AF adapter messing metering?

bawang · Joined: 26 Mar 2008 Posts: 548 Location: Indiana, USA

I got this adapter (Picture #1) about a couple of months ago and I hardly use it on my setup, Canon 20D. I noticed that the the chip is programmed at f1.4. In the beginning the chip was not making good contact (according to seller) as the f indicator kept blinking from 1.4 to 0. I was asked to lightly sand-paper to improve contact. I didn't have the courage to sandpaper, instead I mount/remount repeatedly to simulate sandpaper process. The contact in now good, it stays at f1.4. But my metering is messed up. I used manual setting mostly.

When this adapter is used, I have to compensate EV at more than -1, comparatively with a non-chipped M42 adapter to get similar exposure. I believe the adapter uses dandelion chip or at least it looked like it.

I noticed the seller also sells adapters with a different chip as seen in the bottom picture (Picture 2), but this one is always on Buy It Now while the one on top (Picture #1) is listed as auction. This made me more curious, and inclined to think that the one in Picture 1 (that I bought) is has issues.

Anyone else experience this?

Picture #1

Picture #2

Flor27 · Joined: 13 Sep 2007 Posts: 1195 Location: Paris, France

Hello;

I have both chips from the same vendor.

#1 is the new one, f:1.4 55mm
#1 is the new old, f:1.8 50mm

I have yet no problem about mettering on EOS 5D with Ee-S focusing screen, but I did have with large aperture lenses on EOS 350D (f < 1:2 ).

EXIF datas reported to the body by the chip have no influence so far with the mettering as it's done at real aperture.

I guess it's because MF lenses were designed for 24x36mm film, and on our cropped sensors, a bigger part of the image circle hit sides of the camera room, thus it can cause some all over ghosting with full aperture (sides reflect some light even if paint in black), tricking the mettering sensors that are located on side of the prism and I guess are less affected by those room's side reflected light ...

What do you think ?

Edit : Oh, and some people using MF lenses on Oly 4/3 DSLR are using some trick to achieve better constrasts because of that problem : there are puting a black carboard ring on the lens's back :

http://gfsnt.no/hexanon/

http://forum.hardware.fr/hfr/Photonumerique/Objectifs/panasonic-objectifs-monture-sujet_16974_4.htm#t2470279

bawang · Joined: 26 Mar 2008 Posts: 548 Location: Indiana, USA

Hmmmm.....it seems that on mine it messed up regardless of the lens f.

I got the same messed-up reading when attached to a Vivitar 2.8/90, Flek 2.8/35, Takumar 28, Takumar 2/50, Yashinon 1.4/50, Yashinon 3.5/28 and other M42 that I have. I couldn't try with a S-M-C Tak 1.4/50 as it is not compatible with that one.

When I use the non AF adapter, everything is fine, i.e. proper exposure.

Orio · Expire: 2012-12-04

It is a common problem with all older EOS cameras, when stopping down.
It is related to the focusing screen and the metering algorhythms (the EOS cameras do not meter the actual light when stopping down in AE or P modes, they calculate it from the open wide value, and somehow the calculations fail when non-Canon lenses are used.

Adding the EE-S screen (to those cameras that support it) fixes the problem almost completely.

THe most recent EOS cameras (from the 400D onwards) should be less sensitive to this problem, as the light metering system was changed by Canon, to become more conforming to the international standard.
-

bawang · Joined: 26 Mar 2008 Posts: 548 Location: Indiana, USA

Flor27 · Joined: 13 Sep 2007 Posts: 1195 Location: Paris, France

5D' Ee-S precision focusing screen has no split prism, it's just a plain mat screen but much better for manual focusing as blur occurs more quickly out of focus. But it's also darker at apertures < f:2.8 than the default Ee-A screen.