whwiden

I have been experimenting with Darktable as a replacement for Lightroom and Capture NX-D. For others who may wish to do so, I have a few initial observations about how to do comparisons.

As background, I am doing my testing using a Nikon D7200 and the 35mm 1.8 dx lens, shooting a NEF file with a large basic JPG. I have left the camera settings at SD, ISO 400, auto white balance.

I first used a NEF file to export a JPG using Capture NX-D and compared it to the basic JPG generated by the camera. No surprise, but the two seemed identical (and I made no adjustments using Capture NX-D).

I then loaded the same NEF file into Darktable and similarly generated a JPG without making any adjustments. The output was very similar--but with two differences. The base curve automatically selected by Darktable was something called "nikon like alternate". This was a little bit darker in color/tone than the other JPGs. However, if you right click on the base curve line in Darktable, you will get many other base curve choices, including one for the D7200. When you select this base curve, the colors of the JPGs all appear identical. The second difference I noted related to defringing. The camera JPG and the Capture NX-D generated JPGs did not display purple fringing. However, the default JPG generated by Darktable did display some purple fringing. However, this was corrected easily by activating the "defringe" module in Darktable. Here, you have many adjustments in Darktable--the two generic options being "global average fast" and "local average slow". Local average slow did a more complete job to my eye, and fine adjustments were possible using sliders.

With the selection of the specific base curve and the defringing option turned on, I was able to get a starting point to experiment with further changes which appeared, at least as to color, virtually identical to the Nikon generated options. If you want to experiment with Darktable to see what other settings might do, this is a good starting point.

A few other matters might be of help for testing. First, you must be in the lighttable tab to export files to JPG. You do this from the lower right portion of the screen. Also, on the right side of the screen you can go into the "history stack" and click discard to basically start over with your adjustments. This is very useful for testing. When you start over, I find that you have 3 initial settings active. The picture orientation, the base curve and the white balance. You go into the "darkroom" tab to activate modules and to change them. Here is where you would change the base curve setting and activate the defringe module. I have attached a basic NX-D export and Darktable export, so you can see the differences. The last JPG is the Darktable version with the d7200 base curve and defringing activated.

It is somewhat of a coincidence that all of a sudden there is great interest in dumping lightroom based on Adobe announcements. I have not wanted to sign up for Lightroom 6 (I still use 5.7) and I do not care for the idea of a subscription. But what got me interested was my desire to travel with a small and inexpensive computer which I might use to make some photo edits. For $350 I was able to get a low end version of an 11 inch Lenovo Yoga 710. I installed Linux and it runs Darktable reasonably well--the same computer would not work well using Lightroom, I do not think. I prefer to use Linux to go online when traveling as I perceive it to be more secure, etc. So, I have been looking at this option for awhile.

My sense is that Darktable can do very well. Certainly as a replacement on the road for Capture NX-D and maybe even for Lightroom. I have been able to make the same kind of adjustments to ordinary photos. I am not yet able to get my film negative scans using a macro lens to look proper in Darktable--I am very good at this in Lightroom--but I do not do this on the road in any event.

I hope this helps those considering Darktable. (For what it is worth, I am using the most recent unstable testing version of Darktable but the stable version has most of the features you want, other than perspective adjustments).

Capture NX-D export


Basic Darktable export


Darktable-d7200-bc-defringe

BHC Loc: Strawberry Valley, JF, USA

Sorry, these tired old eyes can't see the difference. Maybe I could have, had you checked the "Store Original" box and allowed me to enlarge them.

whwiden

Here are three close up crops that show the point I was discussing better. The first is the camera generated jpg. To my eye, it compares well to the third jpg which was generated by darktable with the defringe and the d7200 profile applied. The middle jpg was generated with darktable without adjustment. It shows the purple fringing (which Nikon apparently does automatically) AND it also shows a slightly darker color associated with the automatic nikon like base curve that darktable applies. This darker coloration is adjusted by the darktable d7200 base curve.

What I was hoping to show for interested parties is that you can start from the same baseline as Nikon generates from camera or Capture NX-D and then experiment from that point with adjustments to see how they compare to NX-D and to Lightroom.

I clearly messed up by not showing some details in the initial post.

camera generated jpg


darktable generated jpg no adjustment


darktable generated w/defringe and d7200 profile

dpullum Loc: Tampa Florida

Few if any in UHH lay out the ground work parameters and rational for their actions. You were detailed and approached the evaluation of Dark Table in a scientific way... my hat is off to you whwiden. Too often UHH people use a wet finger in the air opine evaluation wetting their finger in the same rear facing orifice where they keep their heads.

I rarely use my RAW... and have for the most part used my pocket 24/7 20 mp Sony HX50 which does not have raw. I am trying to break away from Paintshop Pro X9 and PSE 13 and learn Topaz Studio. I own all the suite by Topaz. By offering the basic Studio free and breaking down plugins into smaller lower cost segments, they hope to capture a fast edit market.... fewer will be using base edit programs... especially the clunky ol Adobe Cult subscription only programs..... Free ever improving base of Studio is a better price.. Yep.

Times are a Changin' as Bob Dylan... Nobel Prize Winning frog voiced boy now old sang in the 1960's'

Adobe should have taken a warning from Kodak... "But digital, NO, we make film." Yep, usta.

sailorsmom Loc: Souderton, PA

Nice shots, whwiden!

Hal81 Loc: Bucks County, Pa.

You can see the difference if you look at the shadow of his hat in front of his ear.

BHC Loc: Strawberry Valley, JF, USA

OK, I can see that there are differences in the pixel detail. BUT, I would not think any the less of any of these shots if I looked at them from a normal viewing distance in any size. I believe we are so addicted to minor artifacts in images that we miss the whole purpose of the image. Take a very close at any painting by a famous artist and I’ll bet you there are a few brushstrokes that look unusual, distorted or out of place. Some people can’t see the forest for the trees!

whwiden

Thank you for your kind remarks. I hope to post some more information soon.

whwiden

For most shots, details like this matter very little. I agree with your general point. Indeed, I often shoot basic jpg and just try to get the shot correct in camera. However, in some cases (a landscape destined for a large print) details like this may matter to some. Wedding photos might be another example. I find that you need to look into details to evaluate the capabilities of the product. My hope is that some may find a free alternative to a costly program and subscription service that they are comfortable does not involve a compromise of one sort or another. In my personal case, I also would prefer to not use Windows operating systems or an expensive Apple computer for travel (though I am not adverse to Apple, unlike Windows).