Telephoto Binocular Comparison

Viewfinder magnification is unrelated to what is considered a "Normal" view.
The eye has a view of about 45% horizontally, as does a 50mm lens, and that
is why the 50mm lens is considered normal. A few people consider the digonal
of the format to be the normal view. For 35mm, 24x30, that work out to @
44mm.

Divide 50mm (normal view) with the 60mm or 70mm you suggest, and you'll come
up with the approximate viewfinder magnification of your camera (@.70) . So
if your viewfinder was really 1:1, your normal view, with both eyes open,
would again be 50mm.

--
THE REAL LEICADDICT
"The Gonzo God of SnapShots"
"Nicholas O. Lindan" wrote in message
ink.net...
"Alan Justice" wrote

One can determine empirically what is "normal." Put a zoom lens (e.g.,
28-70) on a camera with a 100% viewfinder. Look through the viewfinder
with
one eye (camera vertical) and compare the size of objects to what you
see
with the other eye. Probably easiest on a tripod. To my eye, the size
of
objects is equal in both eyes when the lens is around 60-70 mm. Could
someone confirm this? Isn't that what should really be called "normal".

Leica M3's worked this way -- the finder was very close to 1:1. But it
was 1:1 even with a 135mm mounted to the camera, it was just that the
camera put up an itty-bitty frame outline of the 135mm lens's view.

If you have a Speed Graphic the same thing happens with the wire frame
finder:
it's always a 1:1 view, but with a longer lens the frame is farther from
the
peephole and so delineates a smaller area of the scene.

In a 35mm SLR camera the angle of view the eye sees in the finder
is constant and is a function of the magnification of the focusing
screen image by the eyepiece lens. This is why the view looks distorted
for wide angle and telephoto lenses. If the view through a camera
looks 1:1 with a longish lens it is because the viewing magnification
is low and the viewscreen appears smallish. If the view into the
finder looks right with a wide-angle lens then the viewing magnification
is high and the view screen looks big.

Using 1:1 in the viewfinder as the criterion for a normal lens will
result in endless argument from a Nikon F user, who will assert
40mm is 'about right', and a Canon Rebel user, who will insist the
Nikon guy is all wet and the correct focal length is 70mm.

All in all the view through the finder has not a shaved farthing to
do with what constitutes a 'normal' lens.

In determining the 'normal' lens, the lens's focal length, the size
of the final print and the viewing distance of the print all have
to be in kopacetic harmony.

Try it: Select a 4x6" print of a shot made with a 20mm lens. Looks, er,
'wide anglish', right? Now look at it from 3 1/3" (yup, that's inches)
away (it helps to look through a large magnifying glass unless one is
terminally myopic) -- the resulting 'view' is now both natural looking and
wide angle at the same time. For viewing from a more comfortable distance
of 3'4" the negative should be blown up to 4x6 feet! The angle of view
of the scene in real life and the angle of view when looking at the
photograph should be the same; then, and only then, does the picture
look 'right'.

And that's why most photos look better when they are printed big, really
big, especially if taken with a wide angle lens. It is not that
philistines
only care about the size of a photograph and not it's 'art'.
Photographers insisting
on 6"x6" 'images' mounted on two foot square mat boards are screwing the
puppy for their viewers -- the reason most folks ignore these examples of
self-proclaimed 'fine art' isn't because they can't appreciate the
artiste's
tremendous aesthetic sense -- it's because they can't properly see the
bleedin' 'image' in the first place, at least not without leaving
nose-prints
on the picture glass.

A 50mm lens on a 35mm camera will produce a 4x6" print with the correct
perspective if the print is viewed from a distance of 8 1/3", and all
in all this is about right and 50mm is considered 'normal'. If the
picture is taken with a P&S with a 35mm lens the correct viewing
distance for a drugstore print is reduced to 6". And, if giving
P&S prints of the kids to granny, who threads a needle at arms
length, a 12x18" print would be appropriate.

There is another definition of what constitutes a 'normal' lens, but
it has little to do with taking pictures, it has to do with manufacturing
cost. If the lens focal length is about equal to the image circle
then a nice optimum is reached regards to design complexity,
manufacturing costs and lens performance. The best lens for the buck
is to be had when it's focal length is equal to the diagonal of the
negative.
For 35mm the diagonal is 43mm, leading to the popularity of 45mm
lenses on rangefinder cameras of old where clearance for the mirror
is not an issue -- mirror clearance being the major factor extending
SLR lens focal lengths to 50mm. In the end it is not that 50mm or so
gives normal perspective, it is a matter of economics. Because of
this economic reason the standard print size is 4x6 (5x7" or 6x9"
is a better size for most folks, but again economics rules the day).

So, the correct "normal lens focal length" is really a function of the
size of the final photo and the distance from the photo to the viewer.

--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.

On Wed, 10 Dec 2003 10:23:29 -0500, "Leicaddict"
wrote:

Viewfinder magnification is unrelated to what is considered a "Normal" view.
The eye has a view of about 45% horizontally, as does a 50mm lens, and that
is why the 50mm lens is considered normal. A few people consider the digonal
of the format to be the normal view. For 35mm, 24x30, that work out to @
44mm.


Ah... what the eye sees is irrelevant. A "normal" lens is one that
reproduces straight lines as straight lines on film (i.e., no barrel
distortion, no converging lines, when the film plane is parallel to
those structures/lines)

And your above calculations are absolutely Wrong on several counts.
1. using your dimensions you'd get 38.4 mm as an answer
2. the dimensions of the 35mm camera's shutter opening is however not
as you've written but 24mm x 36mm
3. Your assumption is incorrect. you do NOT do the calculation based
on the diagonal of the shutter opening. What you really need to do is
figure out the diagonal of the SQUARE that is inscribed in the CIRCLE
of light that the lens is capable of projecting on the film plane.

So in 35mm what you want to do would be figure out the diagonal of the
36mm X 36mm SQUARE inscribed in the circle.

Which, using simple geometry principles, means that 36 squared plus 36
squared equals the square of the diagonal

So 1296 + 1296 = the diagonal squared

And the square root of 2592 = 50.91

Now you try. Do the same calculation for the 2.24 inch format and see
what you get as an answer. Hint: you should get about 80.6 mm
depending on how you round the metric conversions.

-- JC

On Wed, 10 Dec 2003 18:05:39 -0600, Jim Townsend
wrote:

J C wrote:

On Wed, 10 Dec 2003 10:23:29 -0500, "Leicaddict"
wrote:

Viewfinder magnification is unrelated to what is considered a "Normal" view.
The eye has a view of about 45% horizontally, as does a 50mm lens, and that
is why the 50mm lens is considered normal. A few people consider the digonal
of the format to be the normal view. For 35mm, 24x30, that work out to @
44mm.


Ah... what the eye sees is irrelevant. A "normal" lens is one that
reproduces straight lines as straight lines on film (i.e., no barrel
distortion, no converging lines, when the film plane is parallel to
those structures/lines)

So a 1200mm lens is a normal lens if it reproduces straight lines with no
barrel distortion ? And an 8mm lens is also a normal lens as long as it's
perfectly rectilinear ?

Do you have a link to a credible web site with this info ?


No I don't have a link. But think about it for a second. So what if a
zoom lens gets you closer to the subject? If it does not distort the
geometry, then it is also produces a normal image and can be
considered "normal" because it reproduces geometry faithfully.

The definition for normal lens is the *shortest* focal length that
does not distort the geometry, then this shortest length lens and
everything above this length also does not distort the geometry.

IF the 8mm lens in your example did not distort the geometry then yes
it is a normal lens. BUT for an 8 mm to be a normal lens the size of
the image projected on the film will have to be very small (5.65 mm x
5.65 mm to be precise).



-- JC

in article , J C at
wrote on 11/12/03 2:13 pm:

IF the 8mm lens in your example did not distort the geometry then yes
it is a normal lens.


So all rectilinear lenses are normal by your definition? Interesting... I
suppose a word like 'normal' will always have many definitions. But I do not
think you can claim your definition is "the definition".

BUT for an 8 mm to be a normal lens the size of
the image projected on the film will have to be very small (5.65 mm x
5.65 mm to be precise).

Is this another definition, or a clause of the first definition?

And what if you have a non-rectilinear lens which has an image projected on
film which is the diagonal of the film (or a square formed by the longest
side)? For example, a cheap 28-300mm zoom which has bad linear distortion in
the "traditionally normal" range of 35mm to 60mm?

It is interesting that so many people think that wide angles lenses are not
rectilinear. These cheap zooms have a lot to answer for!

in article , Jim Townsend at
ess wrote on 11/12/03 4:25 pm:

I have a link or two Open Google, type in 'normal lens 35mm'

I'll post some for those who can't be bothered googling:

http://www.tutorgig.com/encyclopedia...ds=Normal_lens

http://www.colloquial.com/photo/formats.html

http://www.kcbx.net/~mhd/2photo/view.htm

And to finish, a bit of debate, which brings in the issues of binocular
vision, and holding prints from certain distances, etc.:

http://www.photo.net/bboard/q-and-a-...?msg_id=000JXL

On Thu, 11 Dec 2003 10:25:30 -0600, Jim Townsend
wrote:

ones I browsed stated that 35mm camera lenses in the area of 50mm are what are
considered 'normal' lenses because they closely approximate what the human eye
sees.

That bit about "they closely approximate what the human eye sees"
should be considered a rather layman's definition and rather
imprecise.

Because, I don't know about you but even with only one eye open, I see
a wider field of view than does a so-called normal lens. So a normal
lens is NOT in every respect, what the eye sees.

Therefore the better definition is a lens that does not distort
geometry (field of view be damned).


-- JC

On Thu, 11 Dec 2003 17:07:30 +0000, Ben Micklem
wrote:

For example, a cheap 28-300mm zoom which has bad linear distortion in
the "traditionally normal" range of 35mm to 60mm?


The key concept there is that cheap = poorly designed lens.



-- JC

The "Nowhere Man" wrote
To a man who is not@real
who [unrealistically] wrote:

[Web sites] I browsed stated that 35mm camera lenses in the area
of 50mm are what are considered 'normal' lenses because they closely
approximate what the human eye sees.

Balderdash. They are 'normal' because that's the optimum point for lens
design: image circle = focal length. A 'normal' lens is the lens that
'normally' comes with the camera. A 'normal' lens has now become
a plastic-element 28-80 f4.5-6.9 atrocity. See if the definition
of 'normal lens' doesn't change to some sort of AF/IS zoom in the
next 80 years: Some fool will post "I read on a web site that
an Image Stabilization lens is a 'normal' lens because it can
see clearly even when being jiggled, just like the eye ...."

From our present '40-50mm lens for 35mm film' optimum point of
design arise all sorts of silly myths: "approximates the human eye";
"gives a 1:1 view in the finder"; "as foretold in Ezekiel"...

A 50mm lens does, however, provide a 'reality approximating' photograph
when printed to 4x6" and viewed from about 8" away. A wide angle
and a telephoto do not: it is not the lens that is not normal though, it
is the an inappropriate print size/viewing distance that creates the
sense of expanded or compressed perspective.

That bit about "they closely approximate what the human eye sees"
should be considered a rather layman's definition and rather
imprecise.

Absitively.

Because, I don't know about you but even with only one eye open, I see
a wider field of view than does a so-called normal lens. So a normal
lens is NOT in every respect, what the eye sees.

With one eye, my field of view while looking straight ahead is about
150 degrees. And that's because my nose gets in the way. With both
eyes open my field of view is 180 degrees horizontally. Vertically
FOV is 150, what with eyebrow ridges and cheekbones.

If one is to get what the eye sees then a 16mm Nikkor fish-eye is about
right. And, surprise, surprise, surprise, a fish-eye lens is just what
folks have sitting there on either side of their noses. Ma Nature played
it clever, though, she made a hemispherical focal plane for the lens and
then did a whole lot of signal processing so we thing there is no
distortion in our view of the world.

To recreate nature's true perspective normal view, a fish-eye image
is projected with another fish-eye onto a hemispherical screen - the
viewer sitting at the center of the hemisphere. See, for instance:

http://www.realsims.com/pop-737NG-Dome.htm
http://www.mew.co.jp/e-press/0204_0206/0212-03.htm
and a whole host of others.

Again:

[Web sites] I browsed stated [blah, blah, blah...]

Who do you think knows something about recreating 'normal eye
perspective' on a (nominally) 2-D surface -- some yo-yo on the
web or Boeing, Lockheed, Matsu****a, Phillips, IBM Research,
PARC ...

And, how would you like the pilot on your next flight
be trained -- does anyone think looking at 4x6" drugstore
prints is good enough for practicing visual landings?

--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.

"J C"

For example, a cheap 28-300mm zoom which has bad linear distortion in
the "traditionally normal" range of 35mm to 60mm?

The key concept there is that cheap = poorly designed lens.

No.

Cheap == cost optimized design - low manufacturing cost - low consumer price

In general, though, you get what you pay for. And something with a lower price
will perform poorly compared with something with a higher price [like nobody
knows this...].

A low cost design often takes far more skill and engineering man-hours to
pull off than a cost-no-object high performance design.

What do you think is easier to design:

An F-17 with a multi-million price tag

A plane with all the capabilities of an F-17 and the price tag of a Piper Cub

Cheap (as in low cost) doesn't always mean poorly designed or made. Sometimes
it means heavily optimized, extensively tooled and produced in very high
volumes - and none of that comes from something that is 'poorly designed'.

--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.

"Jim Townsend" wrote

Troll bait. Ker-Plunk.

--
Nicholas O. Lindan, Cleveland, Ohio
Consulting Engineer: Electronics; Informatics; Photonics.