| From: "F. Craig Callahan" <[EMAIL PROTECTED]>
| Subject: Re: EOS: AF at f/5.6?
|
| Thomas Bantel wrote:
|
| > Yes. When you think about it, this is to be expected. The tube doesn't
| > have the 2 extra contacts (like the TCs) that are needed to transfer
| > the information about aperture changes. Because only lenses meant
| > to be used with the Canon TCs have these contacts at the rear end,
| > while TCs and tubes don't have them, it is not possible to get correct
| > aperture readout with more than one TC.
|
| Ok; the *lens* has 10 contacts (one contact is double-width). The two extenders
| each have an identical set of contacts on the end that connects to the lens: 11.
| So far, so good-- the 11 contacts on the front of the extenders connect with the
| 10 contacts on the lens.
|
| However, the EF25 extension tube has the same 11 contacts that the extenders do,
| so it also connects to all the pins on the lens. Also, the extenders and the
| tube all have identical contacts on the rear: 7 (including one double-width),
| the same as on the 28-70/2.8L. The EOS 3 has 8 contacts, so it matches up with
| the contacts on the extenders and the extension tube (and 28-70). BUT, the lens
| has contacts that the body does not use, but which both the extenders and the
| extension tube do use (or at least connect to).
|
| It would therefore appear, from visual inspection, that the extenders and the
| extension tube are passing exactly the same information to the body, *if these
| contacts merely pass through the information from the lens.* If the body is
| "seeing" the extenders, then the extenders must be telling the body of their
| presence in some way. No?
|
| There is something going on inside the extenders that is not going on inside the
| extension tube. Whether that something is a microchip in the extenders, or
| dead-end pin connections in the tube, or something else altogether, is something
| I can't answer without cracking these babies open--something I am not prepared
| to do at this time. ;-)
|
| Both my EOS 3 and EOS 5 have 8 contacts; my lenses that do not accept the
| extenders have 7/8 contacts, whereas the two lenses that do accept the extenders
| have 10/11 contacts. Physically there appears no way for the extra contacts on
| these two lenses to be utilized with existing bodies. This isn't terribly
| surprising--recall that FD lenses had an AE pin long before Canon introduced an
| AE SLR to the market; and they also had a "reserved" pin that, to my knowledge,
| was never used.
|
| The mystery continues. . . .
|
| fcc
Here's my 2 cents.
First, I was surprised to hear that your 25mm extension tube had 10/11 contacts
on it. My 12mm has only 7/8. (I use your contact counting notation, which
accounts for the fact that some interfaces have 2 contacts in the place where
others have 1 wide one.)
Second, my educated guess on the use of those extra three contacts. I think that
a Canon lens with 10/11 contacts uses the extra 3 contacts to sense when it is
mounted on a teleconverter. 3 contacts could be simply grounded in various
patterns to yield up to 8 different magnification codes. Any teleconverter needs
only to have some dead-simple wiring to ground the appropriate subset of
the 3 contacts. Note also that the special case of no teleconverter present
simply corresponds to none of the 3 contacts grounded, which the lens would
take to mean, in effect, a "magnification" of 1. The lens senses the magnification,
if any, and then knows how to report its aperture and focal length (as modified by
the teleconverter), and also how to respond to commands from the camera, which
has been told the true effective aperture by the lens.
Third, I don't know about the optical significance of the stacking order, but as
far as the data communications, the teleconverter should be in full 10/11 pin
communication with the lens. With my 12 mm extension tube and its 7/8
pins, the TC would have to be right against the lens. With your 25 mm tube
and its apparent 10/11 pins, it could be between the lens and the TC.
In any event, if my theory on TC-lens interaction is correct, when you stack
TCs, only the one on or closest to the lens will affect the reported effective
aperture. Canon could have made a much more complex TC, employing
custom digital circuitry, that would allow arbitrary stacking.
DGW
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
