Copy link to clipboard
Copied
For instance, I have a 24-70 Sigma Art lens.
At 24mm it shows barrel distortion, which gradually changes to pincushion distortion of varying degrees over the range to 70mm.
To correct the distortion, Lightroom would have to apply a different set of values depending on which focal length was used; but applying the lens profile is just a single yes/no option.
Obviously that can work perfectly for a prime, but I'm wondering - is the LR profile for a zoom lens a set of values that automatically correct for the different kinds and levels of distortion across the zoom range, or is it just a 'one size fits all' best guess correction?
If it's the latter, a zoom lens profile would be almost useless or even counterproductive for anything requiring straight lines, so the question has practical implications.
The lens profile for the Sigma 24-70mm F2.8 DG DN | Art 019 has different corrections for 128 combinations of focal length, focus distance, and aperture (see below).
Lens profiles are stored in .lcp files, which are text files you can view in any text editor, and their structure is self-explanatory. On Mac, the profiles are stored in /Library/Application Support/Adobe/CameraRaw/LensProfiles. There's an analogous path on Windows, of course.
Here are all the combinations of focal length, focus dis
...Copy link to clipboard
Copied
I just did a quick grid pattern test with a Sony 24-70, which also goes from obvious barrel (24) to obvious pincushion (70) uncorrected.
Both ends are corrected in the right and opposite directions. So the answer is, yes, the lens profile does use different values per focal length, and then I would assume interpolation in between.
That said, if you need very accurate straight lines, I would not use a zoom lens. Primes will always be better - even a cheap 50 mm kit lens! They will also be a lot sharper in the corners, always a weak point with zooms.
Copy link to clipboard
Copied
Well - until recently I wouldn't have considered using a zoom vs primes either, but to be honest I'm finding the 24-70 Sigma so good that I've started to question the real world utility of that approach. For most work I rarely need or want apertures over 2.8, my clients rarely use the images larger than a double page print magazine ad or the very occasional wall mural, and the brutal truth is that the aesthetic criteria these shots need to hit are relatively indifferent to the kind of technical image quality differences I'm seeing between primes and this zoom. So if the distortion is corrected well enough, there's real world benefits to me in using the zoom for some scenarios, particularly since I rarely have an assistant to change lenses for me while shooting.
Copy link to clipboard
Copied
Fair enough 🙂
Copy link to clipboard
Copied
And thank you for running that test, that's good info!
Copy link to clipboard
Copied
The lens profile for the Sigma 24-70mm F2.8 DG DN | Art 019 has different corrections for 128 combinations of focal length, focus distance, and aperture (see below).
Lens profiles are stored in .lcp files, which are text files you can view in any text editor, and their structure is self-explanatory. On Mac, the profiles are stored in /Library/Application Support/Adobe/CameraRaw/LensProfiles. There's an analogous path on Windows, of course.
Here are all the combinations of focal length, focus distance, and aperture for that lens:
24 0.18 2.970854
24 0.18 4
24 0.18 4.970854
24 0.18 6
24 0.22 2.970854
24 0.22 4
24 0.22 4.970854
24 0.22 6
24 0.3 2.970854
24 0.3 4
24 0.3 4.970854
24 0.3 6
24 0.44 2.970854
24 0.44 4
24 0.44 4.970854
24 0.44 6
24 0.89 2.970854
24 0.89 4
24 0.89 4.970854
24 0.89 6
24 10000 2.970854
24 10000 4
24 10000 4.970854
24 10000 6
24 3 2.970854
24 3 2.970854
24 3 4
24 3 4
24 3 4.970854
24 3 4.970854
24 3 6
24 3 6
35 0.2 2.970854
35 0.2 4
35 0.2 4.970854
35 0.2 6
35 0.25 2.970854
35 0.25 4
35 0.25 4.970854
35 0.25 6
35 0.33 2.970854
35 0.33 4
35 0.33 4.970854
35 0.33 6
35 0.5 2.970854
35 0.5 4
35 0.5 4.970854
35 0.5 6
35 1 2.970854
35 1 4
35 1 4.970854
35 1 6
35 10000 2.970854
35 10000 4
35 10000 4.970854
35 10000 6
35.4 3 2.970854
35.4 3 2.970854
35.4 3 4
35.4 3 4
35.4 3 4.970854
35.4 3 4.970854
35.4 3 6
35.4 3 6
49.9 3 2.970854
49.9 3 2.970854
49.9 3 4
49.9 3 4
49.9 3 4.970854
49.9 3 4.970854
49.9 3 6
49.9 3 6
50 0.27 2.970854
50 0.27 4
50 0.27 4.970854
50 0.27 6
50 0.34 2.970854
50 0.34 4
50 0.34 4.970854
50 0.34 6
50 0.45 2.970854
50 0.45 4
50 0.45 4.970854
50 0.45 6
50 0.68 2.970854
50 0.68 4
50 0.68 4.970854
50 0.68 6
50 1.35 2.970854
50 1.35 4
50 1.35 4.970854
50 1.35 6
50 10000 2.970854
50 10000 4
50 10000 4.970854
50 10000 6
70 0.38 2.970854
70 0.38 4
70 0.38 4.970854
70 0.38 6
70 0.47 2.970854
70 0.47 4
70 0.47 4.970854
70 0.47 6
70 0.63 2.970854
70 0.63 4
70 0.63 4.970854
70 0.63 6
70 0.95 2.970854
70 0.95 4
70 0.95 4.970854
70 0.95 6
70 1.9 2.970854
70 1.9 4
70 1.9 4.970854
70 1.9 6
70 10000 2.970854
70 10000 4
70 10000 4.970854
70 10000 6
70 3 2.970854
70 3 2.970854
70 3 4
70 3 4
70 3 4.970854
70 3 4.970854
70 3 6
70 3 6
Copy link to clipboard
Copied
Thanks John - couldn't really ask for more than your answer and D. Fosse's in camera test.
Great to know that the profile is that thorough behind the scenes.