By the way, the only reason (and that's a big one) for me to get into Cobalt would to get rid of Fujis profiles. I've really begun to dislike them because of the oversaturated highlights and all the usable simulations being weighted towards green.
Makten wrote:
By the way, the only reason (and that's a big one) for me to get into Cobalt would to get rid of Fujis profiles. I've really begun to dislike them because of the oversaturated highlights and all the usable simulations being weighted towards green.
Can you not tweak them to your satisfaction with the other settings? WB shift could help a lot with the tendency toward green, for example.
Makten wrote:
By the way, the only reason (and that's a big one) for me to get into Cobalt would to get rid of Fujis profiles. I've really begun to dislike them because of the oversaturated highlights and all the usable simulations being weighted towards green.
Have you tried Asita? That's my favorite for non-cyan blue skies and for skin tones.
The artistic grades are relatively easy--it is the subtle ones that you can use on almost anything that are really hard to do well.
I have made LUTs for video work, more for convenience and standardization as a base grade (now can be supplemented by AI), but these days I don't bother since there are so many highly tested LUTs available that are provided by companies and people that do nothing but create and test them all the time.
The sheer level of hours and repetition presented vastly exceed my ability to match the same quality, so I tend to have a shortlist of looks I keep on hand from 3rd parties, and adjust to suit my needs.
I have a color calibrated monitor with an attached calibration sensor I use right now to do my post processing, and that is really the bare minimum for color work. A full end to end calibration routine is very tedious, and something I have done back when printing. This isn't trivial work.
cogitech wrote:
Can you not tweak them to your satisfaction with the other settings? WB shift could help a lot with the tendency toward green, for example.
No, white balance affects the hue of all colors. I mean that the Fuji profiles have a ton of green "content" that is hard to to anything about without serious PP.
but I do see the whole profile and preset industry critical as it is more or less a business mainly build on marketing as almost everything they offer can be done at no cost with a little understanding of digital color and time. so instead of spending money on expansive profiles I would recommend learning to build your own profiles with the software you use or invest in camera profiling tools when the standard profiles are not good enough as a starting point.
I have zero interest in building profiles. I want to shoot photos.
I would recommend you not to apply your personal preferences as some sort of truth that is the same for everybody else.
brafman wrote:
Have you tried Asita? That's my favorite for non-cyan blue skies and for skin tones.
Yes, but it still has the oversaturated highlights and also too much yellow and green, and too little red. I simply don't like the "pastel" Fuji look (anymore).
Since getting a simple MFT camera (Olympus E-P7) I realized how much more I like the OOC colors from it, and I can't replicate it with the GFX files at all. Of course it's possible, but I don't know how to and I hate post processing. I simply don't want to put effort into stuff that has very little with photography to do.
One issue that comes up a lot is when people presume the all differences can be objectively sorted into "better" and "worse" categories.
Sometimes a difference is just a difference. You might like A and I might like B.
A second issue is that not all measurable differences are meaningful differences. Data may tell us that X has more or less of some thing that Y, and there's no denying that this is true. But it often turns out that no once in the real world can actually see the difference between X and Y.
A third issue, particularly when it comes to color, is that "accurate" color is quite often NOT what we are interested in. We are interested in color that looks the way we want it to look, and that involves aesthetic choices, decisions about compensating for color shifts (do you really want your sunset photo to be white balanced for midday color), and more. And these things have far, far greater effect on the color characteristics than the tiny differences that we often argue about in threads like this.
In my view, while there is some value in profiting to get more accurate color, the greater advantage is making the work flow consistent from end to end. There used to be this notion that, for example, monitor profiling's goal was to provide an "accurate" view of what an image might look like, for example, when printed on some kind of paper by some kind of printer. But in my experience, that isn't even possible — an image produced on a glowing-from-within monitor will never look the same as a image produced by pigment on some white-ish paper, top illuminated by some kind of light. The value isn't in the impossible task of dealing with all of that — it is in creating a consistent relationship between what the image looks like on the monitor and what it might look like in a print. But you still have to make the print and evaluate it before you know.
The whole color science thing is important... to the researchers and engineers who build the hardware and software systems to manage it. But that's now how we typically work when we actually make photographs.
Raamiel wrote:
I'm the creator of the Fuji emulation Cobalt profiles.
Personally, I always appreciate skepticism, because it leads us to question our work and stimulates us to improve.
In the latest release of the Fuji emulation we have put our utmost effort to be as faithful as possible to the originals.
Fuji's internal development pipelines are unknown and are certainly different from the DNG standard, however our target is the final colorimetry of the output.
To capture as much information as possible we used our new 999 patch target.
@Raamiel, Can emulation packs for Capture One be a solution for dng files that have been processed by DxO PureRaw?
edit: Or is it that the noise removal can affect the expected camera data and make the ICC Profile less reliable?
The built in CO 'Curves' for Fuji do not function as normal on the dng files like they do with native .raf
p.2 #10 · New Cobalt profiles for Fuji (version 4)
RoamingScott wrote:
We are almost 3 pages deep and so far all we know is that these make Fuji files look like...Fuji files?
The original post pointed out two additional features of these profiles:
- All film simulations are included in the package, including Nostalgic Neg, Bleach Bypass, and Classic Neg, which are only available on newer Fuji cameras.
- They can be used with any camera from any brand.
p.2 #11 · New Cobalt profiles for Fuji (version 4)
RoamingScott wrote:
We are almost 3 pages deep and so far all we know is that these make Fuji files look like...Fuji files?
Indeed, it's the finest I've come across. I am able to employ the Cobalt Fuji profiles on all of my cameras. As an illustration, I have a sample from a shooting with the family yesterday using the M10-R (Cobalt Fuji Classic Chrome).
LEICA M10-RSummilux-M 1:1.4/35 ASPH. lens35mmf/1.41/90s160 ISO0.0 EV
p.2 #12 · New Cobalt profiles for Fuji (version 4)
Fred Miranda wrote:
Indeed, it's the finest I've come across. I am able to employ the Cobalt Fuji profiles on all of my cameras. As an illustration, I have a sample from a shooting with the family yesterday using the M10-R (Cobalt Fuji Classic Chrome).
I've really begun to dislike them because of the oversaturated highlights and all the usable simulations being weighted towards green. 
