> [EMAIL PROTECTED] wrote:
> - Undocked EF100f/2.8 (you just don't click the lens on the mount, so
> there's no electronic communication)
> ...
> All of them showed a linear difference of 3 stops tested by photographing
> a kodak gray card and comparing with a slide of the same card photographed
> with the 100f/2.8 USM and the EOS 30.
> ...

Hi Gerard,

Perhaps we should explain further how we proved that the difference was 3
stops. The original discussion was on the eosdoc mailing list when we were
writing up that document:
   http://www.eosdoc.com/manuals.asp?q=E7EMFlens
 
The first requirement was that we could meter through the SAME lens
electronically coupled, then uncoupled, on the same EOS body with the same
target. For consistency, I suggest a lightbox (or LCD monitor screen) and
Centre-weighted metering at ISO 100.

First meter the scene in Av mode with the lens wide open. Record the
reading.
(e.g. F/2.8 1")

Then meter the same scene with the same lens wide open, but electronically
uncoupled. This can be achieved using WJM's trick of partially dismounting
the lens on the EF bayonet, so that there was not electronic connection
between lens and camera. Be careful not to drop the lens or the camera!!!
And if you use an Image Stabilizer lens, make sure that the Image Stabilizer
is switched OFF (0).

The difference in meter reading (e.g. F00 1/8") indicates the metering bias
FOR THE CAMERA BODY.

To demonstrate that the metering is linear, we needed to meter with the lens
uncoupled AND STOPPED DOWN. This is easy if you can activate DOF preview
before uncoupling the lens so that the aperture blades remain stopped down.
If you do not have DOF preview, set M mode with your desired aperture and a
shutter speed of 2 seconds, and unlock the camera with the shutter released.

Cheers
Julian Loke

*
****
*******
***********************************************************
*  For list instructions, including unsubscribe, see:
*    http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************

Reply via email to