Z6III dynamic range discussion

[wbenjin]

For people who cares about camera sensors' technical performance, it's well known that photographic dynamic range (PDR) for Z6III from Photons to Photos:



It shows that Z6III's PDR not only lower than Z6II and ancient D600(by a full stop) , but even catch up by non-BSI APS-C camera Z50! This seems a bit odd to me, so I tried to find another sensor measurement on internet, unfortunately there is very few data out there, I'll start from DigitalCameraWorld (DCW)'s measurement:

Nikon Z6

Nikon Z50

It shows very little DR difference between Z6III and Z6II, and definitely higher than Z50.

Same with signal-to-noise ratio (SNR):

Again, very little difference.

Next is measurements from RTINGS:


Nikon Z6III


Nikon Z6II

It tells a similar story as Photons to Photos, with less difference between Z6III and Z6II, down to 0.4EV.

Next is review from Youtuber Gerald Undone, he is measuring N-RAW video at ISO800 so I expect the PDR performance should lower than RAW photo at ISO100. Also he is the only review that specify the measurement tools and full analysis data.



I don't know how to interpret imatest data but if we look at Slope-based DR at 1.0 it shows 11.4 EV.

For now, we got many dynamic range data, with different measuring method, software and standard, at this point I don't know what to rely on. A properly conducted measurement should list all equipment that used in, include test target, mathematical and electrical standard, analyzing software and configuration, etc. Only a Youtuber done that, thought his measurements are for video.

Also many reviewers underexposed the photo and pull back using software like LrC, show that Z6III indeed has more noise than Z6II, but in my opinion they are not comparing DR between camera, but more like comparing SNR, because more noise doesn't mean it lost the darkest details, in many fields the dynamic range can be greater than SNR. Also the RAW files of modern cameras are not really that "RAW" in many ways, some with noise reduction already in it (such as Canon and Sony), internal calibrations, and almost all have PDAF and dead pixel interpolation, all make sensor performance measurement more complicated, even I think only the manufacturer itself know the real performance data of their sensors.

After all, my point is simple: modern FF cameras already have good enough performance, and vast majority of photographers just don't care about those technical data. We have many more factors to consider when choosing a camera, such as autofocus, ergonomics, price and lens support. I admit I was a bit concern about Z6III's DR, but after research above I'll just buy it to replace X-S20 and have a good sleep.

 

[Rodenmg]

You’ve already spent way too much time worrying about it.

 

[sirhawkeye64]

For people who cares about camera sensors' technical performance, it's well known that photographic dynamic range (PDR) for Z6III from Photons to Photos:



It shows that Z6III's PDR not only lower than Z6II and ancient D600(by a full stop) , but even catch up by non-BSI APS-C camera Z50! This seems a bit odd to me, so I tried to find another sensor measurement on internet, unfortunately there is very few data out there, I'll start from DigitalCameraWorld (DCW)'s measurement:

Nikon Z6

Nikon Z50

It shows very little DR difference between Z6III and Z6II, and definitely higher than Z50.

Same with signal-to-noise ratio (SNR):

Again, very little difference.

Next is measurements from RTINGS:


Nikon Z6III


Nikon Z6II

It tells a similar story as Photons to Photos, with less difference between Z6III and Z6II, down to 0.4EV.

Next is review from Youtuber Gerald Undone, he is measuring N-RAW video at ISO800 so I expect the PDR performance should lower than RAW photo at ISO100. Also he is the only review that specify the measurement tools and full analysis data.



I don't know how to interpret imatest data but if we look at Slope-based DR at 1.0 it shows 11.4 EV.

For now, we got many dynamic range data, with different measuring method, software and standard, at this point I don't know what to rely on. A properly conducted measurement should list all equipment that used in, include test target, mathematical and electrical standard, analyzing software and configuration, etc. Only a Youtuber done that, thought his measurements are for video.

Also many reviewers underexposed the photo and pull back using software like LrC, show that Z6III indeed has more noise than Z6II, but in my opinion they are not comparing DR between camera, but more like comparing SNR, because more noise doesn't mean it lost the darkest details, in many fields the dynamic range can be greater than SNR. Also the RAW files of modern cameras are not really that "RAW" in many ways, some with noise reduction already in it (such as Canon and Sony), internal calibrations, and almost all have PDAF and dead pixel interpolation, all make sensor performance measurement more complicated, even I think only the manufacturer itself know the real performance data of their sensors.

After all, my point is simple: modern FF cameras already have good enough performance, and vast majority of photographers just don't care about those technical data. We have many more factors to consider when choosing a camera, such as autofocus, ergonomics, price and lens support. I admit I was a bit concern about Z6III's DR, but after research above I'll just buy it to replace X-S20 and have a good sleep.

This topic has been discussed quite extensively (and to death) over the past 5 months. Yes the Z6 III's DR is not as good as say the Zf or Z6 II,but better than the Z8 and Z9. But in practicality you may not see it much in your photos as a result so I don't personally know why this keeps coming up. It's about the same as pixel peeping lens test charts at 300% you will see differences upon very close inspection, but it doesn't really translate to much in real world. If it was a 3 or more stop difference, then maybe, but 1-1.5 I don't think you'd see it.



--
PLEASE NOTE: I usually unsubscribe from forums and comments after a period of time, so if I do not respond, that is likely the reason. Feel free to PM me if you have a questions or need clarification about a comment I made.

 

[wbenjin]

You’ve already spent way too much time worrying about it.

You're right, just a engineer's little hobby trying to figure things out:-D

 

[Tim G]

Having just changed my ZF for a Z6iii I have seen no difference in real world use. I haven’t done any ‘testing’ and don’t intend to. The Z6iii is an outstanding camera. If the Z8 is a mini Z9, the Z6iii is a mini mini Z9 (loved the look of the ZF and got sucked in by the ‘retro’ aspects however couldn’t get to grips with the handling especially with my larger lenses). Cheers.

 

[zephxiiii]

I did a DR test between my D7100 and D80 which is much larger range (10.66 - 8.01 = 2.65 stop difference!) on a sunny day lake scene with trees in dark shadows...they looked near identical. I didn't post in here as may not be the most best comparison and wasn't done messing with it. So the difference in real world use between the Z6 II and III is... I mean you really have to go looking for it and won't be visible in normal use.

 

[labalaba]

You won't see much difference in well-exposed photos at low ISO. I remember back when I moved from D300 to D700. Couldn't see the difference. Then one day I had to shoot in challenging conditions, and was amazed at what I could get with the D700. To give you an example of when it matters: I frequently shoot insect macros inside tropical rain forests, where it is quite dark. Sometimes I use a flash of course, but I need to know when I can use natural light and when I can't. I will often be using f11-f16 or so for depth-of-field, and need to keep the shutter speed up when hand holding 200mm and 300mm lenses. I can easily end up in the ISO 1600-6400 range. When selecting cameras, it is definitely a factor for me how they will perform in those kind of ISO ranges, and a 1 stop difference in performance matters to me. These might not be considerations for other photographers.

 

[sirhawkeye64]

You won't see much difference in well-exposed photos at low ISO. I remember back when I moved from D300 to D700. Couldn't see the difference. Then one day I had to shoot in challenging conditions, and was amazed at what I could get with the D700. To give you an example of when it matters: I frequently shoot insect macros inside tropical rain forests, where it is quite dark. Sometimes I use a flash of course, but I need to know when I can use natural light and when I can't. I will often be using f11-f16 or so for depth-of-field, and need to keep the shutter speed up when hand holding 200mm and 300mm lenses. I can easily end up in the ISO 1600-6400 range. When selecting cameras, it is definitely a factor for me how they will perform in those kind of ISO ranges, and a 1 stop difference in performance matters to me. These might not be considerations for other photographers.

At higher ISOs sure there may be a difference but I think (And feel) at least when I looked at the charts, that it's all relative.

 

[Berend Botje]

I did a DR test between my D7100 and D80 which is much larger range (10.66 - 8.01 = 2.65 stop difference!) on a sunny day lake scene with trees in dark shadows...they looked near identical. I didn't post in here as may not be the most best comparison and wasn't done messing with it. So the difference in real world use between the Z6 II and III is... I mean you really have to go looking for it and won't be visible in normal use.

Just for context, the D80 used the older CCD type sensor, while Nikon moved to the CMOS type sensor beginning with the D90. This helps explain the large jump forward between the D80 and the D7100 in terms of dynamic range.

 

[Leonard Shepherd]

You’ve already spent way too much time worrying about it.

You're right, just a engineer's little hobby trying to figure things out:-D

Perhaps not exactly

What is perhaps very important to recognise this measure is but one measure - and that it does not use traditional DR convention.

The measure is reported elsewhere as starting with a signal to noise ratio of 20 rather than the more common 1, and with a resolution equal to the circle of confusion. This is much lower than commonly encountered in stills photography.

This could be why some other site tests conclude about 2 stops more DR.

One can debate as much as one wants which measure might be a most relevant to everyday photography - as long as one recognises other test methodology seems often to record about 15% more DR.

The DR conundrum is in some ways similar to lens testing with a 1000:1 very high contrast test target hardly ever encountered in real world photography :-|

Similarly if measuring lens performance at 50% contrast perhaps there is a need to take into account that Canon imply 60% contrast is normally required for a satisfactory image.

 

[vockman]

The Z6III sensor is a different animal than the D600-Z6II sensor. Not as good in some ways (dynamic range) better in others ( more image clarity and raw files that often need less tweaking)

The Z6 III performs at its best with good precise exposure and WB. Fortunately the new viewfinder makes it easy to expose precisely on the fly and the AWB is better than ever. At low ISO I find exposing well for the highlights is a bonus and at higher ISO exposing well for the shadows is more critical (if you want the shadow detail there: At higher ISO there isn't much highlight tonal range anyway).

Z3 files respond better to all the new LR AI filters and noise reduction.

BUT to add to the challenge; The Highlight and Shadow slider response in LR is very different and less well targeted than for the Z6 files.

overall, for shooting responsiveness and and image quality, I feel like the Z6III is a step up.

YET when I had my first multiple days, last week, of studio headshots since getting the Z6III, I decided to use my Z6 (ISO 100/ studio strobes) just because I wasn't in the mood yet to deal with any unexpected challenges from the Z6III files. I should have made some comparisons with the Z6III but never got around to it.

So I have decided to keep one of my Z6's for landscape work and studio work until I feel I have mastered processing the Z6III files

Curious about how the upcoming Adobe adaptive presets will work out.

 

[wbenjin]

You’ve already spent way too much time worrying about it.

You're right, just a engineer's little hobby trying to figure things out:-D

Perhaps not exactly

What is perhaps very important to recognise this measure is but one measure - and that it does not use traditional DR convention.

The measure is reported elsewhere as starting with a signal to noise ratio of 20 rather than the more common 1, and with a resolution equal to the circle of confusion. This is much lower than commonly encountered in stills photography.

This could be why some other site tests conclude about 2 stops more DR.

One can debate as much as one wants which measure might be a most relevant to everyday photography - as long as one recognises other test methodology seems often to record about 15% more DR.

The DR conundrum is in some ways similar to lens testing with a 1000:1 very high contrast test target hardly ever encountered in real world photography :-|

Similarly if measuring lens performance at 50% contrast perhaps there is a need to take into account that Canon imply 60% contrast is normally required for a satisfactory image.

Thanks for your reply, this is just what I want to express: So many different measurements with different procedures and standards, which is often not stated clearly. So people definitely shouldn't look at a particular measurement and panic around.

 

[raunak]

Yes the Z6 III's DR is not as good as say the Zf or Z6 II,but better than the Z8 and Z9. But in practicality you may not see it much in your photos as a result so I don't personally know why this keeps coming up.

Not throughout the ISO range though..

At lower ISOs or at their respective base ISOs, the Z8 has higher DR (11.31 @ISO64 for Z8 vs 10.44 @ISO100 for Z6 III)

At higher ISOs (800 onwards), the Z6 III is slightly better but the difference is smaller (0.2 EV approx). I agree that for most practical purposes it would be difficult to show the difference without pixel peeping.

 

[Kenneth Almquist]

For people who cares about camera sensors' technical performance, it's well known that photographic dynamic range (PDR) for Z6III from Photons to Photos:



It shows that Z6III's PDR not only lower than Z6II and ancient D600(by a full stop) , but even catch up by non-BSI APS-C camera Z50! This seems a bit odd to me...

After all, my point is simple: modern FF cameras already have good enough performance, and vast majority of photographers just don't care about those technical data. We have many more factors to consider when choosing a camera, such as autofocus, ergonomics, price and lens support. I admit I was a bit concern about Z6III's DR, but after research above I'll just buy it to replace X-S20 and have a good sleep.

The only site you list that I’m really familiar with is the first one. All the cameras you show have at least 8 stops of dynamic range at ISO 800 and above. That’’s plenty for most purposes.

One exception is if you want to shoot a scene with high dynamic range and push the shadows really hard, and can’t use exposure bracketing. I haven’t gotten great results when trying this; I think the issue is that pushing the shadows reduces the contrast, which tends to make the image less interesting, and also (since the contrast is lower than that of the actual scene) tends to make it look unnatural. If you plan to do this, then you have a reason to look for a camera with really high dynamic range at low ISO values, but I don’t think that applies to most photographers.

In low light situation where you have to shoot at high ISO values, dynamic range is likely to be more of a problem. For example, at ISO 6400, the cameras you show range from a dynamic range of 5 for the Z50 to a bit over 6 for the Z6 II and Z6 III. And when you are shooting in low light, you typically want the images to look dark, so you aren’t using the full dynamic range. The one stop difference between the Z50 and the Z6 III will definitely matter. On the other hand, the difference between the Z6 III and the Z6 II is very small--almost certainly too small to notice in real life even in side-by-side photographs of the same subject.

 

[zephxiiii]

I did a DR test between my D7100 and D80 which is much larger range (10.66 - 8.01 = 2.65 stop difference!) on a sunny day lake scene with trees in dark shadows...they looked near identical. I didn't post in here as may not be the most best comparison and wasn't done messing with it. So the difference in real world use between the Z6 II and III is... I mean you really have to go looking for it and won't be visible in normal use.

Just for context, the D80 used the older CCD type sensor, while Nikon moved to the CMOS type sensor beginning with the D90. This helps explain the large jump forward between the D80 and the D7100 in terms of dynamic range.

Yah.

The biggest thing I have noticed and want to note with D80 vs range of D7100 is that your exposure window being smaller means you really have to nail exposure to prevent clipping in HDR situations. With the D7100 and newer sensors there's a bit more forgiveness there.

On mirrorless and WYSWYG live exposure w/ histogram this is far less of an issue. I'd question if you'd even notice it with II vs III.

 

[hectorsm]

I have shot over 2,000 photos (Mostly landscape) with my new Z6iii and I have not noticed any difference between my new 6iii and my old D600 except for a slight difference in the colors. I think the Z6iii slightly lower DR numbers are meaningless. The difference is too small to be noticeable.

My suggestion is to just enjoy taking the photos and forget about these numbers. Here are a few pictures I took last week in Montreal and Quebec City, Canada.

 

[msu79gt82]

You’ve already spent way too much time worrying about it.

But thanks to the OP anyway

 

[msu79gt82]

Having just changed my ZF for a Z6iii I have seen no difference in real world use. I haven’t done any ‘testing’ and don’t intend to. The Z6iii is an outstanding camera. If the Z8 is a mini Z9, the Z6iii is a mini mini Z9 (loved the look of the ZF and got sucked in by the ‘retro’ aspects however couldn’t get to grips with the handling especially with my larger lenses). Cheers.

I agree. My Z6iii AF allows me to capture images I was never able to master with my Z6.

 

[Leonard Shepherd]

Thanks for your reply, this is just what I want to express: So many different measurements with different procedures and standards, which is often not stated clearly.

There is a well respected site that has yet to publish results for the Z6 III - though it reports nearly 14 stops for the Z6 II - using the older "traditional" DR testing methodology.

So people definitely shouldn't look at a particular measurement and panic around.

As it is difficult in bright outdoor sunlight to find a normal landscape scene with 9 stops DR; up to 11.5 stops using one method does particularly bother me, and perhaps 13.5 stops using other methodology does not bother me at all - at or near base ISO.

Few in the "debate" seem to have considered the limitation of 8 stops or less DR by around 5,000 ISO using numerous recent camera bodies. :-(

--
Leonard Shepherd
In lots of ways good photography is similar to learning to play a piano - it takes practice to develop skill in either activity.

 

[primeshooter]

I have shot over 2,000 photos (Mostly landscape) with my new Z6iii and I have not noticed any difference between my new 6iii and my old D600 except for a slight difference in the colors. I think the Z6iii slightly lower DR numbers are meaningless. The difference is too small to be noticeable.

My suggestion is to just enjoy taking the photos and forget about these numbers. Here are a few pictures I took last week in Montreal and Quebec City, Canada.

Not being funny; but you're never gonna notice differences between most cameras in shots like these, so it really is no surprize that you've noticed no difference from your old D600: just saying.