So - I am using a Nikon Z 50 + 16-50mm lens. The camera applies lens distortion correction in camera. In fact, for this lens at least you can’t turn it off.
My question - when I “import” a jpeg into PhotoLab with Lens Corrections enabled what happens? Does it - recognise that lens distortion corrections have already been applied and do nothing? Does it apply them again blindly, resulting in double-correction? Or, is it smart enough to recognise that the camera has already made a correction but it improves on this? (Somehow, I doubt the third).
Looking at some images - and clicking the Distortion setting on and off in PhotoLab I notice that in general it does make a small change. (I can see the image move). It seems, though it is difficult to be sure that when the image was shot wide a small correction is made, but not when shot at 50mm - which would make sense if the lens is known to have issues wide. But the fact that there is a change shows that it is applying something. So this would seem possibly to imply the first variant above; it is applying a correction blindly.
Does anyone know what happens in this case?
Thanks
Update. - further tests
I did another test with a 24 mm F1.7 lens. (The issue with the kit 16-50mm lens is that the camera automatically sets distortion control and that could be a special case which confuses things). With the 24 F1.7 lens I took two images, one with distortion control OFF in the camera and one with it ON. Photolab has the relevant Optics Module and I have set it to apply optical corrections on “import”. In both cases the image appears to change in the same way and to the same degree when I toggle the Distortion button in PhotoLab. This leads me to believe that PhotoLab is indeed applying lens distortion correction blindly regardless of whether or not in camera distortion control is set.
Does anyone have any more information on this point?
I may be wrong and will have to test for myself, but I always believed that correction was done when converting RAWS too JPG / TIFF etc. (The last years I only use RAW anyway…) So, without checking I would say that the JPGs should not be affected.
In this case I am “importing” jpegs - not RAW. The setting is (I am sure) applied in camera since the camera will apply distortion control to jpegs. My setting in Preferences in PhotoLab is - for jpegs apply only Optics Modules corrections. This causes the Distortion setting to be applied. Of course you can toggle it off. And, as I say above, my tests are clear; the lens distortion correction is applied to the images by PhotoLab seemingly regardless of whether any correction has been applied in camera…
I think @Wolfgang had in mind PhotoLab Auto-apply Presets setting ‘6 - No correction’.
I have distortion correction enabled in camera, just to check if recomposition is needed – typical use example are city shots at 24mm. My PL presets, except for one used for architecture, do not set ‘Distortion correction’, which I prefer to control manually, especially if people are involved. I have ‘6 - No correction’ set as default for both raw and rgb files. I use camera jpegs for quick selection only. PL is designed for RAW processing, for jpegs editing there are probably better choices.
You have a Nikon. Take a shot in raw and jpg. VIew them in pl with ‘no correction’ and find the differences. Eventual you can use the preset ‘optical correction only’.
You can also view the nef in an imagebrowser and save it. The embedded jpg will be saved without any metadata, when I remember me.
Use a picture with many vericals and horizontals like buildings.
That is correct. Unfortunately, DxO allows only one preset to be defined for either RAW or RGB image files. DxO also provides just one optics module for a given camera/lens combination for out-of-camera JPEG images. The optics module specifies how distortion correction is to be applied if selected, and is based on a particular set of in-camera behaviors - presumably the default settings. You’ll have to decide for yourself if it’s appropriate to use DxO distortion correction on a given out-of-camera JPEG and how much of that to apply to the image. I prefer to apply no preset corrections to out-of-camera JPEGS, but rather, use the JPEGS occasionally as a reference while adjusting RAW files for geometry, color, contrast, and so on.
So - it does seem this is probably the case. PhotoLab applies the distortion setting if you apply it and it has the relevant Optics Module. It doesn’t check to see if it has already been corrected in camera. (I am talking purely about jpegs). It is no problem. As others have suggested I can just turn off the setting in PhotoLab and leave it on in the camera. (Or vice versa).
The next step is, of course, to compare the in camera distortion correction with what PhotoLab does - which I think will involve graph paper.
Yes, and also pay attention to the other optical corrections that PL applies (vignetting, lens softness correction, chromatic aberration) … which means you may be better off leaving your camera neutral. So, take a couple of tests to find out what suits you best.
@JoPov - i think these are considered as part of the “Optical Corrections” preset. If you “import” without any corrections, “No corrections” they are not applied. But they are if you “Import” with “Optical corrections only” (including to jpegs).
Hi @Wolfgang - yes. I am aiming for a complete finished image for jpegs in camera, including saturation and sharpness. (I copied Ken Rockwell’s settings). Only I have Active D lighting on. I think it does something a bit similar to PhotoLab’s Smart Lighting.
They act similiar when set to Automatic. The system does something you can hardly control, in other words: not really recommmended !
Use DxO’s Smart Lighting in spot weighted mode → search here in the forum as there are some tips. The overall contrast can be controlled by drawing rectangles over the highlights and shadows that should still show texture.
But Nikon’s Active D-Lighting, as far as I know, first reduces exposure by 0.7 to 1 EV and then increases some shadows, and these “corrections” are not visible to you. I believe you even have to use matrix metering.
see → Balancing Photo Exposures with Nikon's Active D-Lighting
As far as I know, Active D-Lighting simply underexposes in RAW. It then instructs the jpeg generator, whether in-camera or in NX Studio via makernote, to apply suitable brightness correction and tonal curve. SmartLighting is far more complex (even in the Uniform mode), since exposure may be corrected by different amount, depending on the local “context” (it’s the DxO’s secret what that means). It seems that PL, like much other non-Nikon software, does not take into account the active D-lighting settings provided in makernotes associated with the RAW image.
EDIT: Actually Nikon jpeg generator seems to take some “context” into account when processing RAW with Active D-Lighting but generally I prefer the results from SmartLighting, especially in extreme cases when colors may get washed out. When I want to be “safe” with highlights, I choose different metering method or just set ExpComp=-0.7 or so.
