Cross posted from a question in the 18-135 thread.
Originally posted by mee Yes. Actually, I own this lens.
Even stopped down to f/5.6 mine doesn't seem to behave like the sharper images in this gallery.. it is closer to the upper middle in sharpness if there was a level between the really sharp images listed here and the really soft. Which, after seeing some really soft series of images in this thread followed by really sharp, led me to my question.. how many of these are truly the lens and how much is actually PP cleanup for the limitations of the lens? Any PP work gives an inaccurate view of any lens if it is rather involved in a thread I thought about the natural aspects of the lens in question.
Maybe you should send your lens in for alignment, but first i'd make sure you can't get good images from it.
There are a few assumptions you're making here which are inaccurate. PP doesn't make an image sharper, it makes it look sharper. Here's a recently posted picture.
And a 1:1 crop from from that picture unprocessed.
You could easily argue, this lens isn't very sharp.
The dog may look sharper in the finished image, but that's not the PP bringing out something that isn't there. This was an overcast dull day with not a lot of contrast. SO two things are done to bring up the level of contrast, I add definition, which is in essence micro-contrast, and I use the sliders in levels to adjust the brightest and darkest levels shown so that I use the full palette of my output device. Neither of those things change the level of sharpness. What they do is they increase the amount of distinction emphasizing differences that are actually there but may not be discernible by the human eye. The fact that you can't see them, the way they came off the camera doesn't mean they weren't there. That is the biggest mistake people who don't do PP make. They assume that the camera recorded it the way it was.
In actual fact, one of the reasons I process as quickly as possible, is because I want to remember how it was, and in PP create what I saw at the scene. People tend to think, it's not like the original image because you processed it. That's exactly backwards, most of us PP so that we bring out in the image what we saw in the scene.
Another factor is this image was hand held. Right off the bat, I have no expectation that this image would be razor sharp. That it's as sharp as it is is a testament to Shake Reduction. If I'd thought this could be a gallery print, I would have set up the tripod and done it right. But, I was out for a walk with 3 other people, so that wasn't likely to happen.
Now another couple of images. These shot with an A-400 on a tripod with 2 sec delay. A much brighter day with more natural contrast
The whole image
1:1 pixel peeper- all sharpening turned off.
This image looks much sharper and detailed, but, I had more light, so the original image had more natural contrast. I was shooting on a tripod, and the subject as well had more natural contrast. All those things combined make this a sharper looking image right off the camera. The reason I don't have similar images from the 18-135 is, I rarely shoot it on a tripod, and it's a walk around lens, so the light is often less than perfect.
Here's the DA 18-135 used on a tripod, on a bight sunny day with some contrast in the subject area. Shot at 18mm, so , not even in the lens' comfort zone. And this is a 1:1 crop. All sharpening is turned off. To me, this 1:1 crop looks as good as many hand held images of the un-cropped images, that have been reduced in size and therefore had their sharpness artificially increased. The pine needles are distinct, the curved lines of the broken pottery are lines are clear and smooth.
The scene
1:1 pixel peeper- all sharpening turned off.
My computer monitor is 92 DPI, and this image was taken with a k-5. So This image would look the same quality as it does in the Pixel peeper, printed at 53 inches wide. With a bit of sharpening applied, it would look even crisper than it does as it is now with sharpening turned off.
So brief synopsis
PP gives you an opportunity to opportunity to make the picture look like what you saw at the scene, and bring out detail that may have been present but invisible in the un-processed print. (And that you may have observed when experiencing the scene live.)
IF you are going to look at lens sharpness, you must work from a sturdy tripod with a 2 sec delay. No, ifs, no ands, no butts.
Images taken of high contrast subjects in high contrast lighting situations will look sharper than images taken on dull days without much contrast even taken with the same lens and camera.
Or to be more blunt, if you haven't tested your lens in good light, on a tripod with a 2 second delay, with a subject with decent contrast, you really have no idea how good it is. You might have some idea about how steady your hand is, but that would be the most you could accomplish.
Last edited by normhead; 01-27-2014 at 09:23 AM.