Stop Struggling With Shadow Recovery!

[gbodave]

That's possible. Also, though, there's really no advantage to shooting underexposed anyway. If you shoot under at 12,800 and pull the overall image brightness up by a stop back home, it'll still look like 25,600. So, might as well shoot what you need in the field.

Thanks for the succinct, realistic answer about what's best when in the field. I frequently see so much reference to scientific or mathematical theories, lists, prompt cards about arriving at a desired point I wonder how much time is ever left to actually take photos.

 

[Steve]

Thanks for the succinct, realistic answer about what's best when in the field. I frequently see so much reference to scientific or mathematical theories, lists, prompt cards about arriving at a desired point I wonder how much time is ever left to actually take photos.

LOL I hear ya. I actually have a love / hate relationship with these kinds of videos. I think the field takeaways are really important, but the background info on how sensors / ISO works always seem to cause the conversations to stray into academic details that aren't strictly necessary for a great shot

 

[Marco74]

That's possible. Also, though, there's really no advantage to shooting underexposed anyway. If you shoot under at 12,800 and pull the overall image brightness up by a stop back home, it'll still look like 25,600. So, might as well shoot what you need in the field.

You ave right Steve, very few times it makes sense to take pictures underexposed.

The main problem here is that we have learned with the reflex to set the limit of the ISO sensibility to the higher value acceptable to us, and here it depends a lot on personal tastes.
Now with the ML comes the problem of focusing accuracy. With low contrast atmosphere, the camera body struggles to lock in focus the correct point.
This also happens with some reflex cameras, but now it can also happen that on the field everything seems to work correctly, but in reality, at home, you realize that it is not.
To close my post, I would like to advise that it is always better to think outside the box, because the "rules" valid until yesterday, it is said that they are still valid today with the new technology.

 

[Steve]

You ave right Steve, very few times it makes sense to take pictures underexposed.

The main problem here is that we have learned with the reflex to set the limit of the ISO sensibility to the higher value acceptable to us, and here it depends a lot on personal tastes.
Now with the ML comes the problem of focusing accuracy. With low contrast atmosphere, the camera body struggles to lock in focus the correct point.
This also happens with some reflex cameras, but now it can also happen that on the field everything seems to work correctly, but in reality, at home, you realize that it is not.
To close my post, I would like to advise that it is always better to think outside the box, because the "rules" valid until yesterday, it is said that they are still valid today with the new technology.

Keep in mind that since the mirrorless cameras are using on-sensor PDAF they can also be affected by noise as well. If the noise overwhelms the detail it can potently cause issues achieving focus since the camera is relaying on finding the a phase differential based on subject detail. If all it sees is noise, it won't focus.

 

[Marco74]

Keep in mind that since the mirrorless cameras are using on-sensor PDAF they can also be affected by noise as well. If the noise overwhelms the detail it can potently cause issues achieving focus since the camera is relaying on finding the a phase differential based on subject detail. If all it sees is noise, it won't focus.

Yes, you are right. Here technical issues arise that go beyond my skills.
What I have described is what happened to me recently with my hiking partner.
He habitually adjusted his Z6 with a limit of 12800 ISO as that is the equivalent value to the adjustment of his D5.
On the other hand, after several tests, I realized that it is better to go up to 32,000, as previously written.
Now I don't want to start a debate on the noise resistance of one model or the other because this is not the place.
But in conclusion, after half a morning of taking pictures of some Deers shortly after dawn, he realized that his ideas were wrong.
I do not know beyond 32,000 iso if digital noise can create problems for the PDAF sensor, but until then, I have not encountered any problems.
Here is an example of a photo of that morning at ISO 25,600, 1/60" F.5.6 at 600mm.

You can only see EXIF info for this image if you are logged in.

Just to give an example of what I am saying.
The photo is slightly lighter than the light it actually was and it, not a crop, the subject was to close to us.

 

[bjanes]

The D5/6 electronics dramatically reduce camera-induced noise (read noise) as ISO increases. Camera-induced noise is highest at base ISO, but isn't often seen because at base ISO it is swamped by the great SNR of the captured light. However if you have to increase base ISO shadows hard, due to a lack of light, you are likely to see it.

https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Nikon D5,Nikon D6,Nikon D7000
https://www.photonstophotos.net/Charts/RN_ADU.htm#Nikon D5_14,Nikon D6_14,Nikon D7000_14

Compared to the D7000 (a great example of an ISO-invariant sensor), the D5/6 sensor has 5 times the read noise in DNs (Digital Numbers, the values recorded in a raw file) at base ISO. See the second link. But as ISO increases, the D5/6 sensor keeps it's read noise near base ISO DN values, while ISO-invariant sensors like the one in the D7000 increase proportionally with ISO, doubling with every stop, so you never feel like noise performance is improving with ISO on ISO-invariant sensors as it does on non-invariant sensors like the D5/6.

+1.
A great post with good links.

A bit of theory may be helpful to those inclined. If you look at a log-log plot of signal vs SNR (signal to noise ratio) as posted on Dxomark, read noise becomes important towards the left of the chart. The mid-range SNR is largely due to shot (photon noise) and PRNU (pixel response non-uniformity) comes into play on the right. In the mid-portion of the chart, photon noise is predominant and doubling the exposure raises the SNR by a factor of the square root of 2.

Dynamic range can be estimated by when the SNR curve crosses the noise floor. For engineering DR at base ISO, this takes place at -13.3 EV, so the engineering DR is 13.3 stops. A SNR of 8:1 is more reasonable for practical photography and the base ISO curve crosses this noise floor at about -9 EV. DXO uses percent saturation for the x-axis and I have converted this to EV. They use decibels for the y-axis SNR and I have converted this to a fraction for easy reference.

Input referred read noise in electrons is plotted by Bill Claff here. Input referred read noise translates to Bill's shadow improvement DR. Read noise is virtually constant for the D7000 and this sensor is said to be ISO invariant. Raising the ISO has no effect on the read noise, but the dynamic range is halved for each doubling of ISO. The Nikon D850 has a dual gain sensor and the high gain switch is at ISO 400. At this ISO and beyond, increasing ISO has little effect on read noise but decreases the DR as with the D7000. Bill's PDR plot for these sensors can be seen here

With the D6, read noise progressively decreases with higher ISO and if one is interested only in shadow SNR it makes sense to raise the ISO. However, as Jim Kasson points out in his essay on shadow noise with the Nikon Z cameras, there is no free lunch as the higher ISO decreases dynamic range as he notes in the quote.

However, in the part of the curves near the Claff PDR, the read noise is of such importance relative to the photon noise that the curves from two whole-stop ISO settings actually cross, which means that for those shadow densities and lower, you’ll see less noise at the higher ISO. Seems like something for nothing, doesn’t it? It is and it isn’t. Remember that you’re always better off for dynamic range at the lower ISO when you get away from those deep shadows.

For the technically oriented, Jim's post has a treasure trove of information.

Cheers,

Bill

 

[kimball]

For the technically oriented, Jim's post has a treasure trove of information.

Personally I absolutely love this stuff, thanks for sharing. I didn't see a link to Jim's blog post, so I googled it and found it here. And as an aside: Steve, sorry we've derailed your thread on your video. I realize you're way more interested in the practical, field consequences of gear (for good reason!) and not so much the engineering and theory behind it, at least not just for the mere sake of it.

Not only is Jim's post a treasure trove of info, but Jim himself is quite the renaissance man: he is extremely well versed in the science of photography (he worked as a color scientist for IBM for decades), while also being quite the artist. His photography has been featured in galleries, and for his latest exhibition at the Monterey Museum of Art he put together a wonderful video about his development of what he calls a "Chronograph":

His youtube video is a really impressive tour of his work. Highly recommended.

 

[ruley74]

I don't use LR so can't say for sure your answer. I do use Affinity Photo and Pixelmator Pro quite a bit as well as Nikon's software. The process I use is I zoom in the image on the shadow area I want to pull up. I go to 100% or maybe a little more. I adjust shadow, brightness, exposure etc. sliders until I start seeing unacceptable noise then back off until I can live with the noise. When the photo is viewed at "normal" size, if I think the shadows are still too dark or there is still much noise, I then use the delete key and move to the next image.
(edited to correct my horrible typing skills)

I'm pretty similar, start with the overall exposure tweak and then move to a slider for final adjustment. I use DXO and but have LR, i only referenced LR as standard tool. With DXO i have a limit on the slider that i don't go past as i feel past a point the image is negatively affected, so if i can't achieve the desired look with that boundary the image is discarded.

I the question was never in relation to processing an image, simply curious on correlation of stops within the range the slider affects.

 

[bleirer]

The lightroom exposure slider is indexed in stops, +5 to -5, and the slider is linear, but I you make other adjustments such as curves or blacks, shadows, whites, highlights then it can seem nonlinear because of the impact of the other sliders. So I think the range you could achieve is theoretically greater than 5 stops, not that you'd want to.

 

[bjanes]

Personally I absolutely love this stuff, thanks for sharing. I didn't see a link to Jim's blog post, so I googled it and found it here. And as an aside: Steve, sorry we've derailed your thread on your video. I realize you're way more interested in the practical, field consequences of gear (for good reason!) and not so much the engineering and theory behind it, at least not just for the mere sake of it.

I can't speak for Steve, but I don't think we have derailed his thread but have merely supplemented it. If you read his e-book "Secrets for Exposure and Metering for Nikon" he has two chapters discussing pretty advanced technical matters that are well worth reading:

• Advanced Exposure Techniques
• Advanced Exposure Considerations

Bill

 

[Steve]

I can't speak for Steve, but I don't think we have derailed his thread but have merely supplemented it. If you read his e-book "Secrets for Exposure and Metering for Nikon" he has two chapters discussing pretty advanced technical matters that are well worth reading:

• Advanced Exposure Techniques
• Advanced Exposure Considerations

Bill

Nope, not derailed IMO.