Hallöchen!
Torsten Bronger writes:
> [...]
>
> I may revoke this plea if too many pictures are made available,
> since at the current stage, I am only interested in finding out
> whether my method is applicable to other cameras, too.
Please don't send in further pictures for this. Thank you to
everyone who has participated. So, what are the new observations?
1. As I have already pointed out but want to emphasise again,
Darktable's basecurves are not responsible for totally off
skintones or otherwise bad photos, as was suggested on
darktable-users recently.
However, people who also shoot in JPEG (like me) may want a
really well-matching basecurve, so that RAWs and JPEGs look
equivalent. This is the main motivation for using measured
basecurves I think. Besides, the basecurve of my Sony NEX-7
simply was not reproduced by any basecurve in Darktable yet.
(Another one may be a matter of taste: For the Canon 450D, for
example, I like Canon's basecurve more that DT's one.)
2. Especially the shadow behaviour of the basecurve has a noticable
impact on brightness *and* colour. Therefore, the accuracy of
the approximating fit must be really high there. I set my
control points in logarithmic distance on the x axis to realise
that.
As a related note, I think it is sensible to offer optional
log-log scale in the basecurve module.
3. The basecurve of the manufacturer can be measured reliably,
however, it is not a fully-automated process. A human being has
to check whether the input picture is suitable. Moreover, the
three curves for red, green, blue, and average may differ in the
almost-black area, so you have to choose wisely which one you use
for fitting.
The JPEG is always smaller than the RAW. To get a perfect
overlay, you usually have to centre the JPEG over the RAW,
however, in case of the Canon 6D, there was a one-pixel error
with this method. This didn't spoil the plot, but one should
check the relative position if the curves look fishy.
4. I have too little data to say how many different basecurves
e.g. Canon uses. FWIW, the 5D and 350D were "canon eos like
alternate", and the 60D and 450D were "canon eos like".
Beseides, their basecurves are close enough to DT's presets for
most people's purposes I assume.
5. I like to use dcraw even for calibrations meant for Darktable.
But for basecurve measurements, I had better results with
Darktable with an empty stack for the RAW-->linear TIFF
conversion. I suspect WB was slightly off in dcraw in one case
for some reason.
Tschö,
Torsten.
--
Torsten Bronger Jabber ID: torsten.bron...@jabber.rwth-aachen.de
or http://bronger-jmp.appspot.com
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
darktable-devel mailing list
darktable-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/darktable-devel
