MF resolution question

Lewis Lang wrote:

. . . . . .

Don't Pentax users have the ability via an adapter, to also use their 645 and
67 lenses on their 35mm film (and DSLR) bodies?

Sorry, forgot about those. I am not sure which lenses could do that, but probably
only a few.

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

Hemi4268 wrote:

However, in practice not even all 50 mm lenses for various 35 mm systems have
equal resolution

Actually they do. What you mean to say is not all 35mm bodies have the exact
same mirror focus position.

You could put ground glass behind the mirror, focus more accurately, and not have
the resolution difference be a matter of focusing error. However, I agree that
resolution is affected by focus accuracy, though it is not the only criteria.



A perfect lens will have the following resolution and depth of focus in noon
summer sun.

Have you ever owned a perfect lens? Who made it?



f-1 2000 lp/mm 1 micron of DOF
f-2 1000 2
f-4 500 16
f-8 250 64
f-16 125 256

You can see that at f-4, the mirror must be within 16 microns to be perfect.

A human hair is 400 microns.

Yeah, and funny how the digital camera manufacturers always try to use visible
human hair to prove that there is enough resolution. It might be acceptable, but
it is a poor measure of quality.



Typical standard for most good camers is about + or - 50 microns for the mirror
position. You have to get to about f-8 for everything to be equal.

Which would mean that unless one was really lucky, the chances are that everyone's
camera has a focus set-up that is off, including rangefinder systems. Also, would
it be more accurate to state that focusing error can only be eliminated in an SLR
when stopped down past f8?



Larry

Okay, so if I understand your f8 theoretical limit, then a Leica 50 mm f2.0, a
Nikon 50 mm f1.4, and a Vivitar 50 mm f2.8 should all give exactly the same
resolution rendering of the same scene when used at f8? Of course, you stated
"perfect lens", and I have never seen one, except maybe a pinhole (which is not
really a lens).

Somehow, I just don't think that in practice, all 50 mm lenses are the same across
different brands (or even within the same brand). The number of elements,
placement, and even coatings will affect image quality. Maybe it will be very
close in the centre, but there are measurable differences at the edges. This is
where MTF charts come in handy, since they point out the differences quite
clearly, and many also show f8 performance.

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

False

A man of few words. I like that.

Larry

Michael Scarpitti wrote:

. . . . .

An example, take an 80 mm lens for a Hasselblad, and use a mount
adapter to place it onto a Nikon SLR. Using the same film in the 35 mm Nikon,
and the Hasselblad, an identically sized slice from either film (24 by 36 mm)
should show exactly the same resolution (detail).

False

Okay smart ass, I actually did this test. The film was Kodak E100VS and inspected under a 10x
loupe, and then a low power microscope. The same Hasselblad lens was used on a Nikon body with
adapter, and on a Hasselblad body. Both were tripod mounted, and the scene was exactly the
same, though obviously the angle of view was greater overall in the 6x6 shot. While there may
have been a chance of focus error, or even differences between the quality of the roll film
and the 35 mm film, it was not perceptible in the loupe, nor through the microscope.

You are welcome to do the same test. Somehow I doubt you will try it. Prove me wrong.



However, in practice not even all 50 mm lenses for various 35 mm systems have
equal resolution, nor do all 80 mm lenses for medium format have the same
resolution. While the differences might be slight, or barely perceptible,
they will always be there.

True, contradicting the two prior statements.

Either your reading comprehension, or you just like arguing. I got the highest grades ever in
my logic classes when I attended college, so feel free to actually try to make a cohesive
argument from your erroneous comment.

For the benefit of others: The exact same Hasselblad lens was used in the first comparison,
except the camera body changed. In the second statement, my contention is that a 50 mm Nikkor
will not be the same as a 50 mm Leica, nor a 50 mm Canon, nor any other 50 mm of a different
make. I would never state that an 80 mm Nikkor is the same as an 80 mm Hasselblad, since the
construction is different. There are lots of MTF charts to show this.

Use that 80 mm Hasselblad lens on a Nikon body, and the resulting smaller piece of film will
be rendered the same as a 24 by 36 mm crop of the exact same lens actually used on a
Hasselblad body. As long as there is not a mount adapter, nor focusing error, that one lens
will render the same resolution on the two different systems. FIlm flatness might be one other
issue, though in that respect, the smaller 35 mm film might actually do better. It would take
some seriously powerful measuring gear to spot any difference.

Seriously Scarpitti, you are most welcome to prove me wrong. In fact, if you present a logical
argument, and you prove me wrong, I will even reply and point it out. I'm waiting . . . . . .

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

Gordon Moat wrote in message ...


What I thought, perhaps erroneously, was that we were talking about a
50mm Hass lens (a WA design) and a 50mm Nikon lens (a standard
design), taking a 24x36mm section of the Hass lens image.

In that case, the Nikon lens will win, because of the smaller film
area that is covered.

Is that or is that not what we are discussing?

What I thought, perhaps erroneously, was that we were talking about a
50mm Hass lens (a WA design) and a 50mm Nikon lens (a standard
design), taking a 24x36mm section of the Hass lens image.

In that case, the Nikon lens will win, because of the smaller film
area that is covered.

Is that or is that not what we are discussing?


A 50mm lens is a 50mm lens is a 50mm lens. The image size is exactly the same
otherwise it would be something other then 50mm.

So a 50mm Hasselblad gives exactly the same SIZE image as the Nikon lens. True
though that the coverage is different.

The other issue is what is detail. Is it the look of the wedding dress or is
the fact the best man is in one of the photographs and not the other. One type
of detail describes the lace in the dress and the other type discribes who was
at the wedding.

Larry

(Hemi4268) wrote in message ...
What I thought, perhaps erroneously, was that we were talking about a
50mm Hass lens (a WA design) and a 50mm Nikon lens (a standard
design), taking a 24x36mm section of the Hass lens image.

In that case, the Nikon lens will win, because of the smaller film
area that is covered.

Is that or is that not what we are discussing?


A 50mm lens is a 50mm lens is a 50mm lens.

Complete drivel. It can be a wide angle, normal, or telephoto
construction, for med format, 35mm, or 16mm motion picture camera,
respectively.

The image size is exactly the same
otherwise it would be something other then 50mm.

So a 50mm Hasselblad gives exactly the same SIZE image as the Nikon lens. True
though that the coverage is different.


...and that's important bit....

The other issue is what is detail. Is it the look of the wedding dress or is
the fact the best man is in one of the photographs and not the other. One type
of detail describes the lace in the dress and the other type discribes who was
at the wedding.

the point hasd been made. address it.


Larry

A 50mm lens is a 50mm lens is a 50mm lens.

Complete drivel. It can be a wide angle, normal, or telephoto
construction, for med format, 35mm, or 16mm motion picture camera,
respectively.

Sure it could be wide angle, normal or tele. The bottom line is the image will
always be the same size regardless with any 50mm lenses. Coverage can vary but
this has no bearing on detail unless you want to say coverage equals detail.

..
The other issue is what is detail Is it the look of the wedding dressis
the fact the best man is in one of the photographs and not the other. One
type
of detail describes the lace in the dress and the other type discribes who
was
at the wedding.

the point hasd been made. address it.



So what is detail? Is it the lace in the dress or is it the fact that more
people are in the wedding party with the Hasselblad image vs the Nikon image.

Detail to me is the texture of the lace.

If detail to you is the guy on the end of the picture then we are trying to mix
apples to oranges.

Larry

Michael Scarpitti wrote:

Gordon Moat wrote in message ...

What I thought, perhaps erroneously, was that we were talking about a
50mm Hass lens (a WA design) and a 50mm Nikon lens (a standard
design), taking a 24x36mm section of the Hass lens image.

The original question started with 35 mm film to roll film (120,220), and just comparing the
same piece of 24 mm by 36 mm film. Since manufacturers are fairly consistent with film
production, the same type of film for 35 mm and roll film use would have the same resolution.
However, what the original poster seemed to miss was that enlarging a medium format image to a
standard print size involves less magnification. That is the factor that gives the higher
quality.

Given that someone might want to crop out a 24 mm by 36 mm slice from a medium format image,
then the only advantage to medium format would be cropping ability. Not that it is a bad
choice, but I think that as much of the film area as possible should be used when making
prints, and just crop for aesthetic effect.

So somewhere in the middle of the thread, Larry (Hemi4268) jumps in with the 50 mm lens thing.
Then he puts forth his theory about perfect lenses, and all focal lengths are equal (I am
summarizing). Anyway, both you and I disagree, because there are no perfect lenses (not even
Zeiss nor Leica).



In that case, the Nikon lens will win, because of the smaller film
area that is covered.

I sort of agree, since the Nikkor would have been designed to be optimal for the smaller image
circle. The distortion is also an issue on the Hasselblad (Zeiss) lens, but only at the edges.
If a 24 by 36 mm slice was taken from the middle of the Hasselblad 50 mm shot, it would be
much closer to the results from the Nikkor.

The easy way to confirm this is to look at the MTF charts for both. Just using a simple theory
as Larry has suggested ignores the practical consideration of lens construction. It is
disappointing that he choose to twist your comments, rather than replying in a more technical
manner.



Is that or is that not what we are discussing?

Sorry if I seemed out of line. My posting to Larry included mention (and my contention), that
not all 50 mm lenses are equal, and that MTF charts prove this quite well. While the results
at f8 can be quite close, careful inspection will still reveal differences.

When I added in a specific question to Larry about the 80 mm, it was due to my having actually
done that comparison. The Hasselblad (Zeiss) 80 mm is quite low distortion, and much better
than their 50 mm choices. I also feel that this lens is better than the Nikon 80 mm, except
for the f2.8 maximum aperture. Use the Hasselblad lens on a Nikon, and the resulting images
are slightly higher resolution. Again, a look at MTF charts will confirm this.

Larry's simple theory is not totally wrong, but it just does not hold up in practice. While
the differences in lenses of the same focal length may be small, they are there. Oh . . . and
Larry, in case you are reading this, if you find that perfect lens, let me know . . . I might
want to buy or rent one.

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

Oh . . . and
Larry, in case you are reading this, if you find that perfect lens, let me
know . . . I might
want to buy or rent one.


One almost perfect 1.2 lens I know of was a Switar 16mm motion picture lens
coming in at about 1200 l/mm at f-1.2.

The $150,000 Tropel stepper lenses comes in very close to perfect at about 4000
l/mm at f-1 in uv light.

Really, just the plain old Nikon 50 lens does a very good job touching about
633 l/mm in noon summer sun at about f-3.
Although, depth of focus is somewhat small at 9 microns.

Really, an optical bench is the only way to go in doing these test. True, many
lenses give way different MTF curves with some having very good high contrast
values and still others giving very high low contrast values.

Hasselblad lenses in general have very high high contrast values at low
resolutions. I would say almost 105% on anything below 40 l/mm. When you add
Verichrome Pan film with it's internal boost of 120% on resolutions below 20
l/mm, this combo gives one killer 10x10 inch 4x image.

The bottom line though is image detail (texture of lace in a brides dress)
only has 3 inputs.

Focal Length
Resolution
Distance

Film size has nothing to do with it. A larger film size only captures the guy
at the edge of the picture that otherwise would be cut off.

Larry