Re: [Leica] 50/1.2 or 50/1.5 or 50/2?!

["Henning J. Wulff" <henningw@archiphoto.com>]

At 7:21 PM -0400 8/4/00, Dante A Stella wrote:
>Here's something a little weird.
>
>I just got back my Canon 50/1.5 LTM (1951) from cleaning.  At some point
>I got it into my head that it was time to see if the transmission was
>higher through a 7/3 Sonnar type than other constructions.  I had
>already tested it against a clean 50/1.2 Canon and determined that there
>was no speed difference (in terms of meter reading), or at least one
>within half a stop.  Interestingly, with the same target (beige wall),
>the results were (with the meter set to 1600)
>
>Canon 50/1.5 @f/2 = Between 1/250 and 1/500 sec
>Nikkor 85/2 @ f/2  = 1/250 sec
>M-Hexanon 50/2 @f/2 = 1/250 sec
>M-Hexanon 90/2.8@ f/2.8 = 1/125 sec
>
>This is not the way it's supposed to work - all of them should read the
>same (or as equivalents).  Any ideas?  It shouldn't be a huge difference
>like that because the Canon is single-coated and the Hexanon multi.  If
>anything, the more modern glass should be brighter.
>
>I suspect the old Canon transmits more light, even stopped down, even
>without multicoating and LD glass.  Or it's really a 1.2 and the f/2
>mark is where 1.5 really is.

There are a number of things at work here.

1) f-stops are geometric calculations, that don't take into account the
transmission characteristics of the lens. Coatings, no. of air/glass
surfaces, internal reflections and type of glass all influence the amount
of light that makes it to the meter. Your Canon lens might flare more for
example, causing more light to reach the meter.

2) meters react differently at times to lenses with different
optical/physical geometries. If you find that when you go from the
second-widest stop to the widest stop there is less than a one stop
difference when metering, that can be partly due to the greater degree of
vignetting that faster lenses in particular have at their larger stops. The
angle of the metering cell can also be a factor, especially in conjunction
with 'shinier' metering spots on the curtain. Edge rays might have a
greater influence on the metering result than axial rays.

In the end, take pictures and measure the density of the central area of
the negative (or slide). That's the real test of the effective relative
aperture, or 't-stop'.

You might find that if you follow your meter's advice, that the density at
the center of your negative taken with the Canon 50/1.5 @f/2 as above is
less than that of the M-Hexanon @f/2, but the edges of the Canon shot are
denser or as dense. Also, you might find that if you take a picture of a
wall, it looks as if the Canon is more than one stop faster than the
Hexanon, but if you take a picture of a normal scene the amount of
information that the Canon returns is considerably less than that of the
Hexanon due to the (probable) greater flare of the Canon, meaning that a
lot of flare light had contributed to the film density of the wall shot,
but was not observable due to the lack of contrast in the wall.

Usually the geometric apertures are quite accurate, and are not a concern.
The other factors mentioned above, film and developing variability etc. are
much more important.

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