Editing high dynamic range images in PhotoLab 5

DxO PhotoLab
61

“G” key works fine! Thank you. I’ve been wondering about that for a while, but never got to ask.

Regarding the aperture, I’m thinking that from now on, I should try for something near f/10 if possible. I may get another chance this evening, weather depending.

Seriously though, suppose I switched from f/13 to f/10. That’s sort of around 1 stop, roughly. That would have meant with more light reaching the sensor, I could have probably doubled the shutter speed, maybe around 1/720th or whatever is close to it. Higher shutter speed = less chance of movement from hand-held photos. Sounds like a win/win situation to me, less chance of errors from movement or diffraction.

2 Likes
62

This post about “technique” could go anywhere, but I put it there because it is as good here as anywhere else…

Those who have followed the longer threads going to and fro between @mikemyers and @Joanna have noticed that Joanna often posts about using f/10 for best results, which made me wonder why f/10 and what the dependencies are. A few things can be regarded as common knowledge:

  • close the aperture to get sharper images - up to a certain point
  • close the aperture beyond that certain point and images will look less sharp

What is it that defines that certain point? It’s the point where the improvement in imaging quality of the lens is overruled by diffraction, also depending on pixel density of the sensor used. While the perfect lens would be able to resolve any amount of detail, diffraction will make that lens go soft anyway…

Here’s how you can calculate the f-stop, at which diffraction will take over

  • 3200 x H / P, where H is the sensor hight in mm and P is the number of pixels in that direction

Examples:
My 24 Mpixel EOS M6 has a diffraction-takeover f-stop at 3200 x 14.7 / 4000 = 11.76
→ Going beyond f/11 is counter-productive, no matter how good a lens I’d use.
→ Good practice would be to stay 1 stop below that value, f/8 and f/9 should be okay.
A full frame 24 Mpixel camera has a takeover at 3200 x 24 / 4000 = 19.2 (stay below f/13.6)
A full frame 54 Mpixel camera has a takeover at 3200 x 24 / 6000 = 12.8 (stay below f/9)

The limiting f-number drops with decreasing sensor size and higher megapixel numbers, we all knew that. The formula above is a model for calculating said limit with a theoretical, perfect lens and for shooting with the optimal compromise of sharpness and DOF.

1 Like
63

Assuming this applies to the 24 megapixel sensor on my Nikon D750 and to my Leica M10, that means a “safe” value to select for my aperture would be f/11 (and not f/16).

This means to use the techniques Joanna has been teaching me for high dynamic range photos (sunsets), I will use ISO 100, place the camera in spot mode, add +1.7 exposure compensation, meter on the brightest part of the image, and with the aperture set at f/11, I will adjust the exposure “meter” at the bottom of the viewfinder to “centered”.

As I understand it, diffraction is a physical property, and it’s impossible to change it, but doesn’t Apple do something like that in software to minimize the effects of diffraction on their tiny sensors?

I guess I was not doing right by myself years ago, when I might select f/16 or f/22 thinking it would improve my image - and I now understand it will degrade my image. Thanks for making this so clear.

64

I hope it doesn’t muddle the water too much by asking this question here, but it does seem appropriate. I was going through every setting on my Nikon D750, and there was a setting highlighted in my viewfinder, “ADL”. I’m pretty sure this is the explanation (for what Nikon refers to as “Active D Lighting”:

https://www.nikonusa.com/en/learn-and-explore/a/ideas-and-inspiration/balancing-photo-exposures-with-nikons-active-d-lighting.html

Knowing that my images will be edited in PhotoLab, and that I’m only shooting in RAW, is this a useful tool? …and if so:

“Active D-Lighting’s level of adjustment can be set to operate automatically or manually at four different levels (Low, Normal, High, Extra high); it’s also possible to bracket pictures to get one with Active D-Lighting and one without so you can make the decision later as to which looks best to you.”

For Joanna, since I’m sure your Nikon D850 has the same thing, do you turn it off, or is it useful to you?

65

Mike, leave that Active D-Lighting OFF.

I used to use it at a low setting, but had to realize, that PL doesn’t handle it properly
(different in Nikon’s own software). :slight_smile:

1 Like
66

Well, there might be the rare case you need all available depth of field and you simply make use of it.
Personally, I avoid going beyond f8 (f11 maximum), depending on the lens …

67

Wolfgang, I turned that setting on many years ago, at the default value, and then forgot about it. All the photos I have posted in this forum from my D750 have been shot that way.

I suspect it only affects jpg mode, not ‘raw’; I need to verify this.

I’ve got to admit, for photos I’ve been taking with this forum in mind, all of you were correct, and the D750 is so much easier to get the types of photos I’ve been taking recently than my M10. With the M10 I’d be “guessing”, but with the D750 I’m more assured of getting what I want. On the other hand, ISO 100 isn’t the best choice for me for any kind of “action” photo, even in sunlight, as the shutter speed is quite slow.

I did use the 3.5-4.5/24-85 for all photos today, and hope to use it at sunset. I will read the two links you’ve posted, and I’m a little into the D750 book that I bought.

68

Unfortunately it also affects raw and PL doesn’t recognize it. – Better to switch it OFF.

The provided links (see on Mansurov’s site → Review → Lenses for other lenses) give you an idea about the possible resolution of different lenses. The overview shows at what focal length / f-stop the lens is best in the center, the middle and the corners …

1 Like
69

This is the most detailed answer I could find, and as a result ADL is now turned off.

https://photo.stackexchange.com/questions/37958/should-i-use-active-d-lighting-when-shooting-raw

Will I ever use it? I dunno, but for the images I’m taking for processing in PhotoLab, I won’t.

It has another side effect - when it’s turned on, the image I will see on the back screen is a ‘jpg’, which will certainly be influenced by ADL.

My own conclusion is while some people love it and others won’t use it, there no longer seems any point to it if I’m manually correcting images in PhotoLab.

70

I turned off ADL years ago. All it seemed to do was avoid over-exposure by under-exposing - something I was perfectly capable of doing myself and only when I wanted to, not when the camera decided.

According to the calculation spreadsheet I made up, based on George Duvos’ calculations, f/10 is the absolute optimum aperture, but he reckons that you can go up to f/14 with only slight softness due to diffraction. If you can reasonably use f/10, why not stick at that instead of f/11? After all, it gives you a tiny bit more light.

My iPhone 12 has a 12Mpx camera with a 4.3mm focal length with a fixed aperture of f/1.6. If I run its blur circle of 8µ through TruDoF-Pro, that gives me a diffraction limit of f/4. But, when you take into account the full-frame equivalent focal length of 26mm, you are dealing with a fairly wide angle lens. In fact, if I calculate the hyperfocal distance, it comes out at 67cm with a closest sharp distance, without diffraction, of 34cm (13"). In that case, why would you need any computational correction to avoid diffraction when you are getting everything acceptably sharp, without diffraction from 13" to infinity?

71

I can’t post the original RAW, because the file from the D850 is 95MB, but here are the before and after exports of a shot I did at about 13:00 this afternoon of a very dark bollard against a very bright sea.

I measured using centre weighted against the sea, with the usual +1.7 compensation. I didn’t use spot measuring because all the specular highlights were upsetting the metering as waves appeared and disappeared.

The only correction on the second one is Smart Lighting, spot measure with one zone on the brightest sea and a second on the front face of the bollard, of 50.

_JNA0022_M_DxO
_JNA0022_M_DxO1024×1536 143 KB

_JNA0022_1_DxO
_JNA0022_1_DxO1024×1536 169 KB

The only reason for doing the shot was to demonstrate what can be done in extreme contre-jour lighting as the image is not exactly an artistic masterpiece.

1 Like
72

I didn’t realize the D750 could be set at f/10 so easily.
Done deal, from now on, that’s what I’ll use.

I did go out after a sunset photo tonight, after going after squirrels and iguanas late this afternoon. The 750 made iguana hunting effortless, and I finally got the shot I really wanted, a male and female together, with the male showing off. The squirrel was easier to capture, but 24-85 isn’t long enough for squirrel hunting. I need to process the images, and post.

What is a “bollard”??

The lower image clearly shows more tones at the bottom, while the upper image looks black.

73

Just a Nikon question. What is the difference between Active Day Lighting and heighlight weighted metering for a raw image?

Another link dealing with diffraction Diffraction Calculator | PhotoPills. When playing with it you must enter the camera model and the number of megapixels.

George

74

Might I ask how you are setting the aperture? Turning the front wheel on the right side, in front of the shutter button, means it changes in ⅓ stop increments. Helen asks, were you using the manual aperture ring on the lens?

And then I found another way of wasting/passing half an hour laughing until I cried at this post…

Especially some of the tongue in cheek comments to the post :rofl:

Indeed. It still amazes me just how much detail PL can pull out of what looks like a totally under-exposed RAW image.

I get a sneaking feeling, it’s a bit of marketing hype, creating two mysterious terms for “under-expose on purpose” :crazy_face:

I’ve seen that and it is not as accurate as TrueDoF-Pro because, whatever you enter, it always assumes a CoC of 30µ, when it is actually linked to pixel pitch as well as sensor size.

75

I think it is another coc. It is the max dot a human can separate from a circle at a given distance. From this value the mac coc on the sensor is calculated. No time to check it now.

George

76

Hi George,
apart from the manual’s information, the ‘best’ answer about Active D-Lighting
→ https://www.nikonimgsupport.com/eu/BV_article?articleNo=000005935&lang=en_GB&setRedirect=true

ADL affects the dynamic range, depending on the settings and works with P, A, S, M metering modes. From the side note in the manual about random noise, uneven shading, shading and halos (pdf p.204 / print p.176) I conclude, there is some in camera processing going on. – I only ever used it at Low setting, but switched it off with DxO. (Maybe one has to do a ‘scientific’ test with FRV.)

Highlight-weighted metering can only be used with Spot metering. In my quick & unscientific test it reduced the exposure by 1 EV. The manual (pdf p.385 / print p.357) indicates how to use it with the assigned Fn button (I have set Fn to My Menu 1st topic).

So, ADL is still somewhat mysterious to me. Hopefully somebody else knows better. :slight_smile:

77

From what I have read from George Duvos’ writings, when talking about diffraction, it is more accurate to talk about blur spot diameter. This can be determined by simply taking the pixel pitch and multiplying by two; that being the size necessary to record a single dot, in case it covers more than one pixel.

Using that size works reliably with TrueDoF-Pro to give excellent control of DoF without diffraction. But is does mean that, with something like the Nikon D810, the measured spot diameter of two pixels gives a blur spot diameter of only 10µ which, in turn, gives an optimum aperture before diffraction of only f/5.6, which is too restrictive. Hence, why I have settled on 20µ as a compromise because it, at least, allows me to use f/10.

78

Diffraction is independent from sensor size or pixel count. A bit of, but is dependent only on the top angle of the cone aperture-sensor. That cone produces an airy disk and becomes vissible when it is 2 till 2.5 times as big as the sensel size.
Or when it becomes bigger as .3mm on a print. And thats what is ment in Photopills calculator I think.

George

1 Like
79

Joanna, and Helen - the problem is in my head, not with the camera or lens. I have lived my whole life using the “click stops” on my lenses - f/1.4, f/2, f/2.8, f/4, f/5.6, f/8, f/11, f/16, f/22, f/32… …unless the lens had f/3.5… and likewise with shutter speeds, although different cameras confuse this with different choices. Yes, now that I tried it last night, the Nikon D750 easily goes right to f/10. Until now, I wasn’t even aware I was doing this - it just happened seemingly automatically. How do I change the aperture - I always did it on the lens. With the D750, with no manual control on my lens, I use the “sub-command dial”, and it obviously and clearly stops at f/10.

It’s not that I’m not “thinking”, as I must be thinking, but my mind is concentrating on something else, and certainly not on the aperture. I must have “seen” f/10, but didn’t notice or think about it until now. Again, “me bad”. 99% of my mind is thinking about the image, and the controls sort of get in the way.

Thanks to all of you I set the camera in order - set camera to M, set metering to spot, set ISO, set aperture to f/10 (now), set exposure comp to +1.7, focus, center exposure meter while aiming at the brightest part of the image, and carefully take image. (I have found if I don’t follow the full routine, I likely forget something.)

I thought all the other features of the D750 were de-activated, but I hadn’t noticed ADL yet; now it’s off.

Aha! I know what they are, as that’s how ships are tied up at the docks, unless people use “cleats”, but I don’t remember “bollard”. Maybe there’s some other word for it in the USA? Pylon comes to mind.

However, I probably knew and forgot this: Bollard is originally (AFAIK) a term of nautical jargon. Aboard a ship or on a dock, a bollard is the post about which one wraps a mooring line to secure the ship to the dock.

Yes, I loved the comments too!!!

It’s mysterious to me too, but since I’ve selected PhotoLab to be my image editor, there is no point in using anything on my camera that can create issues with PhotoLab that I’ll never figure out because of not even realizing ADL was turned on.

I’d like to say Nikon has far too many things that can be turned on, but for some people it’s probably good that Nikon makes them available. I’d prefer to have access to them, and not need them, than the reverse.

I’d also like to add that what I’ve done with my D750 feels like I’ve converted it into a simple “box camera”, although a ver fancy one. :slight_smile:

80

Heheheh. It’s amazing how many habits we learnt a long time ago can be difficult to “update”

I suppose I got used to using ⅓ stops with LF lenses, where the shutter offers a continuous scale without detents…

IMG_1208
IMG_12081536×1152 108 KB

Yes, it is somewhat deeply buried in the abyss that is the menu system.

Absolutely. In some ways that’s what I like about the Nikon DSLR cameras that I have owned. If someone wants to, they can use them in “absolute idiot mode”, or they can take full charge and use use everything manually.

Which reminds me of my very first camera, at the age of 11. It’s on the top left in this slide from a presentation in which I wanted to show my analogue photography evolution…

Capture d’écran 2021-10-27 à 14.36.23
Capture d’écran 2021-10-27 à 14.36.231272×887 69 KB