ZF Vs Z6III high iso

Looking at the reports from the Z6III high iso IQ, it's got me wondering if the ZF is now the best Nikon option for low light photography. Expeed 7 for better AF + non 'stacked' 24mp sensor. At least that's what's keeping me from dropping the cash on the Z6III

Kittemoth said:

Looking at the reports from the Z6III high iso IQ, it's got me wondering if the ZF is now the best Nikon option for low light photography. Expeed 7 for better AF + non 'stacked' 24mp sensor. At least that's what's keeping me from dropping the cash on the Z6III

Click to expand...

I don't know which reports you are looking at but there is no difference worth mentioning between the Z6III and the Zf (or Z6II which has the same sensor) at high ISO in terms of image quality. There is a difference in dynamic range at low ISO but the real world differences are minimal if noticeable at all in practice.


It might be better to make purchase decisions on the basis of other features of the Z6III. AF in low light is incredibly good.

I've seen two reviews now that show the Z6iii ISO noise is very slightly poorer than that of the Z6ii.

I don't know why that should be though.

Dosstramp said:

I've seen two reviews now that show the Z6iii ISO noise is very slightly poorer than that of the Z6ii.

I don't know why that should be though.

Click to expand...

z6iii has a semi stacked sensor (whatever that actually means) vs not stacked on the z6ii. That's why.

Kittemoth said:

Looking at the reports from the Z6III high iso IQ, it's got me wondering if the ZF is now the best Nikon option for low light photography. Expeed 7 for better AF + non 'stacked' 24mp sensor. At least that's what's keeping me from dropping the cash on the Z6III

Click to expand...

Ugh. I guess I'm going to need to spend a lot of time addressing this in my eventual review. People are getting hung up on one set of numbers and not understanding any of the nuance.

From ISO 800 to 6400, the image quality of a Zf and Z6 III are, within sample variance, equal. Beyond 6400 I'm measuring the integrity of the Z6 III data as being slightly better than the Zf, though visually most people would say they're the same.

There are other factors that come into play. On a Zf if you want any large buffer at all, you're likely in High efficiency raw, which produces noisier shadows. On a Z6 III, the buffer is effectively infinite (if you override one default) using Lossless compressed, which provides better deep shadow integrity.

But there's another factor here that people keep forgetting: the Z6 III focuses better in low light than the Zf. If you don't get the image in the first place then it doesn't matter how noisy it would be (and it won't be noisier, as I just pointed out).

I could go on, but then I'd be writing my review here, not on my site.

ghostfox_1 said:

Dosstramp said:

I've seen two reviews now that show the Z6iii ISO noise is very slightly poorer than that of the Z6ii.

Click to expand...

Click to expand...

There's a difference between "maximal dynamic range" and what a camera does when handled properly.

The Z5, Zf, and even Z6 II have a better maximal dynamic range at base ISO. Given that these cameras come with Auto ISO turned on, just how often is that casual user at base ISO? ;~)

That said, in normal operation so far, I'm not missing maximal dynamic range. Instead, I'm enjoying other attributes of the camera that set it fairly far apart from those other models.

ghostfox_1 said:

ghostfox_1 said:

I don't know why that should be though.

Click to expand...

z6iii has a semi stacked sensor (whatever that actually means) vs not stacked on the z6ii. That's why.

Click to expand...

It's not clear that the partial stacking is responsible for difference in read noise that causes the low ISO dynamic range loss. It could be that the ADCs have to run faster to supply the stack portion in order make the camera run faster. As far as I can tell, the stacked portion is simply a fast parallel move of multiple data points simultaneously. But because it's moving data so much faster to EXPEED7, it also needs to be supplied with data faster, which is certainly coming from changes in the basic image processor portion.

I really wish Nikon would be more forthcoming about their key tech like this. Fully understanding it can make it easier for those of us who want to maximize our use of it. Instead, we have to experiment and reverse engineer, which takes time and doesn't necessarily get us to "optimal."

Thom Hogan said:

I really wish Nikon would be more forthcoming about their key tech like this. Fully understanding it can make it easier for those of us who want to maximize our use of it. Instead, we have to experiment and reverse engineer, which takes time and doesn't necessarily get us to "optimal."

Click to expand...

Out of curiosity, have you asked them and they won't tell you, or you're saying you think they should be releasing this information on their own along with other product info/specs they offer to the public?

NeedMorePractice said:

Thom Hogan said:

I really wish Nikon would be more forthcoming about their key tech like this. Fully understanding it can make it easier for those of us who want to maximize our use of it. Instead, we have to experiment and reverse engineer, which takes time and doesn't necessarily get us to "optimal."

Click to expand...

Out of curiosity, have you asked them and they won't tell you, or you're saying you think they should be releasing this information on their own along with other product info/specs they offer to the public?

Click to expand...

Both. Well, okay, NikonUSA doesn't say they won't tell me, they say they don't have that information or something like that, which I believe. The things I want to know are buried deep in the engineering, so really need Tokyo to disclose.

Some ISO-6400 samples of a Z6III, just to give y'all an idea:











Took these last night at a rock festival in my home town. Nikon Z6III with a Tamron 35-150mm f/2.8-4 Di VC OSD (on FTZ). There's actually a bit of noise, but only visible when pixel peeping. Perfectly usable for me.

The band is 'The Fortunate Sons', a Dutch CCR tribute band. I had a real blast - with the band, as well as with my Z6III ;-)

--
Tempestas Furit Cum Omni Vi
Tempestas In Capite Meo

DVF Photography said:

Some ISO-6400 samples of a Z6III, just to give y'all an idea:



Took these last night at a rock festival in my home town. Nikon Z6III with a Tamron 35-150mm f/2.8-4 Di VC OSD (on FTZ). There's actually a bit of noise, but only visible when pixel peeping. Perfectly usable for me.

The band is 'The Fortunate Sons', a Dutch CCR tribute band. I had a real blast - with the band, as well as with my Z6III ;-)

Click to expand...

I can see the noise clearly when viewing at full size but I would see the same level of noise if you had shot this scene with a Z6II or Zf so they don't really address the issue that some people have with misunderstanding or not examining the available data. In any case, you could easily clean this up completely with Adobe Denoise if you shot raw.

Kittemoth said:

Looking at the reports from the Z6III high iso IQ, it's got me wondering if the ZF is now the best Nikon option for low light photography. Expeed 7 for better AF + non 'stacked' 24mp sensor. At least that's what's keeping me from dropping the cash on the Z6III

Click to expand...

I wouldn’t worry about it. The Zf and z6iii are leaps ahead of the z5, z6, z6ii in so many other ways it doesn’t matter. Pick which body you prefer.

If you are an event photographer though, I think the decision is more between the z8 and z6iii. I’d rather get a z8 used for 3k or refurbished at 3200 than a new z6iii right now. Easy decision if I was up in that price range for me. Nikon made z8 too good of a deal and held the 3200 refurb and 3500 sale price too long. It’s kind of like the d700 value prop in a way of u remember that.

The Z6 III's lower DR manifests as noise in pushed shadows at low ISO but also noise without pushing for shadows at high DR, as the contribution of the Z6 III's higher read noise represents the same percentage of total noise between the two at those tonal levels (high iso lower midtones/shadows = low ISO deep shadows). You can see this in my animations from this post, which compares the to the original Z6, which has identical noise performance to the Zf:

Animation: Z6 vs Z6 III, ISO 12,800

Animation: Z6 vs Z6 III, ISO 25,600

Animation: Z6 vs Z6 III, ISO 51,200

This is also visible in the input-referred read noise figures from Bill's measurements:


Zf vs Z6 III, Input-referred read noise

Thom Hogan said:

ghostfox_1 said:

Dosstramp said:

I've seen two reviews now that show the Z6iii ISO noise is very slightly poorer than that of the Z6ii.

Click to expand...

Click to expand...

There's a difference between "maximal dynamic range" and what a camera does when I really wish Nikon would be more forthcoming about their key tech like this. Fully understanding it can make it easier for those of us who want to maximize our use of it. Instead, we have to experiment and reverse engineer, which takes time and doesn't necessarily get us to "optimal."

Click to expand...

I don’t see why anyone should believe that the public would do better with more information. We see how poorly people handle the dynamic range difference.

Horshack said:

The Z6 III's lower DR manifests as noise in pushed shadows at low ISO but also noise without pushing for shadows at high DR, as the contribution of the Z6 III's higher read noise represents the same percentage of total noise between the two at those tonal levels (high iso lower midtones/shadows = low ISO deep shadows). You can see this in my animations from this post, which compares the to the original Z6, which has identical noise performance to the Zf:

Animation: Z6 vs Z6 III, ISO 12,800

Animation: Z6 vs Z6 III, ISO 25,600

Animation: Z6 vs Z6 III, ISO 51,200

This is also visible in the input-referred read noise figures from Bill's measurements:


Zf vs Z6 III, Input-referred read noise

Click to expand...

What I see is a difference of no consequence in any real world photo unless the shadows are pulled up a lot, save for ISO 51,200, where it's more obvious. Engineering is always a compromise. If you want all the other speed-related features and benefits, plus an affordable price, something has to give. Nikon surprised us, well me anyway, with a semi-stacked sensor to achieve certain goals. They have a range of bodies so people can hopefully get the features that are important to them. I vote Z6iii.

--
Ruby
(If you can't see my posts it's because I often say things that get them deleted!)

Ruby Rod said:

Horshack said:

The Z6 III's lower DR manifests as noise in pushed shadows at low ISO but also noise without pushing for shadows at high DR, as the contribution of the Z6 III's higher read noise represents the same percentage of total noise between the two at those tonal levels (high iso lower midtones/shadows = low ISO deep shadows). You can see this in my animations from this post, which compares the to the original Z6, which has identical noise performance to the Zf:

Animation: Z6 vs Z6 III, ISO 12,800

Animation: Z6 vs Z6 III, ISO 25,600

Animation: Z6 vs Z6 III, ISO 51,200

This is also visible in the input-referred read noise figures from Bill's measurements:


Zf vs Z6 III, Input-referred read noise

Click to expand...

What I see is a difference of no consequence in any real world photo unless the shadows are pulled up a lot, save for ISO 51,200, where it's more obvious. Engineering is always a compromise. If you want all the other speed-related features and benefits, plus an affordable price, something has to give. Nikon surprised us, well me anyway, with a semi-stacked sensor to achieve certain goals. They have a range of bodies so people can hopefully get the features that are important to them. I vote Z6iii.

Click to expand...

The difference is noticeable starting at ISO 6,400 and without any shadow pushing. Whether or not the difference is consequential is of course an exercise for the viewer to decide.

Thom Hogan said:

Kittemoth said:

Looking at the reports from the Z6III high iso IQ, it's got me wondering if the ZF is now the best Nikon option for low light photography. Expeed 7 for better AF + non 'stacked' 24mp sensor. At least that's what's keeping me from dropping the cash on the Z6III

Click to expand...

Ugh. I guess I'm going to need to spend a lot of time addressing this in my eventual review. People are getting hung up on one set of numbers and not understanding any of the nuance.

From ISO 800 to 6400, the image quality of a Zf and Z6 III are, within sample variance, equal. Beyond 6400 I'm measuring the integrity of the Z6 III data as being slightly better than the Zf, though visually most people would say they're the same.

There are other factors that come into play. On a Zf if you want any large buffer at all, you're likely in High efficiency raw, which produces noisier shadows. On a Z6 III, the buffer is effectively infinite (if you override one default) using Lossless compressed, which provides better deep shadow integrity.

But there's another factor here that people keep forgetting: the Z6 III focuses better in low light than the Zf. If you don't get the image in the first place then it doesn't matter how noisy it would be (and it won't be noisier, as I just pointed out).

I could go on, but then I'd be writing my review here, not on my site.

Click to expand...

I don't think this thread is ready yet for actual facts, reason and good sense

On that chart over ISO 800 the Z6iii is a half stop worse, max. It's your contention that half a stop is readily apparant? Without diving in?

I thought in some tests the Z6 II started to show greenish color shift/cast whereas the Z6 III did not when pushing higher ISO, this should be a nice benefit as well of the III.

sandy b said:

On that chart over ISO 800 the Z6iii is a half stop worse, max. It's your contention that half a stop is readily apparant? Without diving in?

Click to expand...

The answer you seek is just two clicks away.

Thanks. ISO 25600 is acceptable. I would probably never go that high.