The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Lightroom DNG profiles and Capture One, a comparison

Bob

Administrator
Staff member
First I will explain that I am a Capture One fan. I have used Lightroom extensively but over the past year and a half have mostly abandoned it due to the controls and flexibility of Capture One especially when it came to processing Phase One P45+ files.

Recently, a dear friend gave me an XRite Color Checker Passport. Besides th nifty cased set of white balance card, color swatches and a standard color checker, it comes with software, both stand-alone and a Lightroom plug-in to create custom dng profiles.
last night, I set up the color checker and studio flash and with a P45+ took two images. One was flash exposed and the other was illuminated with the modeling lamps to give me a tungsten illuminant. With these two images, I used Lightroom and the passport plug-in to create a dual illuminant dng profile for the P45+.

To compare them, and to see how they looked compared to the same image procesin in Capture one, I layered them in Photoshop and added layr masks to expos a portion of each image in each Color Checker square. in addition, I added an octagon, filled with nominal Color Checker values that i obtained from Bruce Lindbloom's web site .
The result was interesting to say the least:



The lower portion of each square is the file as processed by Capture One 5.0.1. The upper left corner is the Lightroom processed file using the "Adobe Standard" profile. The upper right corner is the Lightroom processed file using the custom dng profile created by the passport plug-in. In all cases adjustments were zeroed out and default "film curves" or "medium contrast curves" were used.

the first observation is that the Lightroom files all appear significantly darker than the files processed by Capture One.
In order to investigate this a bit, I added a curve adjustment layers clipped to each of the stacked images. The gray scale was used for reference and then the curve for each layer was adjusted to get the gray scale to match the reference values represented by the central octagons. This produced a more "standardized" result




Well, now we can see some differences in a more comparable way. Even though all of the images had been white balanced to the third gray square, the Lightroom highlights have a slight pink cast. in addition, the Lightroom profiles produced more saturated results arguably more accurate with these adjustment curves applied.

So, what adjustments were necessary?

Here are the curves that i used:



Very interesting. The Lightroom curves were more linear than the Capture One curves. Both the Lightroom "custom dng" and "Adobe standard" profile compensation curves essentially matched. The curves on the right are the two overlayed with the "Adobe Standard" in black and the "custom dng profile" in red. The profile did not affect the contrast curve in a significant way.
The curves added about a stop more brightness to the files processed in Lightroom as can be seen by the uniform displacement of the adjustment curves above the 45 degree line.

the Capture One curve on the other hand was interesting in that it clearly adds contrast and brightens the image from black all the way through the upper quarter tone, then rolls off to emulate a long film-like shoulder (please don't flame me for this). In other words, by comparison, Capture One compresses the highlights while providing more mid-tone and shadow contrast.

As for color accuracy, I will let others judge for themselves. To be fair, the brighter squares where C1 seems grayer in the "normalized" comparison are those most affected by the compensating curve, but one thing is clear. Lightroom requires that its exposure slider be moved up a bit over a stop in order to produce files of equivalent brightness to the same file when processed in Capture One. The difference in midtone gamma explains to a large degree, the different appearance of P45+ files processed in Capture One compared to those same files processed in Lightroom.

Since much of the apparent detail we perceive is derived from this portion of the total tonal range, I might surmise that at least some of the enhanced detail we see in Capture One processed files might be due to this difference in mid-tone gamma.

-bob
 
Last edited:

Jack

Sr. Administrator
Staff member
Bob,

I'd be willing to bet a bottle of Hendricks your C1 gray correction curve is an approximate inversion from C1's "Film Standard" curve :)D). You might want to reprocess the file under the linear curve setting and compare.

Nice work -- I know how much work it is -- and great way to represent it!
 

bradhusick

Active member
Bob,
Awesome way to show the results! Do you think these results are specific to the sensor or can we infer similar behavior from an M9 for example?
 

Ben Rubinstein

Active member
Could we then apply the opposite of that curve in ACR to the RAW files to approximate the C1 film look which I find much missing in the adobe rendition? I assume though that you would need ACR to start with a linear rendition first. Might be worth playing with. If anyone knows how to draw the 'C1' film curve from Bob's diagram (I'm not that technical) I'll plug it into ACR and see what happens. I'm assuming that LR even in the '3 Beta' still doesn't allow you to apply a specific curve. A loss IMO.
 

Godfrey

Well-known member
All this says to me is the the Adobe Standard camera calibration profile for your P45+ isn't quite the same as the Capture One profile.

Get out DNG Profile Editor, create a camera calibration profile that is more to your liking, install it and set it as the default in Lightroom for that camera. Done.
 

Bob

Administrator
Staff member
Bob,

I'd be willing to bet a bottle of Hendricks your C1 gray correction curve is an approximate inversion from C1's "Film Standard" curve :)D). You might want to reprocess the file under the linear curve setting and compare.

Nice work -- I know how much work it is -- and great way to represent it!
It is interesting to actually see the shape of the C1 "film standard" curve.
I repeated the experiment and the C1 linear curve is pretty close to linear.

View attachment 25172

As an exercise, I an attempting to get close to the c1 response curve using the profile editor, but it is not quite the curve I am wanting.
-bob
 

Bob

Administrator
Staff member
That's my point -- so the curve you applied to the original "Film Standard" curve must be very close to inversion, give or take a constant that looks like a 1/3rd stop pop...
So my point was that part of what we like about C1 is its film standard curve. The acr medium and strong contrast curves are different animals entirely.
None of their out of the box curves have such a nice soft shoulder and their tools don't make defining one that easy (sort of, but not exactly)
-bob
 

docmoore

Subscriber and Workshop Member
So Bob,

If you output a DNG from C1 with inverse of standard curve and use that as the capture for the DNG editor? When or if ACR will normalize it do you approach the curve you desire? Corrected for exposure boost that you see with their more linear curve?

Probably not....?

Bob Moore
 

Bob

Administrator
Staff member
So Bob,

If you output a DNG from C1 with inverse of standard curve and use that as the capture for the DNG editor? When or if ACR will normalize it do you approach the curve you desire? Corrected for exposure boost that you see with their more linear curve?

Probably not....?

Bob Moore
Not yet but I haven't given up trying.
It may be a limitation of the dng profile editor, but the curve shapes it produces so far are just wrong for my purpose.
-bob
 

docmoore

Subscriber and Workshop Member
Understood...

I purchased an Eye1 Extreme with the intention of profiling cameras only to find that DNG editor seems to be the limit for Adobe products...need to explore profile generation for C1.

Bob
 

cmb_

Subscriber & Workshop Member
So my point was that part of what we like about C1 is its film standard curve.
Part of what I like about Capture One is having both the Film Standard Curve and the Linear Response Curve plus two others to choose from as well! I use the Film Standard most of the time but do use the Linear Response as well. When images appear to have some highlights clipped with the Film Standard Curve, switch to the Linear Response and a lot of information is recovered.

Bob - Nice work!
 

David K

Workshop Member
Bob, I'm curious what the difference would be if you moved the brightness slider instead of the exposure slider (talking LR here). I've been told that moving the exposure slider shifts the entire histogram and is not the correct way to brighten an otherwise properly exposed image. Most of this stuff is beyond my comprehension so forgive me if the question leaves you scratching your head :)

"Lightroom requires that its exposure slider be moved up a bit over a stop in order to produce files of equivalent brightness to the same file when processed in Capture One. The difference in midtone gamma explains to a large degree, the different appearance of P45+ files processed in Capture One compared to those same files"
 

carstenw

Active member
That's right, the Brightness slider anchors the end points (more or less; not completely), and moves the middle of the range proportionally to the distance to the nearest endpoint. The normal workflow is to tweak the Exposure slider until the histogram ends somewhere near the right side, but not touching, and then tweaking the Brightness slider until the level of light is good.
 
Top