Mike, sorry for that. I simply downloaded your files and that’s how they were saved on my side (retried the download, but no idea why the “|” is left out).
I think the best way for me to learn this, is to start with a “clean” image, and re-do each of the five steps you have listed. When I see the individual steps like this, as I do now, it does seem logical, but that doesn’t mean I can do it yet on my own!
Well, to fully experience you might have to do it yourself.
I left the tool names visible and added some description so it’s easier see / realize, which tool you are dealing with. You might want to disable / enable the masks to see the effect.
“Control point - wall darker” is the most obvious example
“Auto mask - exterior less prominent” invokes small corrections
Just remember, your screenshoots don’t show the (numeric) correction values. Most of the correction masks (your local adjustments) act on their own, but for a satisfying result they (have to) work together.
As the saying goes “Rome wasn’t built in one day” – take Your time!
This is all good - I don’t want to replicate your “numbers”, just what you were trying to achieve, and the only way it will become part of MY toolbox is to do it on my own, and NOT to blindly copy you.
I want to concentrate on ONE thing at a time. I took some good photos of last night’s sunset, and some interesting photos of Halloween on Lincoln Road in Miami Beach, but I want to do the “iguana photo” before I move on. I expect I’ll need some help - maybe. I very much like your finished version (with the possible exception of the. crop) more than mine, and now that I know what changes are helpful, I need to do them, one at a time.
That’s not the intention – there are always different possible solutions.
When suggesting to disable / enable those existing corrections it’s about to see first
before you start from scratch. – Otherwise there is no ‘need’ to give you an example.
– But hey, you’ll discover on your own!
Hello, It is realy interesting theme and I learned alot from it. Thanks. If you allow me some offtopic comment:
Sunstars or starburst is a purely optical effect that is obtained from the edges between the aperture blades.
Each lens that is closed enough can give such an effect, but each is different depending on the number of blades and the size of the edges between them. The more clearly they stand out in the aperture ring, the more pronounced is the stellar effect. Usually more expensive lenses have more rounded blades and are therefore more difficult to make such a star, but not nesessarily. The other important thing is the number of blades - when it is even - the number of rays is equal to the number of blades, and when it is odd - the number of rays is twice as many.
Personally, I’m still learning to use this effect, but what I’ve found is that the smaller the light source is, the more pronounced is the star. The other key condition for a good star is that the background sould be significantly darker than the light source, so this effect is most easily obtained when the sun shines through a small gap between objects.
There are a lot of articles on Google on the subject that can be useful to you for example this, but experimentation is the main way to master this technique, and you should be careful with the Sun - not to burn your eyes or the matrix anyway. I think it is safer to experiment with lamps at night.
Very, very true! To be able to achieve what you did, I think it will be VERY helpful to understand what you did, and how it worked. Lots of learning, just as I do with Joanna, but then I need to create a “blank slate” and start from the beginning. I’m already way ahead, as you “see” things I didn’t. I’m much better at “seeing” other people’s photos, as when I view my own, I’m too likely to accept them the way they are, and less likely to change things. I used to squint my eyes, and look for “shapes”, and then adjust, without being lost in the details.
It also bothers me that the photo I posted reminds me of what I actually saw. Your version is better than what I saw, and what the camera captured, but sometimes it’s difficult for me to think “hey, how can I improve this?” My window of potential improvements is 100 times more powerful than before, but my brain “accepts” what I did, and struggles to find improvements. In retrospect though, your image is what I “imagined” as I took the photo, and is not an exact copy of the image created on the sensor. You and Joanna are both SO, SO much better at this than I am capable of being - I need to correct that, and be much more critical of the images I capture.
And very importantly, I need to understand, and know when and how to use all the new tools I have at my disposal, and how to do this correctly, not blindly. I used to love “ClearView” until I realized how terrible it made my images look.
The “pipe” sign is often used in command line interfaces (terminals) to concatenate multiple commands. Some systems consider the pipe to be a special character that cannot be used in file names. I don’t use it in filenames.
Good show,
Two things to add.
1 first click on a shape takes a sample of the colorbrightness. I am fairly sure it’s autoselection is luminance based. So not color based. But i could be wrong.
So a edge by luminance difference of the object and the surrounding is key for a clean selection.
2 one big drawback is the autoselection field, outer ring does have the same blue masking so you don’t see the selection properly unless you use desaturarion or exposure to light up the painted object.
And with reverse mask you get a gab. Which is the automask area.
I would like to have a “clean autoselection field” button so only the mask is over the selection area.
Now i use ALT to get eraser and clean it my self, much more work.
I think this is where you might try “previsualising” the kind of image you intend to create and try not to be swayed by what you saw, remembering that your eyes can see so much more than the sensor and that the RAW file can record.
I believe the trick is to always assume that what you see when you first open the file is wrong, mainly because what you see is the result of a machine process to visualise the raw data. It is unusual for the demosaïced image to be “perfect” - but it does happen occasionally
Never think that. You are just as capable of doing better than anyone else. It just takes practice
I’ve been using Photo Mechanic to transfer my (organized) images and files onto my Windows, and now Apple, computers, for ages. It results in files and folders looking like this:
Everything is neat and orderly, dates to the left, and details to the right.
I wasn’t aware of any problems until now, either on Windows or on Mac, but things may have changed.
I changed the order for images, as I’m usually searching for the file name first, compared with folders where I start by searching by date.
PhotoMechanic “ingests” all my images from an SD card, and then I can select which ones to keep, and which ones to delete. When I’m done, I have a folder full of images that I think may be useful, and all the “trash” has already been deleted.
What I still need to do, is to find a good way to use “keywords”. Lightroom had their way. PhotoMechanic now has a different way. I think that PhotoLab has yet another way.
As soon as I have time, I need to learn how PhotoLab does it, and decide if that will do everything I need. That’s a question I’ll probably ask about in a week or two, once I try to read about it wherever DxO has that information.
If I had thought about this long enough, I might have realized that, but it became completely clear once I read the above!
Something else that is completely new to me! What was that line, about one picture being better than a thousand words?
This, along with what preceded it, and what followed it, is again something I knew NOTHING about, but now, it seems so simple! Gack!!! …and Amazing!
Another gem of advice!
I need to stop, and go out on the balcony, and just relax for a while. For me, it takes time before all this settles in, and I can use it effectively. …and to re-read this again tomorrow.
Yes, there are many ways to edit metadata, as many as there are apps that do it… PhotoMechanic made a good impression when I tested it a few weeks ago. If PM is your first step in your workflow, I’d use PM as the metadata handler and use PhotoLab for everything else. Using exactly one app for metadata will save you a lot of trouble, no matter if it is PhotoMechanic, Lightroom or PhotoLab.
I think that’s excellent advice. I no longer use Lightroom, but all my Lightroom stuff is “history”. I still have copies of what I did before I upgraded to the rent-a-lightroom plan. But I have no plans to go back to Lightroom, and even if I lose my old editing work, I still have all my images.
I did buy the latest version of Photo Mechanic, that has all their new add-ons, so I’m ready.
Time to see if I’m learning. According to PL5 there is no clipping, although I had to work on the sun to accomplish that. I spent the last half of my editing trying to bring out the radial lines coming out from the sun, after which I made the photo warmer, to match what I felt, although the sky was more blue than what I have created. The export to disk worked - I thought I was going to have a problem with that again, but no complaints from PL5. I like the way the buildings came out, and that I can still see some color in the boats, and I took the photo just as the sun was peeking out from behind some clouds, which worked fine. A split second later, it was too bright.
I like the end result, but I don’t like the way I got to it, with a lot of trial and error. I thought about cropping off the top of the image, but I decided to leave it, as it balances out my watermark.
I should add that I don’t understand the patterns in the sky. I’m used to seeing the radial lines, even if it is so difficult to capture them, but here I also see horizontal lines interacting with the radial lines.
I’m also not sure if I have the horizon level, but if I change the horizon so it’s lower at the left, the buildings lean to the left. I think it’s an optical delusion, and it’s really level, even though it looks otherwise.
As per our discussion, I set the aperture to f/10, Manual, and I raised the ISO to 500, which I now see I didn’t need to as the shutter ended up at 1/2500 second. I tried to keep the meter centered while aiming at the sun, but in retrospect, I had center weighted turned on, not ‘spot’. Me bad.
Now that I think it’s done, I love it. I’m sure Joanna and Wolfgang, and maybe others, will show me what I messed up or missed.
There is also a bird in front of the buildings, that nobody will ever notice, and I suspect some other specs may be sensor dirt, not birds. I need to check this out in a test shot, but that’s for tomorrow.
Setting the horizon level is almost impossible here because we see no horizon. In cases like these, we can use the buildings as reference instead, they are strictly vertical in most cases.
If we draw center lines (disable crop, set grid spacing to 999 and switch the grid on), the center of the grid marks the optical center/axis. vertical lines should be vertical near the center line. Vertical lines near the image edges, but located near the middle horizontal grid line should be vertical too - or leaning symmetrically (inwards or outwards).
As far as I can see, you pointed the camera upwards. The leftmost building seems to lean inwards, while the rightmost building looks straight. You could use the two lines tool (which is difficult to use in such circumstances) to fix this - or the perspective tool and the slider that corrects verticals. Some slight CCW rotation is needed too.
Here’s what I get:
Note that I bulged the tone curve to better see horizontal and perspective corrections. Rotational correction: -0.08. Vertical correction: -4. I also set the crop to better show the rotational and perspective “leftovers”. A somewhat brighter than default background helps to see this better too. _MJM9478 | 2021-10-31.nef.dop (11.8 KB)
Yes, I wanted to capture more of the sky, so I aimed the camera slightly upwards. I never thought that slight amount would be noticeable on the image, but you’ve proved that it is.
I guess the moral for me to remember is keep the camera level, meaning the “base” of the city buildings should be centered vertically in the image.
I’ll have to find the perspective tool later today, and figure out how to use it. The slight rotation is what bothered me while looking at the image - as you noted, I need to correct that first, before I work on the perspective. Maybe Nikon could add an electronic “bubble level” to help people avoid this. I wonder if it comes with the Z9?
No mechanical shutter, no mirror flopping around, lots of on-screen data, perhaps including a bubble level, and perfect focusing on an “eye” when desired… When I read this article from The Verge, I see a major change in the works for all professional cameras, and Leica’s new M11 is already “obsolete” even though it won’t be released for another week or so. To accomplish Joanna’s technique to not clip the highlights, even that could be controlled electronically with the new system, if a camera manufacturer chose to do so.
In the meantime, by following Joanna’s suggestions, I actually have included (part of) the sun, without clipping - I’m amazed that it worked so well. I love the sky, but notice that in your lighter version, the special effects in the sky are washed out. If I do this again, I will do your corrections first, before I do anything else. Thank you!
No. the answer is to realise that you have the same ability as a view camera to apply lens shifts, but in PhotoLab instead of in camera. Keeping the camera level in this case would have meant a poorly composed, out of balance, image.
Seriously though, that kind of clipping is unavoidable with the sun in shot and the truth is that you exposed as well as was possible to be able to deal with it with a minimum of correction to the tone curve.
I need to remind you to select larger areas for the Smart Lighting zones. It’s not just a matter of choosing the darkest and lightest areas, you need to give the tool a bit of context to work with but selecting the immediate area around it.
I’m not sure why you created a Control Line for the sky but selected one of the buildings with the pipette when it was the sky you wanted to change.
As you will see, I also added a subtle Control Point to the buildings to try to control the flare and a standard grad filter on the water.
Just as a footnote, in the early days, you used to be really concerned about “crunchy” water and clouds. I notice that this has changed but now I find you are making things even crunchier than I would
So many thoughts, so let me ask you about one at a time.
I went directly to Biscayne Bay after taking my Halloween photos, so no tripod, not much of anything. If I want to be serious about this, I need to bring a tripod. I also need to level the camera perfectly, so it isn’t canted one way or the other. I didn’t even think about the D750’s built-in level tool, but when the camera is on the tripod, that will be great. Hand-held, not so much, too many things going on at once, and I know I can “fix” errors in PhotoLab. So, my question for you now, is whether I should make the camera perfectly level in all directions, and zoom out slightly so I can crop to what I want later, or should I aim the camera up slightly, knowing I can fix it with the perspective tools as you have shown (and which I never even thought about until you brought it up)? My gut feeling is keep the camera level, and with such a small amount of cropping, do that instead of the perspective tool. (Or, buy a proper view camera.)
About the bubble level, I’ve started taking the Nikon D750 manual with me when I go places, and in may free time, reading it starting at the beginning. I also bought a book on the D750, which I haven’t started to read yet. With what you’ve shown me, I can start doing it this way (but unless the camera is on a tripod, I doubt I’ll be using “Live View”). In any case, instead of concentrating on the image, I need to start concentrating on the tools, and then the image. With tripod AND Live View, everything will get easier.
I’m confused - if I can remove that clipping in PhotoLab, isn’t that effectively what I might have done in lowering the exposure? I did see this in the camera, but not so much as I see in your photo.
For me, everything is a compromise, and I’m pleased that you accepted what I did. Given time to think, I might have left the camera settings as they were, but for dropping the ISO to 100, so two more stops of underexposure, but every millisecond I delayed meant more and more of the sun was coming out from behind the cloud. I don’t think that fast, let alone think and act that fast, but next time I’ll be remembering what you said, and deliberately under-expose some more. (I haven’t tried bracketing, but I’m very tempted!).
I mis-understood. I made the area as large as I could, without including “wasted” stuff. I’ll try it your way next time, and have a larger area, that includes the very bright area I was interested in.
I kept changing my mind, as I was editing, and the image was improving little by little. In retrospect, I wish I had stopped, made a VC, and tried things differently. As it was, I moved the “pipette” around, and the spot where I left it seemed to make the sky much nicer - the radial lines started to appear. I guess it was doing the wrong thing at the wrong time in the wrong place, but I liked the improvement in the image.
One last thing - I prefer the sky in the final image I posted, to the sky in the image you finished with. The radial lines stand out more in my image, and they are not so noticeable in your image. I’m guessing this is because your sky is “lighter”?
Added later - I agree about removing the clouds at the top right corner, but when I see the image without them, it feels “unbalanced”. Maybe it’s my imagination, but without them, the image feels somehow “uncomfortable”. That added “stuff” at the top right makes the image feel balanced to me, but I would have a very hard time describing “why”. It’s just something I “feel”.
I agree with you - From now on, I will be careful NOT to edit the water. It does look more realistic in the un-edited photo.
How about the sky? Was I “wrong” to make the sunbeams so obvious? Maybe I over-did it?
Of course, that could lead to why the buildings are seen so clearly in the edited images, but in the un-edited image I just posted, they are all black silhouettes in front of a colorful sky.
How much editing is “just right”, and when does it become “too much”?
I’m beginning to think that I crossed that line, and Joanna stopped short of crossing it.
