| From: "Tim Munro" <[EMAIL PROTECTED]> | Subject: Re: Re:EOS 100-300mm f5.6L with Tamron 1.4x SF AF | >From: <[EMAIL PROTECTED]> | > Don't know why the camera thinks it is still 5.6 with the 1.4 converter; the | > 100-300L is an older design and the Tamron may not be communicating properly | > with the body. | | The Tamron 1.4x convertors do not have a chip to alter the information from | lens to camera, even the SP series convertor. That is why the maximum | aperture display, and focal length recorded, is always incorrect. The camera | exposure will be correct though without any compensation. The Tamron 2x | convertor does alter the info. Does any teleconverter have a chip in it? When I learned that the EF telextenders and 'telextendable' lenses had more contacts on them than a camera body, I assumed that Canon had taken the following design approach: "Use the extra contacts to encode the magnification of the converter, which simply grounds them in a pattern that encodes its magnification. Since there are three extra contacts, this could encode as many as eight different values (including, of course 'no telextender mounted'). The 'telextendable' lens' microcontroller reads the extra contacts and modifies the aperture and focal length data it sends back to the camera appropriately; it also sets its aperture from the camera commands taking account of the converter upon which it knows it is mounted. Data pass through the telextender itself without modification, and therefore no electronics are necessary therein." Older "EF" teleconverters that didn't have the extra contacts would of course mess up various aspects of autofocus and exposure. The newer ones were presumably simple to design and manufacture, needing just the extra contacts and some trivial wiring, because all the heavy computational lifting was provided for already in the telextendable lenses (basically the L telephotos). Now, some really enterprising telextender manufacturer _might_ be able to produce a telextender with a little ASIC or gate array chip that intercepted data streams in both directions and modified, in real time, the focal length and aperture reports/commands according to its magnification. And also, like the Kenko/Tamron extenders, avoid having portruding elements. Voila! an EF-series telextender that works with _any_ EF lens, can be stacked, and avoids goofing up any aspect of exposure control! Two possible flies in the ointment, however. The first is the likelihood that the lens communication is clocked serial, with the camera being the master (i.e., the clocking source). Depending on the details of the protocol, this might make it logically impossible for the translation of the messages to work, or to be transparent. The other potential problem is the likelihood that the third party manufacturer might not understand the EF lens protocol as well as they need to, and end up getting screwed by some new kink in that protocol that Canon throws into it somewhere down the road. This is of course what's happened with various third party EF lens manufacturers in the past (and maybe the future!). * **** ******* *********************************************************** * For list instructions, including unsubscribe, see: * http://www.a1.nl/phomepag/markerink/eos_list.htm ***********************************************************
