et31 said:
I'm a theoretician. If 14.8EV is measured from the sensor, but only 12.3EV is usable, then I will correctly say that my camera has a 14.8EV max threshold in the limit as the SNR approaches zero, but only 12.3EV capable for effective use.
In fairness, I should acknowledge there's an off chance you actually
do believe DxO measured the D810 sensor at 14.8 stops of DR – after all, theoreticians are sometimes weak on analysis and interpretation of actual data. If that's the case, be aware that what they actually measured was 13.7 stops of DR. Accordingly, if you try to capture a scene with more than 13.7 stops, you're going to clip detail in the shadows, highlights, or both.
You're regurgitating the value they give for 'Print DR' which is two lies for the price of one, since they neither print nor measure an image. Instead, they calculate the DR of an image downsampled to an arbitrary 8 MP. Downsample further, DR artificially goes even higher (artificial in the sense that while DR does mathematically increase, you'll never recover the clipped data). Downsample less, DR artificially goes less high. Regardless, data outside of the actually measured DR (of 13.7 stops for the D810) are unrecoverably lost at capture. So, while you can state that your D810 has 14.8 stops of DR until you're blue in the face, it's not a
correct statement. Theorize upon that.
More likely, you do understand the above and are just flatulating along our approximate vector. In that case, I suggest you emulate Brave, Brave Sir Robin and run away. Scarper. Bugger off.
