Sigma 12-24: what soft corners?!?

Started Dec 19, 2003 | Discussions thread
Flat view
Petteri Sulonen Forum Pro • Posts: 24,585
Sigma 12-24: what soft corners?!?

Here are more of my extremely preliminary comments on the Sigma 12-24.

I got my test shots on the computer and had a chance to play with them a bit. For my part, they have gotten rid of the major doubt I had about the lens: corner softness, at least with the 1.6x field of view crop. See for yourself:



Yes, these have been post-processed: the intent is to make the sharpness difference as easy to see as possible. Here's what I did:

1. Shoot a fabric partition at 12 mm, f/4.5, as close as I could get hand-held to being exactly perpendicular to it.

2. Desaturate, increase contrast, adjust gamma for similar look. This gets rid of effects of poor and uneven lighting, and emphasizes resolution only .

Note: there is NO unsharp mask applied.
Relevant camera settings:
+ JPG Large/Fine
+ Contrast -2
+ Sharpness -1

Here are the crops with something like my usual amount of USM added (same to both crops):



Yep-o, the corner is slightly softer than the center, but the difference is very subtle; much better than I expected. I don't even know at this point whether this is in fact due to aberrations rather than simply field curvature -- I'm very surprised if a lens this wide has a perfectly flat field.

Conclusion: the people complaining about soft corners on this lens at least with the 1.6x crop might want to re-examine their technique. At this field of view, with real photos you're bound to have a very deep scene: this will inevitably mean differences in sharpness, most readily apparent in the corners, which are magnified compared to the center (due to the extreme stretching out that happens when this wide a field of view is projected onto a flat plane).

I'm hoping to do my usual test suite tomorrow, to investigate CA, flare, contrast, vignetting, and color. However, I've seen plenty of test shots and I don't expect any huge let-downs there; I suspect flare will be the biggest issue.

Oh, btw -- I also shot a distortion test shot (at 12 mm again). I'm impressed: there's barely any discernible.

(BTW, don't attempt to judge vignetting on this pic; it looks that way because there's a fluorescent light just facing it -- the bright spot is its reflection.)

Preliminary conclusions:

1. What soft corners?

2. Very low barrel distortion: much better in this respect than the 15-30.

3. I'm very impressed. Maintaining these qualities (lack of distortion and sharpness across the frame) at this field of view is a quite a trick to pull.

We'll see how the lens does tomorrow in my usual tests -- but so far it looks like far less of a compromise than I expected.

And, a word to those of you disappointed with it: don't give up so easy. Ultra-wides take time to tame: AF is hit-and-miss, because the huge DOF doesn't give enough information to the AF sensors to place focus precisely, and because wide scenes are deep scenes, it won't be obvious to get everything in focus as you'd like. Shooting good wide-angle scenes is not a matter of pointing and shooting, and even if you're very good at, say, 18 mm, doesn't mean that you'll be able to get great quality at 12 without a bit of practice.


Portfolio: [ ]
Pontification: [ ]

 Petteri Sulonen's gear list:Petteri Sulonen's gear list
Fujifilm X100S
Flat view
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow