M10 vs M10-R - Real dynamic Range/Highlight Recovery difference

[SOHODE]

Advertisement (gone after registration)

The M10-R has a better highlight recovery than the M10. However, the photonstophotos chart shows that the dynamic range of the M10 and M10-R is very similar and the shadow recovery of the M10 is a bit better than that of the M10-R:

https://www.photonstophotos.net/Charts/PDR.htm#Leica%20M9,Leica%20M10,Leica%20M10-R

Do you think the M10-R distributes more of the available DR to the highlights than the M10 does? and that is why the M10-R has better treatment of highlights or, even, more dynamic range than the M10? but as shown in the charts linked above, that the shadow recovery of the M10-R is not as good as that of the M10 — and in some other comparisons the M10R has more noise than M10 .

I'm wondering if we shoot with M10 1 stop under what we shoot with M10-R (both at ISO 200) and then add 1 stop exposure to the shadows (for M10 picture) in post processing, how much the highlight recovery and shadows details/noise will be different between those camera? Has anyone did this comparison?

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[jaapv]

No significant difference...   A third of an EV is less than the tolerance of your exposure metering.

 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Adam Bonn]

I think Jaap's nailed it TBH.

The M10R is more tolerant of a user mistake regarding highlights (clipped highlights are user mistakes, not camera problems - and we've all done it)

I suspect that seeing as true base ISO isn't a user selectable setting on the M10 🙄, ISO200 and up exist as 'push' so the camera is tending to underexpose which means that the user gets more shadow data to play with.

[SOHODE]

I'm wondering if the chart shows the recoverable highlights too.

Some M10-R users noted they were able to recover highlights 1 stop more than M10. (link below - min 7:50)  I'm wondering if that's just a 1 stop DR shift towards highlights in M10-R or the M10-R dynamic range is really 1 stop wider? Comparing the shadows (details and noise) maybe can give us a better a comparison. 

https://www.youtube.com/watch?v=nyJqsqjKUpg

 

Edited April 21, 2022 by SOHODE

[jaapv]

I guess it has nothing to do with DR. Recovering highlights is done by interpolating the non-blown colour channels to create detail. It is quite possible that there is some difference in the algorithms used. There is certainly a difference in the amount of detail to recover. 

[SOHODE]

Clipped highlights are not always user mistake. In high contrast situations when you try to expose the subject the bright part of the picture gets blown out and this is limited to the dynamic range. Wider dynamic range give you the ability to recover some of the blown out highlights. 

My question was if this value is really one stop wider in M10-R? or it's mostly a software adjustment in M10-R firmware. (or both of them)

I'm not really sure if that chart is seeing/including the recoverable highlights too.

Thanks

[jaapv]

Advertisement (gone after registration)

Well, that was what I said… At any rate, coming from decades of slide film with a DR of five stops, I am not very much impressed at underexposing a bit to save a highlight with a thirteen stop range. 

[SOHODE]

Jappv - I agree. 

Some people think Leica just shifted the DR towards highlight for M10-R so it can keep more details in highlight after the clipping point and also I read some people said the shadows details and recovery is better in M10. If that's the case I'm wondering if shooting a half stop under exposed with m10 (to save the highlight from clipping) and then adding a half stop to shadows exposure in post-processing can give a very close result to M10-R? What do you think?

[Jeff S]

@jonoslack distinguished between dynamic range, base ISO and highlight recovery in discussing the M10-R compared to the M10.  He was not comparing dynamic range; rather the advantage in his experience of recovering blown highlights (for whatever reason) with the M10-R.

 

Maybe Jono will chime in here.

Jeff

[jaapv]

8 hours ago, SOHODE said:

Jappv - I agree. 

Some people think Leica just shifted the DR towards highlight for M10-R so it can keep more details in highlight after the clipping point and also I read some people said the shadows details and recovery is better in M10. If that's the case I'm wondering if shooting a half stop under exposed with m10 (to save the highlight from clipping) and then adding a half stop to shadows exposure in post-processing can give a very close result to M10-R? What do you think?

That would only mean that they adjusted the exposure curve to pull down the highlights and brighten the midtones and lift the shadows adjusting the black point at the same time in-camera before writing the DNG   That would certainly produce this effect if they adjusted the ISO value and exposure metering to match. Consider that a pixel produces an analog signal from base noise to 100% fill. The curve produced in between those two points can be modified. 
The user can do a similar manipulation in postprocessing if he underexposes slightly, but this would make the process easier  

I am sure that Leica can do more tricks than I can think of when designing their firmware. 

[jaapv]

Looking at Jonos post we can see that he started with quite a flat image which supports my supposition. 

[GFONG]

I don’t know much about technical aspects.   Below is my guess based on my experience and information from Internet.

The M10’s highlight blown out is because of the base ISO is around 160 (read from Internet). When using ISO 100, it is actually using the ISO 160. As a result,  in some circumstances the highlight is blown out because of 1/2 stop overexposed. That’s why Lecia changed the min ISO in AutoISO to 200. Also people in Internet suggested to use ISO200 or -2/3 EV to avoid blown out.

For the noise, I found that the M10P is one to one and a half stop better than the M10R. For example M10P’s images at ISO 8000 is cleaner than M10R at ISO 6400. Also by looking at M10R’s images at ISO 400 and above, you can see the noise.  I don’t know why people in Internet keep saying the M10R high ISO is better than M10.

 

 

Edited April 22, 2022 by GFONG

[Adam Bonn]

15 hours ago, SOHODE said:

Clipped highlights are not always user mistake. In high contrast situations when you try to expose the subject the bright part of the picture gets blown out and this is limited to the dynamic range. Wider dynamic range give you the ability to recover some of the blown out highlights. 

If the DR of the scene is within the DR of the camera but the user doesn't manage to capture the full DR then it's a mistake (that we've all made) or intentional and no problem

As Jaap says above, the listed difference of DR between the 10 and the 10R is very slight....

DR is not limited to highlights. A camera that offers 1 stop of HL recovery and 1 stop of Shad recovery has less exposure latitude than one that offers 0.5 stops of HL recovery and 2 stops of Shad recovery.

The trick with any digital camera is to make an exposure that works well with your RAW software and RAW file.

The M10 doesn't have a user selectable base ISO. Digital ISO works by the camera underexposing the shot then amplifying it. As the M10 has (AFAIK) an invariant sensor, this means that the M10 will always have a bias towards shad recovery in post.

[Adam Bonn]

8 minutes ago, GFONG said:

The M10’s highlight blown out is because of the base ISO is around 160 (read from Internet). When using ISO 100, it is actually using the ISO 160. As a result,  in some circumstances the highlight is blown out because of 1/2 stop overexposed. That’s why Lecia changed the min ISO in AutoISO to 200

To expand on this.

When the M10 is set to ISO 100 it is indeed really ISO 160*. The problem is that the camera then attempts to make 160 into 100 by overexposing the RAW file then artificially darkening it.

The problem is that the original overexposure clips highlights. Much like if the user clips highlights at capture then no amount of moving the exposure slider to the left will fix it, the same is true in camera.

Conversely ISO 200 does the opposite - the camera underexposes the shot then artificially lightens it. 

The underexposure at 200 more protects the highlights than overexposure at 100. (over exposure and highlight protection don't really go hand in hand)

*In fact every ISO setting on the M10 is really 160! All digital cameras have only one ISO. The other settings are all push (base ISO + x stop under exposure, then x stop artificial brightening) or pull (base ISO + x stop over exposure, then x stop artificial darkening)

It's generally considered to be sound advice to stay away from pull ISOs unless one is very careful metering the scene or wishes to add contrast to a flat scene

(by 'artificially brightening/lightening' I mean tone mapping at point of DNG creation in camera)

[SOHODE]

4 hours ago, GFONG said:

For the noise, I found that the M10P is one to one and a half stop better than the M10R. For example M10P’s images at ISO 8000 is cleaner than M10R at ISO 6400. Also by looking at M10R’s images at ISO 400 and above, you can see the noise.  I don’t know why people in Internet keep saying the M10R high ISO is better than M10.

 

 

That's exactly what I'm looking for. Is there any picture samples/comparison on the net or you have done guys?

[SOHODE]

10 hours ago, jaapv said:

That would only mean that they adjusted the exposure curve to pull down the highlights and brighten the midtones and lift the shadows adjusting the black point at the same time in-camera before writing the DNG   That would certainly produce this effect if they adjusted the ISO value and exposure metering to match. Consider that a pixel produces an analog signal from base noise to 100% fill. The curve produced in between those two points can be modified. 

 

 

So it approves that why the DR charts of both cameras are similar. So do you think the only major difference between M10 and M10R (if we don't call it improvement) is the resolution?

"lift the shadows adjusting the black point" Doesn't it make more noise in M10-R?

 

[SOHODE]

4 hours ago, Adam Bonn said:

DR is not limited to highlights. A camera that offers 1 stop of HL recovery and 1 stop of Shad recovery has less exposure latitude than one that offers 0.5 stops of HL recovery and 2 stops of Shad recovery......As the M10 has (AFAIK) an invariant sensor, this means that the M10 will always have a bias towards shad recovery in post.

Have you compared the shadow recoveries of both cameras? or noises? I don't see any comparison on the net. That would be great if someone could do that. All is just comparing the highlight recoveries.

[SOHODE]

4 hours ago, Adam Bonn said:

 

*In fact every ISO setting on the M10 is really 160! All digital cameras have only one ISO. The other settings are all push (base ISO + x stop under exposure, then x stop artificial brightening) or pull (base ISO + x stop over exposure, then x stop artificial darkening)

 

I think it's actually a weakness of M10 that doesn't let use the native ISO (160)

What is the native ISO of M10-R? Isn't it 100?

[Advertisement]

Hallo,

Look here: Leica M

[jaapv]

8 minutes ago, SOHODE said:

So it approves that why the DR charts of both cameras are similar. So do you think the only major difference between M10 and M10R (if we don't call it improvement) is the resolution?

"lift the shadows adjusting the black point" Doesn't it make more noise in M10-R?

 

lifting the Black point reduces noise by blocking, at the expense of detail, lowering the black point will increase  noise but may give some more detail.  All, of course,  at the very bottom of the curve. 

[jaapv]

4 hours ago, Adam Bonn said:

All digital cameras have only one ISO.

Err... not quite. Nowadays we have dual gain, AKA dual native ISO.

https://www.premiumbeat.com/blog/dual-native-iso-explained/

It does not change the basic sensitivity of the sensor, but it certainly improves noise behaviour.

On a sidenote: the use of ISO values for sensors is terribly imprecise. It is basically a film value, denoting the density of the exposed film - objectively measurable. However, for sensors it is defined by equivalence to film which leaves the readout values and conversion of the signal wide open, in practice giving the manufacturers the. freedom to assign it any way they want.