Interesting.
The flash sure looks like it fired in both files, the shutter speed was OK, aperture was normal, the metering scheme looks normal, according to the EXIF, but clearly there is no evidence of flash in your image. Very strange.
Normally I'd say your flash was dead, but if you see a flash from both the internal and external flash then obviously they are alive.
It is almost as if the shutter is totally out of sync with the flash.
However there are a few puzzling things:
- why are your out-of-camera K-3 JPG's so small (2 or 3 MB). The ones I've seen are generally larger - 5 to 13MB.
- the EXIF here still seems to be missing tons of stuff - eg where are the flash exposure data points?
As you can see from the attached text file (derived from from one of the sample JPEG's in the K-3 samples area, using exiftool v9.34 - view it with word wrap on
), there should be data about how the camera was metering every AE point, every flash metering point etc.
Puzzlingly, exiftool v9.39 seems to be skipping lots of data collected by the earlier
exiftool v9.34, for some reason. If you have the time, could you run your images through exiftool 9.34 and see if it reveals that the flash metering segments are actually recording anything?
You should see a large block of data that looks like this (except with non-zeros in it, ideally):
Flash Metering Segments : 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
I'm just trying to see if the camera recorded anything about how to meter the scene for flash.
Also if you try shooting an image with live view and flash, does it work then?
Just trying to figure out how the camera is viewing the scene.