Is my EM10's light meter off?

Joined
Apr 27, 2014
Messages
127
As some will know I've just bought an EM10, and I'm wondering if my camera's meter is off causing it to overexpose? I shoot in aperture priority and noticed that most of my shots had -0.7 or -1 exp comp. When composing the shot I use a combination of the look on the LCD, histogram and knowledge of the shot to set the exposure. My histograms in lightroom and aperture also look OK.

So I've just run a quick test, using wide (ESP) metering and centre weighted, and my 45mm f1.8, 20mm f1.7 and 14-42mm EZ, all using bracketing set to 0, -0.7 and +0.7 ev. On all shots, -0.7 looked the best, and this reflected in the histogram. I've tried both indoor and outdoor shots.

Have others found this, and is it a quirk of the camera, or is my meter off?

Cheers.
 
Joined
Apr 27, 2014
Messages
127
Try the test with Spot Metering so you know for sure what the meter is seeing.
Thanks for your reply. Going to sound a really silly question, but how would I know if this is over exposing the shot or not as surely it would depend on what I am metering off, so if I'm metering off something dark I would expect the rest of the frame to be overexposed and blown out? In the shots I have done it's just been general shots, such as my garden where I'd expect a nice even spread of luminance, but I've found that I get a strong peak and clipping in the highlights, and little in the shadows.
 

Rickf

Mu-43 Rookie
Joined
Apr 7, 2014
Messages
22
I've actually had an email discussion with a well known and well regarded photography blogger who noted that because the EM10 is targeted more for the consumer crowd (i.e. not the pros or serious amateurs) it is setup to over-expose a bit by default. I had noticed the same thing myself and had already changed the exposure compensation to -0.3 and sometimes -0.7. It was nice to get validation from someone else that I wasn't imagining it and there seems to be an explanation.

Rick
 

Growltiger

Mu-43 All-Pro
Joined
Mar 26, 2014
Messages
1,868
Location
UK
You really can't tell by taking normal photos with trees, sky etc. For example the camera may have tried to bring out the shadows and allow the sky to be too bright. But you may not want that, so it looks wrong to you.

You need to photograph something like an evenly lit brick wall. Then look at the histogram.
 

Replytoken

Mu-43 Hall of Famer
Joined
May 7, 2012
Messages
3,540
Location
Puget Sound
Real Name
Ken
I've actually had an email discussion with a well known and well regarded photography blogger who noted that because the EM10 is targeted more for the consumer crowd (i.e. not the pros or serious amateurs) it is setup to over-expose a bit by default. I had noticed the same thing myself and had already changed the exposure compensation to -0.3 and sometimes -0.7. It was nice to get validation from someone else that I wasn't imagining it and there seems to be an explanation.

Rick
That seems counter-intuitive for a consumer camera as they are usually biased towards underexposure as to prevent highlight blowout. Thom Hogan also mentioned this in a blog post (http://www.sansmirror.com/newsviews/january-march-2014-newsview/olympus-e-m10-camera-announ.html), so we now have reports of the meter being set to both underexpose and overexpose.

--Ken
 
Joined
Apr 27, 2014
Messages
127
You really can't tell by taking normal photos with trees, sky etc. For example the camera may have tried to bring out the shadows and allow the sky to be too bright. But you may not want that, so it looks wrong to you.

You need to photograph something like an evenly lit brick wall. Then look at the histogram.
OK thanks. What metering mode would be best for this? I did try a whole host of different scenarios though and all did over expose, the garden was just an example. Also, I didn't have any of the sky in. Normally I would put it down to user error but my Sony A77 does not expose like this for the same scenes under the same settings.
 
Joined
Apr 27, 2014
Messages
127
I've actually had an email discussion with a well known and well regarded photography blogger who noted that because the EM10 is targeted more for the consumer crowd (i.e. not the pros or serious amateurs) it is setup to over-expose a bit by default. I had noticed the same thing myself and had already changed the exposure compensation to -0.3 and sometimes -0.7. It was nice to get validation from someone else that I wasn't imagining it and there seems to be an explanation.

Rick
Seems odd, why would they design a camera not to expose correctly regardless of whether it's for pro or consumer use? I've had many consumer compacts all of which have exposed correctly, and my A77 which is an 'enthusiasts' camera. I'm not doubting your discussion by the way, just sounds odd that's all.
 

Rickf

Mu-43 Rookie
Joined
Apr 7, 2014
Messages
22
I cannot defend or explain the position being taken, I'm just passing along the comment because I had noticed the same thing. The actual quote was ". . . it’ll overexpose consistently compared to the E-M1 if you let it," and it was attributed to "consumer tastes" Take it for what it's worth.

Rick
 

BigStreetGuns

Mu-43 Veteran
Joined
Apr 29, 2014
Messages
222
Location
Serbia
Real Name
Ivan
Odd, my E-M10 underexposes on a regular basis by around 0.7 EV. I have to dial in +0.3-0.7 very often.
 
Joined
Apr 27, 2014
Messages
127
Odd, my E-M10 underexposes on a regular basis by around 0.7 EV. I have to dial in +0.3-0.7 very often.
Hmm, strange. Definitely not the same across all cameras then :confused:

I've had a thought about checking my metering. I've got a large grey card I can use. If the histogram's not central surely it would mean my metering's off would it not?
 
Joined
Apr 27, 2014
Messages
127
Here's an example of what I mean. Here's the same scene at 0ev, -0.7ev and +0.7ev. Just to my eyes -0.7ev looks best, but then when I load it into lightroom the histogram clearly shows blown/clipped highlights at 0ev and I need to bring back the highlights, whereas at -0.7ev I don't have to adjust exposure or highlights. This is just one example of many. The paving slabs were wet, but weren't particularly reflective and it was very dull, unlike what you might think looking at the 0ev exposure. If I'm being picky I'd say that -0.7ev is a touch too dark, and might be better at -0.5ev or -0.3ev, although if I raise the exposure in LR it starts to clip.

https://www.dropbox.com/sh/pi7qk2ye6vk6msq/AADK4cnrhZTZ8fJs0knWu2RKa

Edit: I think looking at the links on dropbox the pics look slightly darker than the originals so might be worth while downloading to compare?
 

MajorMagee

Mu-43 All-Pro
Joined
Feb 24, 2011
Messages
1,630
Location
Dayton, OH
That's a very difficult scene for the camera to meter, because you have very bright brights, and very dark darks, and not much in between. Getting the exposure "right" is just a matter of choosing what subject you want to properly expose for and letting the rest go where it may (unless you're able to add artificial lighting).

Alternately you could apply curves to the histogram to balance things more than they naturally were, or you could do an HDR conversion of the image using the three exposure brackets you've already taken.

Reframing your shot to exclude one end or the other of the dynamic range can help in these situations as well.
 

Replytoken

Mu-43 Hall of Famer
Joined
May 7, 2012
Messages
3,540
Location
Puget Sound
Real Name
Ken
Here's an example of what I mean. Here's the same scene at 0ev, -0.7ev and +0.7ev. Just to my eyes -0.7ev looks best, but then when I load it into lightroom the histogram clearly shows blown/clipped highlights at 0ev and I need to bring back the highlights, whereas at -0.7ev I don't have to adjust exposure or highlights. This is just one example of many. The paving slabs were wet, but weren't particularly reflective and it was very dull, unlike what you might think looking at the 0ev exposure. If I'm being picky I'd say that -0.7ev is a touch too dark, and might be better at -0.5ev or -0.3ev, although if I raise the exposure in LR it starts to clip.

https://www.dropbox.com/sh/pi7qk2ye6vk6msq/AADK4cnrhZTZ8fJs0knWu2RKa

Edit: I think looking at the links on photobucket the pics look slightly darker than the originals so might be worth while downloading to compare?
What were you metering off of?

--Ken
 
Joined
Apr 27, 2014
Messages
127
That's a very difficult scene for the camera to meter, because you have very bright brights, and very dark darks, and not much in between. Getting the exposure "right" is just a matter of choosing what subject you want to properly expose for and letting the rest go where it may (unless you're able to add artificial lighting).

Alternately you could apply curves to the histogram to balance things more than they naturally were, or you could do an HDR conversion of the image using the three exposure brackets you've already taken.


Reframing your shot to exclude one end or the other of the dynamic range can help in these situations as well.
Thanks, it could well be that and just a case that Im used to the Sony that meters how I would choose to do it, and the Olly does the other end of the spectrum so not wrong per se.

However, have you checked the histograms? At 0ev there are quite a lot of clipped highlights .
Now look at the -0.7ev shot, not only does the shot show more detail in the highlights, the highlights are still actually clipping ever so slightly in the blue spectrum., and there's no clipping ninths shadows, which you might expect if it was underexposed.

I'll take some more shots of different scenes but it just seems like Im consistently having to decrease exp comp to get the exposure I want. Now it might just be a mis-match between me and the camera, and it's no issue remembering to change the exp comp (easier than using curves IMO), but it just seems odd. As I've said already I have no such issues with my Sony so I don't think it's a case of user error per se. Also, I shoot in RAW so don't want to use the HDR function as that's JPEG isn't it?

In answer to the question re what I'm metering off, it should be in the EXIF if you download the file but I believe I was using ESP (whole frame) metering (can't check at the mo as I'm on my iphone).


Edited as got my histograms mixed up :redface:
 

Growltiger

Mu-43 All-Pro
Joined
Mar 26, 2014
Messages
1,868
Location
UK
OK thanks. What metering mode would be best for this? I did try a whole host of different scenarios though and all did over expose, the garden was just an example. Also, I didn't have any of the sky in. Normally I would put it down to user error but my Sony A77 does not expose like this for the same scenes under the same settings.
Testing it with difficult scenes will prove nothing.

If you fill the picture with an evenly lit subject, such as a brick wall, it won't matter which metering mode you use. I suggest you don't use spot in case the wall has inconsistencies.

Do this test and show us the histogram. I'll bet it looks normal.
 

MajorMagee

Mu-43 All-Pro
Joined
Feb 24, 2011
Messages
1,630
Location
Dayton, OH
Yes, the Histogram has large spikes on each end, and not much in the middle. That's a difficult situation for any metering system.
 

GBarrington

Mu-43 Top Veteran
Joined
Apr 4, 2014
Messages
634
Location
Springfield, Illinois
That seems counter-intuitive for a consumer camera as they are usually biased towards underexposure as to prevent highlight blowout. Thom Hogan also mentioned this in a blog post (http://www.sansmirror.com/newsviews/january-march-2014-newsview/olympus-e-m10-camera-announ.html), so we now have reports of the meter being set to both underexpose and overexpose.

--Ken
Well, I know my E-M10 overexposes by an EV of +.5. I've set it to underexpose by an EV of -.5, and now exposure is dead accurate for me!
 

Replytoken

Mu-43 Hall of Famer
Joined
May 7, 2012
Messages
3,540
Location
Puget Sound
Real Name
Ken
I would say that Olympus might have a QC issue with their E-M10 light meters, given all of the anecdotal stories and wide variations in readings, but I suspect that to properly test a meter for accuracy (which is different than testing as to how the matrix decides to evaluate a scene) the variables need to be reasonably controlled. This is not to say that these anecdotal stories are not correct, but it is hard to draw any definitive conclusions about E-M10's in general. Having said that, I generally want my camera meters to be repeatable. I can make a general adjustment to the meter, so long as it is consistent.

--Ken
 
Joined
Apr 27, 2014
Messages
127
Thanks for the replies. If the results are to be expected why do I then get different results with my Sony, I'm assuming then that different cameras assess the scenes differently?
I did a grey card test and the histogram was central as you'd expect. Been thinking more about that though, and obviously cameras make everything middle grey, so regardless of whether using a grey card, white card etc I'd expect the camera to make everything central, which it does. However, how this then translates into reading a scene I don't know, as to me (at least how my brain works and what I've become accustomed to with my other cameras) the Olly seems to read the scene wrong/differently.

I assume from the grey card test it's actually working 'normally' and I just have to get used to how the Olly reads the scene compared to my other cameras? It does seem strange to me at least that it reads all the scenes in a way that consistently blows highlights :confused:
 
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Copyright © 2009-2019 Amin Forums, LLC
Top Bottom