Don Weiss, that's a very interesting analysis of the
logic of how converters/extenders might work. Although
lacking your expertise on the electronics, I had
reached rather similar conclusions myself, and also
regard your option 1(a) as plausible. Here is a bit
more data to fit into the picture. There are AFAIK
seven Canon devices with forward and rear facing EF
lens mounts: the four extenders (1.4x original and II,
2x original and II), the Life Size Converter, and the
12mm and 25mm extension tubes. All have the basic set
of contacts at the rear, and the 12mm extension tube
has only the basic set of contacts at the front. All
the Extenders, the Life Size Converter, and the 25mm
extension tube have the extra contacts at the front.

The conclusion that a lens with the extra contacts
'thinks it is mounted on the camera' when there is
open circuit across the extra contacts is
irresistible. The simplest way for a device to signal
its presence to the lens would be by bridging some or
all of these contacts, and there are only four options
available, or eleven if bridging to some uninteresting
(e.g. 'earth potential') line in the basic set is also
possible. This does not sound like future-proofing,
since even if the original and II versions of the
extenders signal themselves as identical, there are
already four distinct devices to be recognised. A more
sophisticated device signature could be provided by
means of a ROM chip in the device, and this is of
course open-ended in terms of the number of devices
which could be recognised. Some degree of mixing of
the two approaches may even be possible, so finding
that one dismantled device contains no electronic
components does not imply that is true for all of
them.

Not only is Canon's lens-camera communication protocol
unpublished, but we do not even know for sure what it
communicates. I have heard a rumour that at least some
USM lenses can communicate focused distance
information back to the camera, but the rumour does
not say whether any use is made of this. It is
well-known that the protocol was extended when IS
lenses were introduced, resulting in a minor
limitation in the use of IS lens plus extender
combinations on legacy EOS bodies. Data readout from
an EOS-1V confirms for certain that focal length and
maximum aperture are adjusted for the presence of an
extender when a lens with the extended contact set is
attached. With a zoom lens, the reported focal length
is adjusted correctly throughout the zoom range. If
working aperture is expressed internally in terms of
stops down from the maximum, then of course this would
not need adjustment, but absolute working apertures
appear on an adjusted basis in the camera display and
on the EOS-1V readout. There are, or may be, other
subtle adjustments for AF speed and metering
corrections.

AFAIK the only Canon lenses with the basic contact set
that can physically be mounted on the extenders are
the three TS lenses. TS24x1.4 is actually a useful and
optically quite acceptable combination. The lens
operates correctly with the extender in place (it is a
MF lens, so you just have to remember to turn the
focusing ring more slowly :-)) but the reported focal
length, maximum aperture, and working aperture are
reported as if the extender was not present. The
metering inherently takes account of the 1 or 2 stops
loss of light because of the presence of the extender;
if any further correction is required, it appears to
me to be too small to be noticeable.

The 50/2.5 macro lens has an extended contact set to
connect with the set on the Life Size Converter.
Physically, these contacts appear to be identical to
those used between L-series lenses and extenders. The
LSC should in theory be optically equivalent to a 1.5x
extender with an extra 8.3mm of extension at the front
(this is an elementary optical calculation from known
data, not a piece of inside information!). In fact the
combination is reported as 70/3.2. The reported
maximum aperture is not adjusted for the focus
setting, either with or without the LSC, but remember
this is an AFD lens. The LSC will physically fit any
Canon lens, but although it gives excellent results
with the 50/2.5 for which it is designed, the optical
results with other lenses often show serious chromatic
aberration. It does work, operating both aperture and
autofocus mechanisms, although autofocus usually does
not actually lock on except with the 50/2.5. When the
LSC is used with a lens with the basic contact set, it
of course cannot be detected, and the effect is like
using an extender with a TS lens. Interestingly, when
used with an L-series lens with the extra contacts,
its presence is again ignored, although at least it
does not appear to damage the lens (this test is bad
for the nerves!). The conclusion is that there is some
basic difference between the extenders and the LSC
over how the extra contacts are used.

Now, two questions I cannot answer on the basis of the
equipment I have. First, does any other Canon lens
have the extra contacts apart from L-series lenses
designed for use with the extenders, and the 50/2.5?
The only plausible candidate is the non-USM 100/4
macro; if so, what devices do they link to? Secondly,
what do the extra contacts on the 25mm extension tube
do?

... RS

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com
*
****
*******
***********************************************************
*  For list instructions, including unsubscribe, see:
*    http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************

Reply via email to