Processing comparison (i.e. do you really have a preference?)

On any photography forum it is not hard to find debates and strong opinions on which software does the best RAW processing and whether they can beat the camera JPEGs in terms of quality, color, tone, etc. This topic is not to discuss who’s right or wrong, but to conduct a small Sunday afternoon experiment :slightly_smiling_face:
I was wondering if people really have a preference to a particular (default) processing method, or if that is just very dependent on the type of shot, or some form of confirmational bias. So I took my Nikon D750 camera and shot 5 different scenes. Nothing too fancy, just things inside and around the house. Then I processed them in different ways:

  1. Out of camera JPEG. Neutral profile set in camera.
  2. Raw processed in Lightroom, default settings, no changes
  3. Raw processed in DxO Photolab 3, default settings, no changes
  4. RawTherapee with the default ‘Auto-matched’ profile applied.
  5. darktable with the default scene-referred workflow active. Exposure turned slightly up to bring it into the same ballpark as the others.

In all raw cases, the images were exported as 16 bit TIFF, then symmetrically cropped to 6016×4016 pixels if necessary. All images (including the OOC JPEG) were resized to 3000×2003 pixels using IrfanView (resampled using ‘Lanczos (slowest)’ and no other options active such as sharpening), and finally saved as JPEG without metadata at a quality of 90.

My question for you: per image, rank the five versions from ‘best’ to ‘worst’ in your opinion.
You can explain your ranking if you want, but it’s not needed. Please do not discuss which images might be processed in which way. That is really not the goal here.
Once I’ve collected multiple reponses, I will try to see if there is any preference to a particular processing method. My hypothesis: it won’t be the OOC JPEG :wink:
Important edit: I randomized the images below! The order the different version A-E appear in, does not correspond to the different processing steps above.

Here are the images:

1A


1B

1C

1D

1E

2A


2B

2C

2D

2E

3A


3B

3C

3D

3E

4A


4B

4C

4D

4E

5A


5B

5C

5D

5E

3 Likes

PersonalPersonally, I’m more interested in that hour glass, its so cool.

They all more or less look the same, except where RT pulls ahead slightly by matching the tone curve automatically.

But defaults: meh.

Its not the size of the sword, but how well you wield it.

1 Like

1E, 2E, 3B-3E (tie), 4D, 5B
…whatever this means :stuck_out_tongue:

I even tried to forget which was which. For the first one: 1E, 1B, 1D, 1C, 1A.

2D, 2B, 2E, 2A, 2C

3E, 3B, 3C, 3A, 3D

4E, 4D, 4A, 4B, 4C

5E, 5B, 5D, 5C, 5A

Let me clarify: the five different versions are randomized. They are not in the order as I described the different development processes!

Interesting, lets hope I did not pick all the OOC’s jpegs as being best :sweat:

1 : E B C D A
2 : A D E B C
3 : B E C D A
4 : D B A E C
5 : B E A D C

I do attack each image without any defaults applied. Exception being lens correction, chromatic aberration and even those I revisit during the edit. I do have a rough order in which I do things, up to a point. After that, mostly the technical part, I vary and go back and forth at times.

I am curious about the defaults though, although I’m not going to apply them myself.

1C, 2D, 3B, 4B, 5E

Although all of them look fine for me in isolation and would make a good starting (or ending) point

@Jade_NL

lets hope I did not pick all the OOC’s jpegs as being best

Actually: what if you did? So what?
A “complicated” development method does not necessarily have to be “better”!?!

Have fun!
Claes in Lund, Zweden

:smiley: The server doesn’t seem to like that I send just the list:

1D 1C 1B 1E 1A
2D 2A 2E 2B 2C
3B 3C 3E 3A 3D
4E 4A 4C 4D 4B
5D 5A 5B 5E 5C

That was fun. In most cases, you had to pick your poison.

  1. E, B, D, A, C
  2. C, B, E, D, A
  3. E, B, C, A, D
  4. A, E, C, B, D
  5. D, B, E, A, C

+1 on the hourglass…time’s up!

Ordered from best to worst:

1: C B D E A
2: E B C A D
3: B E C A D
4: A E D B C
5: D A B E C

Very interesting.

  1. BECDA
  2. DAEBC
  3. BECAD
  4. BDACE
  5. BEDAC

1: DABEF
2: ABEFD
3: FDABE
:grin:

When it comes down to it it doesn’t matter at all. But… One does prefer some over others.

In this example I don’t even count the jpegs, those are basically end-products that you can manipulate/edit in-camera. The defaults that are produced by the other four RAW editors are a decent starting point at best, nothing more or less.

With that in mind I do hope that I did pick RT, DT, LR or DxO (have no experience with either, and lastly jpeg.

“My” editor should “win”! :rofl:

EDIT: I a word.

1: B E C A D
2: D A E B C

I certainly like first two choices (1: B E and 2: D A), and three other variants I certainly dislike.

Third scene is a good example of a scene with too wide dynamic range for OOC jpegs. So if someone likes colors/tone curve of OOC jpegs, but wants to bring shadows up - he/she might want to have a raw file with starting point looking the same as OOC jpeg.

My editor isn’t even in the comparison…

@CarVac : Yeah, some good ones are indeed “missing”! I’m not going to rain on Thanatomanic’s thread though…

@CarVac @Jade_NL I seriously considered it, but finally decided to keep things manageable. At some point, we can maybe extend the selection and see how well others fare :slight_smile:

1 Like

F?! You’ve found the secret edit? :smiley:

I’m curios to know which are the OOC jpeg
1:DCEAB
2:CAEBD
3:ECBAD
4:AEDCB
5:DABEC