Basically im thinking, if before the conversion i apply the most neutral profile for my camera ( they call it Repro ) here’s the description.
The Cobalt Repro profile is engineered for peak precision in imaging, offering a linear response that captures the full dynamic range of your camera without alteration.
While other profiles from the Base Pack for Adobe stick to a basic contrast curve, this one breaks the mould, creating an unbridgeable chasm for advanced editing. In essence, this profile delivers the raw image to your screen in its purest form; to beginners, it might just look like a dip in contrast or exposure. But once you realize you’re at the starting line without even touching a slider, the leap in editing latitude will be so extreme, that you’ll feel like you’re working with files from a camera that’s not just newer, but spectacularly better at recovery and correction.
High-Fidelity Capture: Ideal for high-contrast environments, it retains detail in both shadows and highlights, crucial for accurate colourimetric reproduction.
Perfect for Digitization: Essential when scanning negatives or slides, ensuring colours and details are preserved accurately.
Professional Use: Suited for scientific, archival, or artistic work where exactness in colour and detail is paramount.
Post-Production Flexibility: Its linear data provides extensive editing latitude, allowing for detailed adjustments without quality loss.
and choose none in the profile conversion, would that lead to more accurate results ? Has anyone tried it ?
Yes indeed it is, i was just hoping to bypass with a profile that made with as little done to the image as possible.
I ran the pixel shift dng in capture one and chose the no color correction profile and liniar profile, exporrted it and used it in nlp and the results are a bit better but not by that much.
Careful in comparing conversions of a single file that has passed different apps: Capture One, Lightroom and all others have their own idea about how a raw file should look. Therefore, an export of e.g. C1 can look different to the one from Lr.
Additionally, NLP analyses image content and adjusts the conversion accordingly. Whether one conversion is “better” than the other can be a) pure coincidence due to differences as noted above and b) a matter of personal taste or preference.
My best practice is to test a few conversions and then convert the whole lot, taking the results as a starting point for whatever I want the results to look like - and I only test negatives that look different than the average negative (e.g. heavily underexposed) or if the conversions turn out too far away from what I want them to be (e.g. a strong colour cast).
It’s a good idea to consider advice from Nate’s guide - and sometimes it helps to ignore it.
Your best bet in combination with NLP is to avoid any other app before launching NLP.
catalog/import your RAW files with Lightroom. Develop settings don’t matter because NLP takes unaltered raw data from the file. Built-in JPEGs and Lr’s thumbnails and previews take no part in this, they are just a front.
convert images with NLP
If you use C1 and export to e.g. TIFF, C1’s idea of how images should look are baked in. Same with export to linear DNG in which the colour filter pattern (e.g. r-g-g-b for a Bayer type sensor) is replaced by (RGB-RGB-RGB-RGB), a step that requires some math that is based on parameters provided in C1. These parameters determine, how data from sensor photosites are to be combined to produce RGB pixels.
