glass that belongs on it' will be a new series of digitally optimized lenses that will be sold at a 'glorious new price'. Or you could just use the Zeiss lenses.

Henning

:angel:
glory doesn't come cheap
but I won't be paying a leg and a toe for a croppled thing
sorry about my bad english
nuno

 

[photogdave]

This issue of whether current wide-angles work on a full-frame sensor is puzzling me somewhat. I have a 5D and also a Sigma 12-24 zoom lens, which IMHO performs extremely well on that camera. There is a little vignetting, true, and there is some loss of resolution towards the corners, but in both cases it's not noticeably more than you get on film. I don't see any real difference in performance over the entire range. If you stop down a touch, there is some improvement, but even wide open it's quite acceptable.

I agree completely. I think most of the complaints about wide angle performance come from number-crunchers, rather than people actually taking photos.
Not to mention that many people are viewing their digital images at greater magnifications on their computer monitors compared to how they viewed their film images!

 

[Dougg]

Jaap, good group of web references to the perspective topic. Depth of field is another tough one... I was recalling Spotmatic-era Pentax manuals and brochures on Pentax lenses would illustrate each lens with a shot from the same position, each longer lens showing an ever-more-magnified view of the same scene. I would pour over these in trying to decide what lens I "needed" next. But this kind of series is also useful in showing that while the view narrows and is magnified, and the "compression effect" is notable, close examination teaches us that a crop of the wider illustrations has the very same perspective. I found some pdf files of old Pentax manuals, but the copying was so poor that the illustrations were all blocked up and hard to see clearly.

Interesting philosophical point about the pin-hole camera... the aperture disk is its effective "lens" and it indeed has no inherent focal length or angle of view or f-number, these being determined by the position and size of the sensitive material in the box.

 

[rvaubel]

To my eye the wider lens on the smaller sensor clearly shows more distortion. I don't care if the physics are wrong but this is simply how it looks. End of story.

They look about the same to me. Except, of course, the 30D's zoom lens is set for a little wider field of view. Is that what you notice?
Note that the perspective is the same as one would expect.

Rex
Who still can't see "compression"

 

[Finder]

What physics? As far as the wide-angle effect, three-dimensional objects stretching away from the optical axis, there should be no difference. As far as curvilinear distortion, that is related to the quality of the lens design. In current DSLR technology, the APS size sensor has a distinct disadvantage as the mirror box is scaled for a larger format. Given the same angle of view with wide-angle lenses, it becomes a greater problem to project the image to the image plane and greater curvilinear distortion can be a result. Since most zooms today are designed with price in mind, it is difficult to expect miracles.

 

[jaapv]

Not to mention that many people are viewing their digital images at greater magnifications on their computer monitors compared to how they viewed their film images!

A very good point. Few people using so-called 100% crops actually work out the magnification that entails for their particular camera, nor do they figure in the viewing distance to their monitor as related to an enlargement hanging on the wall