On 2012-12-31 00:03, Jose Carlos Garcia Sogo wrote:
> One post on behalf of Pascal de Bruijn
Very much appreciated, thank you!
> Hi,
>
> Time to get some confusion out of the way.
>
> A typical complete camera input profile consist of two things: curves
> and an XYZ matrix.
>
> The default color processing of Darktable (and most other open source
> raw converters) use the Adobe DNG D65 XYZ color matrix (camera
> specific), but that means the curve is missing. Which we complement by
> applying a mostly brand specific basecurve.
Does this mean that if I were to create my own input profile based on a 
test chart and with the correct settings, I would be able to get (more 
or less) natural looking images without a base curve?
> Other open source RAW converters complement the color matrix with a generic 
> gamma curve
> (often 0.45 which is the inverse notation of 2.2). But UFRaw for
> example adds a linearity option (0.10 by default) which actually
> distorts that normal 2.2 gamma curve to increase contrast (read to
> make it look good), and the end result is probably not dissimilar to
> how our basecurves look, except that it's not tuned to a particular
> brand. The problem with this approach is that it leaves you with a
> discrepancy how the camera processes the RAW (most people seem to
> expect images in a RAW converter to look similar to what the camera
> generates). And there is no way to solve that without applying brand
> specific curves.
Fair enough!
> So if you disable the basecurve in darktable it's completely expected
> that a RAW files will look dull, since the data is still linear (at
> least to the extent that the raw sensor data is linear to begin with)
> then.
I did not expect this. Again, if I do my own input profile (and do it 
correctly), would I not be able to do without the base profile?
> The idea that any particular approach would be "better" than any other
> is nonsense, and please do not spread the idea that it would be. It's
> just a preference nothing more.
I apoligize if it seemed that I wanted to advocated a certain approach. 
I honestly just want to understand what is going on. Of course I have my 
own personal preferences for how I want my images to be processed, but I 
certainly do not want to claim them to be the only way of doing things. 
Of course I will think my own approach is superior, but then again, I 
would not have picked it if I did not think so :)
> The difference between the curves:
> Basecurve in camera RGB (before colorin)
> Tonecurve in Lab colorspace (after colorin)
Right now I do not understand what you are saying, but it is late and I 
may have been drinking too much wine. If I have further questions, I 
will post them tomorrow...
> This is made visible by turning off plugin grouping in darkroom mode,
> which I've demonstrated time and time again in my screencasts,
My apologies, I have not seen all screencasts. I will try to catch up 
tomorrow.
>   so you
> can see the static order in which plugins are applied (bottom to top).
> These plugins account for everything being done, not just
> "corrections".
Thank you very much for your intervention, I was hoping this thread 
would catch your attention somehow. It is much appreciated.

I feel like I still do not understand everything, but I will leave the 
rest for tomorrow.

Again, thank you!

Eckhart Pedersen

------------------------------------------------------------------------------
Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS,
MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current
with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft
MVPs and experts. ON SALE this month only -- learn more at:
http://p.sf.net/sfu/learnmore_123012
_______________________________________________
Darktable-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-users

Reply via email to