Re: 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
RAW files come from the camera itself, are (or )used to be the unprocessed photo site data, today usually incorporate some camera data processing, can only be displayed by running them through a software converter, and are not intended to look best — for example, they _must_ be sharpened, contrast is low, the colors are flat, etc.
A TIFF file usually is created even later in the process — at a minimum it is the result of a conversion from a raw file, and more often it is the end result of a post-processing workflow, so it can include all of the usual post-processing variables. It would be extremely unusual (and very inefficient) to display tif files on the web, as they are unnecessarily huge.
To some extent, I contend that the \"cooked\" (to use your term) files is provide the most interesting and useful results. I really don\'t care if the raw file from camera X is \"better\" or \"worse\" by some standard than that of camera Y. I care about what I can do with the file and the eventual final photographic output that I\'m striving for. While the desire to level the playing field and apply the exact same standard to each type of file is supposed to create a more objective comparison, it also leads to comparing things that have less real-world applicability, or from which that real-world meaning must be guessed.
I believe that the most interesting and useful comparison is between prints produced after the files had been taken through idealized post-processing workflow by expert photographers, in each case optimized to produce the best possible output from the files to be compared...
More than once I have learned extremely important things about what difference matter, what differences don\'t matter, and what can be done with files by such a test.
Dan
RustyBug wrote:
Regarding the diff @ jpg vs. RAW ... is there a similar (i.e. an internal processing diff exists) diff between RAW vs. TIFF @ pre-cooked RAW files from the mfr\'s?
I guess the question would really be is TIFF vs. TIFF an even more equitable (reduced variables) comparison than RAW vs. RAW ... in the same spirit that RAW vs. RAW is more equitable (reduced variables) than JPG vs. JPG?
Having not used TIFF yet, I\'m not versed on the diff @ TIFF vs. RAW (which iiuc, is not quite the same as raw). So, it would seem that if mfr\'s are \"pre-cooking\" the files differently, is it possible to get each @ \"uncooked\" files. I\'m just wondering if part of the diff we see in DR is because we are comparing different levels of \"pre-cooking\" that generates more DR via pre-cooking algorithms, among mfrs, i.e. suggesting that the starting points @ \"uncooked\" might be much more similar than when we see them @ \"pre-cooked\".
Feb 13, 2015 at 09:13 AM
Previous versions of gdanmitchell's message #12857768 « 5DSR Low ISO DR 1.5-2 stops better than 1DX ... »
