Here you go. One image is exported out of LR from a CR2, the other after conversion to DNG. All other settings are equal. I've subtracted the two, they are pixel for pixel identical.
Hi Andrew,
I would not have expected anything less, since the conversion to DNG was also done with LR, the results should also be the same. Reading the original Raw
and converting it on the fly, or
converting it to DNG first in a separate step should not change the resulting data upon which the raw conversion in LR is based.
Both conversions should be identical
if done by the same application, but are they also the same when converted by one application (e.g. DNG converter) and read by another application, e.g. Capture One? That's the question that was raised, so we'd need to compare Capture One conversions of DNG versions with original Raws from which those DNGs were made. One would hope they result in the same conversion, but past comments from various users seem to indicate otherwise.
Of course not all anecdotal reports are from equally reputable / scrutinously analytical users who are also somewhat versed in Capture One, so maybe things are actually more equal, at least now, than they used to be. I could run a test myself, although it would only satisfy my curiosity if had nothing else to do, and I only would be able to test with files that I have, which is from a limited number of camera models, and mostly Canons.
It would be nice if someone associated with Phase One could confirm if DNGs and Original camera Raws are always going to produce the same results (assuming the DNG version is supported, like up to 1.4), that would save a lot of us a lot of testing work.
Cheers,
Bart
