Camera: Olympus Pen E-PL1, nearly 8 years old (wikipedia).
I have two lenses, 40-150mm (amazon) and 14-42mm (amazon). I change between them frequently, but am careful when I do so.
It's recently started playing up. What can happen is this: the image shows fine on the screen (this model has no viewfinder) but when taken, it is incorrectly exposed by a large amount, usually over-exposed. Here are two images taken today:
According to the meta-data, both of these images have the same settings (they were actually taken as part of a "burst", but the exposure issue can occur regardless of that):
Top Image:
Flash used : No (auto)
Focal length : 32.0mm
Exposure time: 0.017 s (1/60)
Aperture : f/5.0
ISO equiv. : 1600
Whitebalance : Auto
Metering Mode: center weight
Exposure : program (auto)
JPEG Quality : 75
Bottom Image:
Flash used : No (auto)
Focal length : 32.0mm
Exposure time: 0.017 s (1/60)
Aperture : f/5.0
ISO equiv. : 1600
Whitebalance : Auto
Metering Mode: center weight
Exposure : program (auto)
JPEG Quality : 75
Obviously, given the age of the camera, I'm aware of the possibility that this is its gentle way of telling me that it is time for it to retire to pasture. But I quite like the camera so I'm hoping that it's something like needing to clean the contacts between the camera and lens.
Added in Edit: As requested in comments, here's the output of exiftool -FNumber -ApertureValue -Lightvalue -ExposureTime -ShutterSpeedValue -ISO -ISOValue -BrightnessValue -ExposureCompensation
:
======== P5271839.JPG (normal image)
F Number : 5.0
Light Value : 6.6
Exposure Time : 1/60
ISO : 1600
Exposure Compensation : 0
======== P5271847.JPG (over exposed)
F Number : 5.0
Light Value : 6.6
Exposure Time : 1/60
ISO : 1600
Exposure Compensation : 0
No bracketting was set. The lens (as reported by exiftool
since I was swapping between the two frequently this afternoon) was the Olympus M.Zuiko Digital ED 14-42mm F3.5-5.6 L
. Looking through other images that exhibit the issue, they appear to be all with that lens.
For completeness, I dumped the entire output of exiftool
on both images into separate files and then did a diff
on the two. Apart from file name and the various dates, the differences are (<
is the over-exposed image, >
the other one):
< File Size : 655 kB
> File Size : 5.7 MB
< Special Mode : Fast, Sequence: 9, Panorama: (none)
> Special Mode : Fast, Sequence: 1, Panorama: (none)
< Preview Image Start : 572800
< Preview Image Length : 98426
> Preview Image Start : 5362912
> Preview Image Length : 590858
< Drive Mode : Continuous Shooting, Shot 9
> Drive Mode : Continuous Shooting, Shot 1
< Black Level 2 : 64 63 63 64
> Black Level 2 : 65 64 65 65
< Sensor Temperature : 46.7 C
> Sensor Temperature : 46.0 C
< Thumbnail Length : 1507
> Thumbnail Length : 5946
< Preview Image : (Binary data 98426 bytes, use -b option to extract)
> Preview Image : (Binary data 590858 bytes, use -b option to extract)
< Thumbnail Image : (Binary data 1507 bytes, use -b option to extract)
> Thumbnail Image : (Binary data 5946 bytes, use -b option to extract)
From that, you can see that they were images 1 and 9 of a burst (but I've seen this effect in single images). In comparing with other images, the sensor temperature is at the high end (the range seems to go from 20 to 50) but the difference is slight. I don't know what Black Level 2
means, but again comparing with other images says that those numbers aren't unusual. The most significant difference is the file size, but given the whiteness of the image that could be down to jpeg compression (I have images which exhibit the same issue that are larger than images which don't).
Added in edit (2018-05-28): From the comments it would appear that my underlying question, as in the title, is difficult to answer so let me pose a variation. Of course, if someone can answer the question in the title that would be best, but the following would still be useful to have answers to in the absence of that:
What could be causing this effect?
How could I test for a possible cause?
What could I do to remedy a possible cause?
I realise that because the effect is intermittent, if I take some action then it'll be hard to verify that that was the correct action. Nevertheless, it would be good to know some possible actions to take, given that I'd prefer to delay the "buy a new camera" option for as long as possible.
jhead
. I've just runexiftool
on them both and there's no significant difference between the outputs from the two images. I can post any relevant fields from theexiftool
output that you think would be useful, but posting the entire lot seems excessive. The exposure time reported byexiftool
is 1/60. – Loop Space May 27 '18 at 20:39