PDA

View Full Version : E5 overexposing on legacy lenses


Tordan58
21st October 2013, 11:27 AM
Hi

When using my E5 with my scopes all frames get overexposed. I have benchmarked using the same scopes against other camera body as reference (E620, which produces correctly exposed frames and I know the EM5 did expose correctly). Checking the exposure times pairwise I find a bias of +1EV on average, within 1/3 EV tolerance. behavior is shown both in centre weighted and evaluative metering modes.

However, when switching to Live View mode on the E5 the frames now get correctly exposed!

Anyone here got an idea what lies behind this behavior? I can live with it but it's annoying to have to remember to offset 1EV.

/Tord
PS Using 4/3 digital lenses produces correctly exposed frames.

bilbo
21st October 2013, 11:38 AM
Do you close the viewfinder blind when using LV?

Tordan58
21st October 2013, 11:54 AM
No, I don't.

I switch to LV and magnification as a support tool to secure that the focus is spot on (when there is time enough to do so), and usually I don't take any pictures in LV mode, just watch, then turn off and compose the picture in the OVF. Sometimes I take macro shots from akward low angle using LV mode, but have not taken yet with the E5.

By accident I happened to take pictures in LV mode the other day and that is when I noticed they were underexposed (since I had set A mode and -1EV).

Bikie John
21st October 2013, 12:59 PM
I vaguely remember seeing something about this when the E-1 was new. Existing users of OM kit could get an adapter to mount their OM lenses on it, and there were warnings that metering at full aperture could give misleading results. I can't remember the details I'm afraid, there might be something buried on the forum.

I think the problem was something to do with the way the light reflected off the mirror and into the metering cells. So it makes sense that if you are using live view, where I imagine it is estimating exposure from the live sensor view, the problem is less likely to happen.

Ciao ... John

photo_owl
21st October 2013, 02:08 PM
I vaguely remember seeing something about this when the E-1 was new. Existing users of OM kit could get an adapter to mount their OM lenses on it, and there were warnings that metering at full aperture could give misleading results. I can't remember the details I'm afraid, there might be something buried on the forum.

I think the problem was something to do with the way the light reflected off the mirror and into the metering cells. So it makes sense that if you are using live view, where I imagine it is estimating exposure from the live sensor view, the problem is less likely to happen.

Ciao ... John

basically yes - easiest solution is to create a 43 baffle (cutout) to reduce the excess (it's not going to hit the sensor) light or, especially with fixed aperture lenses/scopes just dial in the appropriate compensation....I think I would stick to the latter.

it's a bigger issue with some lenses than others.

Otto
21st October 2013, 03:41 PM
Whether this is relevant I don't know, but I used to have the overexposure issue with legacy lenses on my E-620, but I don't get it on my E-M5. That would support the stray light argument I guess.

OM USer
21st October 2013, 04:12 PM
basically yes - easiest solution is to create a 43 baffle (cutout) to reduce the excess ...

I, too, have heard of people creating a black paper "iris" to reduce the image circle to four thirds image size when using legacy lenses. It just stops the extra light bouncing around inside the camera.

peak4
21st October 2013, 05:48 PM
basically yes - easiest solution is to create a 43 baffle (cutout) to reduce the excess (it's not going to hit the sensor) light or, especially with fixed aperture lenses/scopes just dial in the appropriate compensation....I think I would stick to the latter.

it's a bigger issue with some lenses than others.

I'm not quite understanding something here; I'm sure it's me that's being dim. :o

I can understand the reason for adding a baffle to increase contrast, and possibly flare, if there is spare light bouncing round the mirror box.

On the other hand if there are more photons proportionally getting to the exposure metering sensor than it expects to get to the imaging sensor, wouldn't that lead to underexposing the image rather than over exposing it?:confused:

bilbo
21st October 2013, 06:40 PM
...if there are more photons proportionally getting to the exposure metering sensor than it expects to get to the imaging sensor, wouldn't that lead to underexposing the image rather than over exposing it?:confused:

It's light Bill, but not as we know it.

Sorry, couldn't resist. :D

Tordan58
25th October 2013, 10:57 PM
Hi Bilbo,

Apologies but I am not following what you are trying to explain.

Since the E620/E520 produces correct exposure with the same lens, does this mean that the incorrect metering is caused by the design of the E5 mirror box?

I can understand that a baffle that cuts off light that would hit a surface outside the 4/3 frame would mitigate flares/reflections, however why it would result in correct exposure is beyond me.

Where should the baffle be positioned? Inside the adapter ring, level with the camera body?