>>4305284Dxomark tests dynamic range with a tonal range chart, and says the average full frame camera can discern about 14 stops of tones at base ISO, and records the labeled/measured ISO discrepancy on another page so you can know if that DR wonder camera is actually high tech or if it's just ISO 50 pretending to be 100
Photons to photos takes a photo of the back of a lens cap or body cap and uses matlab or something to generate an SNR chart against the ISO labels. They say the average full frame camera has less than twelve stops of dynamic range. This is false, and would mean mft has less DR than a computer monitor! FALSE!
Does this tell you something about how their method, and data, pertains to the real world use of cameras? It is totally worthless for anything but identifying at which labeled ISO the sensor switches to a different gain circuit, which is info available in most other camera reviews and the users manual, so the best thing you can do about photons to photos is just no go to that shitty fucking website and never use their dumb ass charts.
If 4um/4chan gearfags really like to use screenshots from a site you can bet it deviates from photographic reality
Dpreview's studio shot tool for instance does not meter their cameras exposures, they used fixed exposure with a fixed flash setting and if a camera comes out darker they +exposure in adobe camera raw (which is one of the worst raw developers to exist). That's not how you'd shoot your camera, so it doesn't matter!
Most third party lens reviewers shoot a test chart close up and develop the raw with unsharp mask turned off, which is unusually brutal on lenses and doesn't reflect how they actually work on real subjects due to field curvature. They also use misleadingly plot "sharpness" as line widths x picture height, when people who actually design the lenses measure performance by contrast level at different special frequencies, which unlike lw/ph bar charts, actually tells you how the lens looks.
