>>4298692>20 years of progress>fucking BILL CLAFF CHARTOooh lines SCARY
1: Bill claff generates these charts based on the signal to noise ratio of a downscaled image. Therefore, is he really testing "dynamic range"? No. He is charting the signal to noise ratio of a compressed image, and the DR cutoff is a line in the sand with regards to SNR, even though you may be able to recover usable detail past that in the original raw (not a squashed jpeg). Didnt you ever think its weird that every other source assigns FF cameras 14 stops, not fucking 11 point fuck? Claff's a fucking retard and so are you for using this idiots charts. He arbitrarily declares that a well known and well used 3 extra stops of DR isn't there.
2: Bill claff generates these charts based on labeled ISO, which has up to a full stop of variance between brands because ISO is based on jpeg engines, not raw exposure, and there are two separate standards neither of which actually need to be followed. Therefore, can you compare cameras based on this chart? No. Not even within the same brand sometimes.
3: Bill claff inconsistently detects forced NR. He loves to spot it on canon, but has ignored sony cameras that are known by everyone who's attempted to seriously use them to have "brittle raws" due to sony's forced post processing - baked in NR and vignetting correction. Raw histogram analysis actually shows almost every camera has forced noise reduction at most ISOs (except for some MF cameras, ie: fuji gfx50r)
Now, what does that "DR" difference look like in a real photo under test conditions? Just fucking noise. Grain. Who cares. The vast majority of the detail is still there.
Under normal use, like rough ETTR+shoot raw, not trying to save an image thats 5 stops underexposed, with NR in post, the 5d classic would perform 99% as well as contemporary cameras in 99% of photos and just be a little grainier. Eyeball test, it's maybe missing one stop or less of recoverable info.
DO NOT USE CLAFF CHARTS.
