Quick M9 vs 6x6 comparison

[Jan Brittenson]

Perspective is purely a result of relative distances to the camera. As long as the camera and all distances are unchanged, perspective remains the same. You pick your shooting position based on the perspective you want, then pick a focal length for framing.

 

[BradleyGibson]

I've rented an M9 and several lenses to try precisely this experiment (M9 vs MFDB). The speed and the size of the M9 and its lenses make for a very compelling offering.

I've been shooting with it for a couple of days now, and find myself getting quite used to the ergonomics.

The M9's IQ is stunning for 36x24 at low ISO, but does not compare favorably with my AFi-II 10 over a similar field of view. The AFi requires a lot more muscle and sweat to use, but I feel there is a reward for the extra effort. Also, the M9's mid-to-high ISO performance is closer to MFDB performance than small format, and thus, isn't that much of an advantage over medium format.

I'll be sending it back tomorrow.

But jeez, it's a nice, tiny package, even with Summilux lenses! Nice job, Leica.

My 0.02,
-Brad

 

[Rolo]

Perspective is a function of camera position, but detail improves with closer proximity and aesthetic with depth of field.

Given a fixed subject, such as a house or full body portrait, say, one would need to back way off with a 35mm body, or use a much wider angle lens, sub 30mm, to capture the height that wll be a given on an 80mm MF outfit. A 24mm lens would need to be shot at f1.4 for the same aesthetic; no 28mm can produce an equivalent to a standard 80mm MF lens at f2.8.

IMO, a wider MF lens, 50/38mm, nor a longer, 110 f2/250 f4 cannot be directly aesthetically matched because of the need to adjust camera position due to the letter box format of 35mm.

Retaining image quality will always be under pressure in a comparison, but of course, wonderful images are regularly produced with the M9.

 

[gero]

I still think I am right. I will do the experiment with my M8, a 15mm, 21mm & a 35mm; I place the camera on a trypod and without moving it, shoot three images with the three lenses. I crop the 15 & 21 images in the computer so that both have the same outer most objects in the frame as the 35mm and will find out.

I don't have my m8 working right now so I hope someone wants to do it as well.

It's all in Einstein's theory of relativity.

 

[mjm6]

Sorry gero, you are incorrect.

The distinction you are neglecting is the coverage angle with different formats. Given a single location, a 150mm lens on a 35mm camera will produce a much tighter crop on the film/sensor than a 150mm on a MF camera, but the relationship of the objects in the images will be identical.

If you are trying to match angle of coverage, you will need to adjust the focal length, but the geometric relationship between objects will still remain the same as long as you don't move the cameras position.

For example, you need a 300mm on 8x10 to match a 150mm on 4x5, to match the coverage of about an 100mm on a 6x7 (cm) camera, to match approximately a 50mm lens on 35mm film/sensor

All of these will have (almost) exactly the same relationship of objects in the image, but the relative images will be different sizes. Since they don't all have exactly the same aspect ratio, it is impossible to make them exactly the same without cropping one or more a little, but that is a minor difference.

There will be differences in the images having to do with the DOF and optical characteristics of the various lenses, but for all in tents and purposes, the images will be functionally the same, but the larger formats will likely have a lot more information recorded in the film due to limitations of film resolution, etc.

Even this is not assured, however, as the optical systems become more and more critical and difficult to manage as you get into larger formats. Larger lenses also are not as optically corrected as smaller lenses (larger in format, not necessarily in focal length!), which will result in more complicating factors.

---Michael

 

[mjm6]

Yes, and it is in that respect that I think you are wrong for the reasons I described! A lens with a 90 degree field of view on 35mm will render the same as a 90 degree field of view lens on say 6x9 or 5x7...the house and trees will look the same. The only difference will be the depth of field at a given aperture.

But, of course, I could be wrong.

This is absolutely correct.

The other way to look at it is a 90mm lens will produce the exact same image reproduction (magnification) on ANY film size you use. The only difference will be how 'tight' the image is framed. On a 35mm frame, it will appear to be a telephoto shot, and on 4x5 sheet film, it will be a mild wide angle lens.

Large format photographers fully understand this because the camera and lenses aren't married. I routinely used the same lens on various cameras from 4x5 inches up to 12x20 inches and occasionally larger.

On the 4x5, a 450mm is a telephoto, and on 12x20, it is a wide angle, but if you took a negative from the 12x20 and cut a 4x5 sheet out of the center of the image, it will represent exactly the same image you will get if you shot that lens with a 4x5 camera at exactly the same location.

I have found that the people who struggle with this concept are often single-format users who never needed to get their heads around the differences in reproduction that comes from switching formats. It used to be 35mm shooters primarily, but digital has brought these topics into the realm of even P&S shooters these days (full, vs. micro, vs. 4/3ds, etc...)

---Michael

 

[gero]

The other way to look at it is a 90mm lens will produce the exact same image reproduction (magnification) on ANY film size you use. The only difference will be how 'tight' the image is framed. On a 35mm frame, it will appear to be a telephoto shot, and on 4x5 sheet film, it will be a mild wide angle lens.

---Michael

Michael, this is true; but also true, is the fact that the relationship between the sizes of two objects (one close and one far) will change with different focal length's, regardless of the size of the sensor that it is recorded in.

The test would be eassiest with a wide zoom.

 

[mjm6]

Gero,

That is not correct. The relationship between the sizes of two objects is purely dependent on the geometry of the arrangement, i.e. the location of the camera to the respective subjects.

What you are forgetting is that most people think of lenses in the specific context of a relative geometry, not an absolute geometry. For example, you want to take a photo of a person, and generally that may be a portrait that mostly fills the frame. with a WA lens, you will move in closer to fill the frame with the subject. With a telephoto, you will move further back to fill the frame (to the same percentage filled, or to about the size in the frame). The movement results in different angles of coverage for the same subject.

When you stay in the same place and change focal lengths (zoom), your angle of coverage for the subject remains the same, and the magnification of that coverage changes. Since the angle of coverage is constant, the proportions of various subjects will be constant.

When you try the test you proposed, you will see this to be true. By all means do the test, it will help you understand how these relationships work, but you'll probably have to go through a bit of sketch pad to really understand how this all works out.


---Michael

 

[gero]

The longer the focal length the more the distance between far and near is compressed even if you don't change the point of view.

If I'd remember how to draw perspectives, I could explain in a drawing; if the point of view doesen't change, the vanishing point can be further away or closer so that the objects get smaller-quicker.

 

[doug]

The longer the focal length the more the distance between far and near is compressed even if you don't change the point of view.

If you make two photos from the same location of the same subject with two different focal length lenses on the same camera and enlarge & crop the photo made with the shorter lens so the framing matches the photo made with the longer lens, the perspective will be identical. Try it.

 

[gero]

Doug, I don't think so, the object far away would be bigger compared to the one closer to you (with the larger focal length).

I'd try it if I had my camera. :angry:


this is not a grate example but, with a telefoto, the girl in the back would be closer to the size of the girl in the front. (she would be more out of focus, but that is depth of field)

 

[MGrayson]

This discussion hinges on what is changing in the two scenarios and what is not. If you swap out a wide angle for a telephoto and don't move the camera, there is no change to perspective. If you swap lenses and move the camera to fill the frame with the same subject, then the relationship between far and near objects changes.

"Crop sensor" discussions get heated over differing assumptions. The full "equivalence" is surprisingly subtle, i.e., what would you have to do to get exactly the same print from two cameras with different sized sensors. From perspective considerations, they would have to be in the same location and the lens focal length would have to scale with the sensor size, but then the DoFs are different, so we have to adjust f stops, but then the shutter speeds would be different, so we have to adjust ISO.... it's a real mess.:loco:

Matt

 

[mjm6]

Gero,

I know how to do perspective drawing; I'm an engineer by training.

The example you cite is exactly an example of what I was talking about above; you are neglecting that to get the foreground to be comparable in size with different focal lengths, you would need to move the camera, thus negating the geometric relationship between near and far subjects.


---Michael

 

[mjm6]

Ok, here's an example...

The images taken with the M9 and an 18mm, 24mm, and 50mm lenses.

Here are the three initial images:

50mm

24mm

18mm

Shot at the same location, with the same exposure. Not on a tripod, but you'll get the idea.

Here are the same coverage crops...

50mm

24mm

18mm

Without the EXIF (or me telling you), it would be nearly impossible to tell which one was which, except that I may have moved slightly, or the falloff may be more apparent, etc.


---Michael

 

[mjm6]

Ok, to beat a dead horse...

Here is a composite of the three crops showing that the relative relationships between them are the same...

Since I did this handheld, they aren't EXACTLY matching (look at the orange bucket for slight misregistration), but they are pretty darn close, considering I had to bend over between shots to exchange lenses.



---Michael

 

[gero]

Michael, that looks convincing. I will have to experience it myself to be 100 but thanks for the help.

In constructing a perspective drawing, can't you choose a vanishing point that is closer or further in the plan before you start drawing the persp.? And doesent that change how fast the objects get smaller as they get further?

 

[mjm6]

Gero,

You can't pick vanishing points without some limitations. The most mportant thing to choose is the projection plane distance from the object. That effectively determine the amount of distortion that will appear in the image, and also the near-far size differences you mention. The further from the subjects the projection plane is, the more 'telephoto' the objects will feel with respect to each other, i.e. that compressed image feel.

---michael

 

[mjm6]

Here's a good page on perspective:

http://www.khulsey.com/perspective_basics.html


---michael