I've found various bodies to expose with various levels of error, no real surprise.
What I find more annoying tho, is an inconsistent error on any given body.
new Pentax K52s with fast zoom and sunlight.
- matrix metering is within 1/3 of Sekonic
- CWA and Spot on the same uniform surface are -1EV from matrix
I thot this may be due to the effect of lens corner shading in matrix EXCEPT that if I move to an indoor location, again on a smooth surface evenly lit by natural outside light thru windows, all metering modes are now the same result!
FWIW, my K52s underexposes considerably and inconsistently when outdoors in sunlight, no matter what metering mode I use and even sunny-16 numbers do not provide proper results but are often nearly 2 stops under... I need to get that thing checked out... I can usually rely on full manual giving consistent results but I have to go by the histogram as the metering's too wonky. Same behavior with various lenses so not sure what its problem is yet, aperture control lever calibration?.. Have to find some time for detailed testing.
My D800 does the same -1EV shift in sunlight when changing between metering modes as the K52s but its matrix mode does a very good job for all my shots and rarely under or over exposes by much in complex scenes and is predictable in low dynamic range shots.
My old Canon 5d2 often underexposed a great deal and again, inconsistently. It occasionally over exposed a scene grossly too, even with no change of scene and from shot to shot. Manual was the only way to get consistent shot-to-shot results.
my 60D and 7d gave remarkably accurate and consistently good metering (&WB) and also agreed very well with my Sekonics or sunny-16 when in manual. 40d gives consistent and predictable metering, even if not accurate.
All my consumer grade bodies, Canon and Nikon, actually seem to meter quite predictably and consistently, if a little conservative at times so they rarely clip highlites or underexpose by much.
So... not surprised to hear that a 6D underexposes a bit.
But how consistent is it between metering modes and various light levels? and shot-to-shot on the same scene?