low light shot of Sony NEX-6 vs Fuji X100

Started Apr 11, 2013 | Discussions thread
Pete peterson
OP Pete peterson Regular Member • Posts: 384
Re: Part 111: I found the metering to be the problem

blue_skies wrote:

Pete peterson wrote:

I did the shots again in aperture priority(with matrix metering, 400 ISO, Ex Comp 0) and they came out as before - ie dark. I noticed that he shutter speed (SS) was wrong. So with matrix metering the camera was setting the SS at 1/15 instead of the correct 1/6 hence why they are dark.

So I changed it to spot metering and the results were better the meter stopped down the SS to 1/5

Does this mean that the metering on the Sony has a problem. Fuji's matrix meter can handle this scene?

Which do you prefer the warm sony exposure (spot metered) or the cooler fuji? The real colour of the white toy is white.

Aperture priority spot metering                                                                   Sony NEX-6

Aperture Priority Matrix Metering                                                               Sony NEX-6

As you can see there is a big difference in results between the metering.

Below are the fuji equivalents

Aperture Priority, Matrix Metering                                                                        Fuji X100

Aperture Priority, Spot Metering                                                                      Fuji X100

Two questions: do you have a grey card and do you use Lightroom?

The first part - a grey (or color) card - allows you to verify the metering algorithm under your lighting conditions, and you can see how the metering translates in an exposure. However, this only would confirm the spot metering (as the grey card would cover most of the screen).

The second part - Lightroom - is a useful tool to verify exposure and histogram. Lightroom has a mode in the "Development" module, which is named "auto-exposure". It basically will center the histogram, which means pulling the exposure up on the Nex image.

I took the liberty of comparing both matrix metered images and am showing them below. I let them speak for themselves.

I notice several problems here -

  1. Neither tool is 100% correct. The Fuji, the Nex, and Lightroom can all benefit from user intervention.
  2. The Nex is prone to 'highlight protection', but as you can see in the histogram under the Nex-6 OOC result, there is nothing that needs protection. So I do not understand why the image is so dark.
  3. To 3) However, I have noticed in low light images, that this ETTL behavior is common in matrix mode, and actually a better exposure IF you post-process it - the DR of the sensor is very high, and the shadows have a lot of details (without noise). 
  4. The LR auto-exposure correction on the Nex fails, imho, in Lightroom. To make the Nex image equivalent to the Fuji image, I have to add an additional +0.75EV to the slider to brighten the Nex image so it matches the Fuji image more closely.
  5. The OOC histograms are not the same - I wonder if this is in part because of slightly different FL and viewing angles (see window) and the matrix metering points picking up different measurement.  
  6. To 4) Notice the big spike on the Fuji in the yellow spectrum that the Nex is missing. 

As to which image I prefer - none of the OOC images are close to being correct. You confuse the matter even more by placing the white toy in the center in the Fuji spot metering mode, and the grey toy in the middle in the Nex spot metering mode. This will bias the metering, but even the white toy will take on the color of the incandescent light, unless the camera knows it is an incandescent light sourse (and its temperature).  Further, the matrix images use different FL and viewing angles, which may confuse precise measurements. If you look at the images as dark/light, the Fuji image, even though a wider FL, has relatively LESS dark than the Sony image, so it would expose to a different average value than the Nex. However, if the Nex sees more dark, it should exposure higher, not lower.  I can only suggest that the Nex measures a large DR between the brights and the darks, and that this triggers an ETTL to protect the highlights from clipping, even though there are none that needs protection. For a professional, the Nex image is a 'safer' image to work with.  Perhaps redo the experiment and use a tripod - to enforce same angle of view, same FL, and set both camera to incandescent, rather than AWB. I expect it might bring the results a bit closer, but I do expect that neither camera will give you what you want, incandescent exposures are tricky, and the Nex is not the best camera without post-edits.

Your images, as posted - OOC

Both images, after processing exposure in Lightroom using "Auto-Exposure". The Nex image has an additional +0.75EV exposure added to it.

[Edit] something wrong with DPreview - trying to add the image again, it does not show (last edit - 2nd try)

same image posted again, it does not show in dp-review, click on the link below to see it

-- hide signature --


Thanks for all this info I really appreciate it. A lot for me to work with.

I've done the pictures again this time using manual WB. It is a sunny morning now under this natural light the white toy looks off-white (not cream), and not as white as the card I used to calibrate the WB, that card actually seems to have a tinge of grey in it.

The pictures were taken under florescent light

Manual WB                                                                                                    Sony NEX-6

Manual WB                                                                                                     Fuji X100

As you can see the Fuji produced a blue tinge (at least I see that on my screen). I think the Sony's manual WB did a better job.

BTW your lightroom exposure on the Fuji is the closest colour to the toy as a see it in daylight

 Pete peterson's gear list:Pete peterson's gear list
Sony Alpha NEX-6
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow