I must point out a couple of things.
1. First about the constant K. Although each meter manufacturer can choose to use a different constant K, I have found that basically only K12, K12.5 and K14 were being used. The difference between K12 and K14 is about 1/6 stop so meter reading differences are not related to the constant K in use.
2. When you have different meter readings between cameras and meters it has to do more with the metering pattern of the cameras and meters. Most cameras are center weighted and meters are not. To compare a surface with the entire surface has the same reflectance (any tone, a gray card is fine) and the surface has to be evenly lighted. The surface must then fill the frame of the camera on entire measuring field of the meter otherwise they won't be the same.
My 2 meters give readings that are withing 1/10 stops of each other. My cameras agree with my meter withing 1/3 stop.
I agree with Chan Tran. At the end of the day, the differences between various makers, various K constant etc. normally do not go beyond 1/3 EV at most. That can be irritating and can have an effect on careful work with slide film (so much so that many light meters can be calibrated so that they agree with each other) but it's not a difference that justifies all this diffidence about the "objectivity" of a light measurement with an instrument.
By the same token light metering with a DSLR is normally not different than light metering with an SLR. Once a proper equivalence between the two ISO values is found (and the two ISO values will be close to each other) it is in principle possible to use the light meter of a DSLR just like one would use a SLR, if need be.
I disagree that the exposure indication of a DSLR should be "ignored" as its metering technology would be different. In fact, I would expect the metering circuits inside a DSLR to be exactly those of a SLR. The light is measured by a SBC somewhere in the pentaprism or mirror box to obtain a certain exposure value. The metering circuits "doesn't know" what's behind the mirror.
I would not exclude that "matrix" metering systems of a DSLR might take into account the dynamic range of the sensor, but that it's another "layer of reasoning" applied by the camera.
Comparisons of light meters is prone to many mistakes. In the video posted in this thread about the profiling of a Sekonic meter the photographer takes the spot reading while probably projecting some of his shade onto the grey card. Besides, the angle of reading of the incident light metering and the reflected light metering was different and that might have a small incidence in the reading.
These small errors in comparison methodology can lead to an excessive mistrust in the instruments. It's a bit like comparing different thermometers and say that they differ without taking into account the exact length at which they have to be immersed.
" A man with one watch knows what time it is, a man with more than one is never sure", the proof of the pudding is in the results. I have several cameras with TTL metering and several hand held light meters, each item of equipment used in isolation give excellent results, I don't compare one with the other because I don't want to end up in a rubber room on the funny farm.
My watches and my meters are matched quite closely but of course they are not exactly the same. I do know which one is most correct though although none of them is absolutely correct.
Originally Posted by benjiboy