Micro landscapes


(ump) #1

The idea of these shots wasn’t to have a sort of landscape in miniature with the lines of branches substituting for hill lines and mountains, and buds and details of bark substituting for trees, so that the images hopefully possess the different levels of depth that landscapes usually do.

https://www.flickr.com/photos/138717866@N05/23281502444/in/dateposted-public/

https://www.flickr.com/photos/138717866@N05/23814898441/in/dateposted-public/


#2

Did you mean, “was to have a sort of landscape in miniature”?
1-3: seem to have some halos on the branches/twigs. 2&3: it almost looks like CA, without the red shift.
1: reducing chroma noise may help the image some
4: Personal taste: I would have focused on the leaf, in the center, and left the background in bokeh. I would have also shifted the focused leaf off center.
2: my favorite: the blue halo is bothersome though.

Macro landscape are a favorite of mine; especially those that include fungi and/or moss.


(ump) #3

W - I “encouraged” the colour fringing in 2 because it unified the small leaves with the large blur.

In 4, the point of the shot was to use a blurred central object to make the other leaves look further away.

I don’t see any halo/fringing in 1 though - could you give a better idea where to look and what for?


#4

I don’t see any haloing in 1 either.


#5

On the main trunk, third branch up, right side of trunk. It’s just that one section, just above the third branch. It looks like a halo artifact. It may not be that, but that bright area along the trunk looks out of place in the photo, to me. If that is what you were looking for, that is great. What you are looking for in your images is all important. :wink:


#6

Oh, to me that’s clearly a background element.

Distracting, sure, but not a halo.


#7

Now for my main critique:

  1. The background is a bit too busy for my liking. You could blur more, or have a lens with smoother blur of equal strength.
    Additionally, you’ve sharpened fine detail too much. Mask off the blurred background, or use a different sharpening algorithm, because that adds even more busyness to the background.
    Lastly, the branch on the bottom of the frame is distracting. I’d crop the bottom of the picture off; maybe make it 5:4 aspect ratio.

  2. I too am bothered by the strength of the longitudinal CA; that’s just a feature of your lens. Maybe defringe could take care of it.
    Otherwise, watch the left edge of your frame there; there’s a little tick coming in that draws my eye away from the right side of the image which should be the main focus.

  3. Nothing to say about processing here.
    I don’t like that the central in-focus branch is eclipsed by a foreground object. It’s really distracting. On the other hand, the diagonal branch on the left works well, though. It tends to draw my eye back into the photo, which is nice.

  4. Nice photo. Probably my only concern is the dense tree at the bottom of the frame disrupting the sense of open space around the central leaf.


(ump) #8

The background is a bit too busy for my liking. You could blur more, or have a lens with smoother blur of equal strength.

That’s not really useful to me: if I liked “bokeh ball” lens, I’d use them. I despise them, which is why I spent several evenings looking at sample images to find a lens with restrained bokeh. I appreciate your reply, but a critique based entirely on personal taste isn’t really useful to someone who doesn’t share that taste.

Additionally, you’ve sharpened fine detail too much.

By what criteria? It’s very possible that I have, but the problem I’m struggling with re. sharpening is exactly finding a defined process for tuning rather than “looks right to me.”

Also: are you sure that you’re not being confused by a lens that’s possibly a lot sharper then the ones you’re used to? This was one of the first images I pure through rawtherapee and I’m not sure that I’d found the sharpening…

I too am bothered by the strength of the longitudinal CA; that’s just a
feature of your lens.

It’s interesting and impressive that people can deduce that without knowing lighting conditions… No; it’s not a particular feature of the Takumar design - they’re about a 3 out of 5 in this regard, even by the standards of modern prime lenses. The shot was taken during blue hour against a glaring sky with the sun just out of frame.

Making pronouncements on the ability of a lens from a single shot without any idea what conditions were is not really a possible thing. To know how bad CA is you need to know the size of the object exhibiting the CA, various physical characteristics, and the light position and intensity, yes?

Maybe defringe could take care of it.

I should test that, definitely, even though I want to keep it.


(ump) #9

…This really cuts to the biggest problem with using rawtherapee: I can’t find any documents telling me how to the the parameters for sharpness and noise adequately. Possibly I should take a look at tutorials for Adobe products and try to translate…


#10

Post-processing is entirely subjective. If you don’t agree, then that’s okay, I’m just telling you my impressions.

No, I shoot with only Contax Zeiss lenses, and I have a Ricoh GR which has absolutely obscene pixel-level sharpness, to the point where I don’t even bother with sharpening.

Anyway, I wasn’t talking about overdone edge transitions. The oversharpening I mentioned was what I see in the full-res view on Flickr, which has a lot of very harsh noisy grit in the background blur where lens sharpness is utterly irrelevant.

What do you mean by “bokeh ball” lens? Ultra fast (think 85/1.2L) or smooth (think 135 STF)? If it’s the smoothness that you dislike, then we can agree to disagree, because in this case I can’t focus on what I perceive as the subject because of the background.

Longitudinal CA is pretty easy to tell when you know what you’re looking for: here I can see that the foreground blur is red-fringed while the background is blue/green fringed.

There’s no need to know how big the object is or what the characteristics are; simply compare what the blur looks like in front of and behind the focal plane.

In this case, defringe might only attack the red foreground fringing, since it goes blue-red-black. The background is just blue to black, so there’s no spurious color transition.