Re: Bessa L 15mm reviews?

["Henning J. Wulff" <henningw@xxxxxxxxxxxxxx>]

>Does anyone have any new reports on the Bessa L and 15mm lens?  I have a
>co-worker going to Japan and want to know if I should have him buy one
>for me.  Also, what is the import tax on cameras?
>
>Gary

The following are two reports I've posted in the last couple of weeks on
the Leica Users Group list (I'm not interested in the Bessa-L body, as I
have Leica M bodies):

1)

The light falloff is due to the same optical facts that to some degree help
explain why it is such a good performer in other ways. The design is a
retrofocus design (but not by a lot), which you can tell because the exit
pupil is larger than the entrance pupil. This also means that the lens has
less falloff than a symmetrical or regular wideangle lens, such as the
Hologon. A lens has light falloff that follows the cos^4 law. This means
that the very corners of a 15mm lens, where the light rays are normally at
55 degrees off the axis, the cosine is 0.5736. Take this to the 4th power,
and you get 0.1082, so that means that the corners get 11% of the light
that the center gets. This is not affected by stopping down. Stopping down
gets rid of the vignetting due to the lens mount construction, which
prevents the full opening from being seen by the corners. This 11% is all
that is left after the lens has been stopped down. In most wideangle
lenses, some effort is made to reduce this, which can be done to a degree,
but not much. The easiest way is to make a retrofocus lens, where the rear
exit pupil is made to be further away from the film plan than the focal
length. For a given aperture, the aperture then has to be larger as seen
from the back. That is the effect I was talking about above. If the exit
pupil is further away from the film plane, the exit angles to the corners
are not as sharp, and instead of being the 55 degrees (half of 110, the
angle that a 15mm lens sees) that would normally be the case, it may now
only be 40 degrees. This increases the cosine to 0.7660, and cos^4 to
0.3444. The corners now get 34% the light at the center, or about 1-1/2
stops less, whereas before they got over 3 stops less. The Hologon has
about 2-1/2 stops falloff without the filter, and about 1/2 stop falloff
with. My estimate is that the Heliar has about 1-1/2 stop or 2 stops
falloff.

If a lens is made 'more' retrofocus, ie, the exit pupil is made further
away from the film plane, you start having a lens like the Nikkor. It has a
large exit pupil which is further away than the whole mirror box, and the
angles the light has to travel are not far off the axis. Falloff here is
due more to the limits of the physical construction than the optical
formula, but there is quite little. On the other hand, the lens is huge,
the front element is gigantic, the possibilities for reflections and flare
are immense with all the glass and severe curvatures, and distortion
becomes extremely hard to keep under control. Sharpness, or more
specifically MTF values fall off drastically into the corners compared with
a more conservative design.

It would be nice if Cosina had made some provision for a filter for this
lens, even if you would need a large step-up ring to actually use one. Then
you could put a center filter in front for those times when you really
needed even exposure. I'll see about getting something made that would
accept one of my LF center filters.

So, with the Heliar you just have to accept some falloff. For general
pictorial use a gentler film than Velvia is called for. Wideangles don't
need contrast enhancing film, as they usually show much higher contrast
than normal and long focus lenses anyways. I would tend to use Sensia or
Astia, or better yet Kodachrome with the Heliar. Velvia helps with tele
shots where the range of image contrast is lower, and additional saturation
helps bring back what the atmospheric haze has taken away. BTW, the new
Ektachrome 100VS is very similar to Velvia, but has a bit more latitude,
and of course more speed. Actually, a good film with this lens would be the
Ektachrome 100SW, which is warmer and has gentler contrast. That should be
a great combination. With B&W film the corner falloff is not an issue,
especially with the chromogenic films.

For metering you just have to realize that the meter with this lens sees
more than the field of view of a 28mm lens, at least on the short
dimension. If you meter with this in mind, your only compensation should
have to be about 1/2 stop for the corner falloff (so your center gets
overexposed by 1/2 stop, and the areas 15mm from the center are
underexposed by 1/2 stop. This would look about right. You have to remember
that most 35mm focal length lenses show about 2/3 stop falloff in the
corners, and that is hardly ever objectionable. The Noctilux wide open has
over 3 stops falloff in the very corners, but that is due to the physical
mounting and element size of the lens.



2)

>It seem hard to believe that the BAS test laboratory found exactly the
>same results
>with both the Voigtlander (Cosina) Heliar 4,5/15mm and Leica (Zeiss)
>Super-Elmar-R 3,5/15mm.
>
>9,4/10 for the Heliar and 94/100 for the Super-Elmar regarding the
>optical performance.
>
>(9,6/10 versus 98/100 for the mechanical quality)
>
>For your information, the hereafter are the optical results
>BAS gave to some other Leica R lenses.
>
>19/2,8 = 98/100
>24/2,8 = 94/100
>28/2,8 = 96/100
>35/1,4 = 94/100
>35/2 = 94/100
>35/2,8 = 96/100
>
>No comment !
>
>I look forward to see if this is corroborated by other sources.
>
>Luggers, what are your experiences with that lens?
>
>
>Lucien

The numbers that BAS picked at the end may be the same, but the method
whereby they got there were likely quite different.

I don't have the 15 Elmar, so can't comment on its performance, but since
it is essentially the same design as the 15 Contax lens, which I have used,
I'll base my comments on that, and on the 15 Nikkor, with which I'm most
familiar, having had it for almost 20 years.

The Heliar and the Contax are very dissimilar lenses. The Contax is
strongly retrofocus, while the Heliar is only retrofocus enough to enable
metering. As a result the Contax has good eveness of illumination, the same
as my 15 Nikkor. The Heliar has fairly strong light falloff in the corners,
which some people will find excessive. I have not made densitrometric tests
on this, but I would suspect that the light falloff is about 2 stops at f/8
and smaller in the very corners. This compares to the 21 Elmarit ASPH,
which has probably just over 1 stop falloff. The Heliar also has a
different color cast than the other Leica lenses, being a bit cooler.
Ektachrome 100SW seems to be a good film with the Heliar, both through its
color rendition and the not as extreme contrast. Velvia does not do very
well, because it accentuates the corner falloff and the naturally greater
contrast range that wideangle lenses usually encompass.

After that the news gets better. The Heliar renders detail, both medium and
fine much better, and at higher contrast levels than either the Contax or
Nikkor. At the center the Contax and Nikkor are quite good, but still
nowhere near the level of the Heliar. Contrast is never as high, anywhere
in the field. This is understandable when you see the amount of glass that
the latter two have. Similarly, the flare levels of the Heliar are
significantly lower. Distortion is also lower in the Heliar, but both the
Contax and Nikkor lenses are quite good in this regard. The Contax seems to
have somewhat better flare control than the Nikkor, but not by a big margin.

As I mentioned in a previous post, the Canon 14/2.8, which I have also
tried, is a lot poorer than the Nikkor in almost every way. I also had a
quick look at the new Tamron 14, and it too has distortion levels which
make the lens completely uninteresting for me. I don't know how it does
w.r.t. resolution, contrast or flare.

In any case, the Heliar is definitely a competitive design. Whether you get
it or a Super Elmar depends on (besides which system you have) whether you
favour eveness of illumination over contrast and flare control and size.
Giving one number as a qualitative value for either of these lenses does
justice to neither and is almost silly. Almost like looking for the highest
rated lens, and buying a 180 Summicron as your first lens based on that
alone.

As an additional comment, I would like to point out that the 21 ASPH
betters the Heliar in every respect except size, price, and angle of view.
But then, if you want a 110 degree angle of view, the 21 just isn't an
option.

   *            Henning J. Wulff
  /|\      Wulff Photography & Design
 /###\   mailto:henningw@xxxxxxxxxxxxxx
 |[ ]|     http://www.archiphoto.com