Archived posting to the Leica Users Group, 2001/11/15
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]
At 12:00 PM -0700 11/15/01, John Collier wrote:
>It is not that the Noctilux and Summilux are not true f/1 or f/1.4 lenses
>but that, under certain lighting conditions, the inherent wide-open
>vignetting will affect the metering area and you will not get a strictly
>geometrical shutter/aperture relationship happening. The meter is not being
>fooled if you do not follow it, it will underexpose. I have dug up the scans
>from the last time this came up and if someone would like them, just let me
>know.
>
>John Collier
As John wrote, the apertures of the Noctilux and Summilux are truly
f/1 and f/1.4, and if you follow the normal sequence, you will get
correct exposure at f/1 and 1/1000 sec if you otherwise get correct
exposure at f/2 and 1/250sec. However, this is only true at the
center of the frame. The further you go to the edges, the more
underexposed the shot at f/1 is relative to the shot at f/2. The
metering in effect compensates slightly for this, because the falloff
is strong enough to affect the total amount of light that hits the
white spot (which is what gets metered) and thus the meter expects a
slightly greater amount of exposure at f/1. If you check this out
carefully, and do the exposure series at a constant light level,
varying aperture and shutter as accurately as you can according to
the meter, you will find that the center of the shots at f/1 is
slightly overexposed compared with the other shots.
If you check your very fast SLR lenses, especially the normal to
wider ones, you will find the same metering anomaly. This applies to
full area, centerweighted and matrix metering; not spot except in
certain cases where off-center spot metering is possible like on
recent Canons.
If your camera does not seem to do this, the area you are trying to
meter is maybe such that f/1.4 at whatever would actually give you a
slightly overexposed (by 1/6 stop or less) shot, while f/1 at the
next higher speed would then want to meter just at the lower edge of
the meter's agreement.
In practice, this is all nothing to worry about. Just match the
little lights as you normally do, and shoot. True, at f/1 the meter
will ask for slight overexposure, but human tendency is to kind of
push things at lower light levels, and this compensates a little bit
for that, and besides, if you look at metering accuracy graphs,
you'll see that most exposure meters are not dead on the whole way
across the aperture range, and tend to underexpose slightly on the
darker end, which also compensates for the effects noted.
Lastly, this whole exposure thing is not as much of a science as is
sometimes made out (unless you are mainly shooting completely evenly
lit brick walls) and bracketing can be useful if greater accuracy is
required. Film can be off 1/3stop as well as processing, and these
factors are often greater than any metering discrepancy. The best way
to handle this is to practice with the film and processing of your
choice, in conditions that you normally shoot, and meter accordingly.
- --
* Henning J. Wulff
/|\ Wulff Photography & Design
/###\ mailto:henningw@archiphoto.com
|[ ]| http://www.archiphoto.com
- --
To unsubscribe, see http://mejac.palo-alto.ca.us/leica-users/unsub.html