[normhead]

So Ian, why is the Z7, a four year newer camera a point behind the K-1 on DxO? And that's without using pixel shift on the K-1.

[luftfluss]

Nikon Z7 DXOMARK Sensor Scores

Overall Score [?]
99

Portrait (Color Depth) [?]
26.3 bits

Landscape (Dynamic Range) [?]
14.6 Evs

Sports (Low-Light ISO) [?]
2668 ISO

https://www.dxomark.com/Cameras/Nikon/Z7




Pentax K-1 DXOMARK Sensor Scores

Overall Score [?]
96

Portrait (Color Depth) [?]
25.4 bits

Landscape (Dynamic Range) [?]
14.6 Evs

Sports (Low-Light ISO) [?]
3280 ISO

https://www.dxomark.com/Cameras/Pentax/K-1

[normhead]

Nikon Z7 DXOMARK Sensor Scores

Overall Score [?]
99

Portrait (Color Depth) [?]
26.3 bits

Landscape (Dynamic Range) [?]
14.6 Evs

Sports (Low-Light ISO) [?]
2668 ISO

Nikon Z7 - DxOMark




Pentax K-1 DXOMARK Sensor Scores

Overall Score [?]
96

Portrait (Color Depth) [?]
25.4 bits

Landscape (Dynamic Range) [?]
14.6 Evs

Sports (Low-Light ISO) [?]
3280 ISO

Pentax K-1 - DxOMark

Thanks, somehow I got that wrong. But none the less, past experience has taught us, a difference to 3 isn't visible, a difference of 5 is, so you still have to ask... 3-4 years later, what's the issue?

Especially since both have the same score in Dynamic Range, which DxO calls landscape. (Measured Dynamic range, not theoretical.) Colour depth is also not significant, leaving ISO, where for some reason the Z7 loses, but also by an insignificant amount.

It doesn't look like a 3-4 yer newer camera to me. Anyone using their eyes not test scores is going to have no reason to buy a Z7, based on image quality. It's definitely no worth selling a K-1 for.

Looking at IR images.... still not much to pick and chose from. With equal processing they'd look out the same.

Meanwhile Nikon Z7 at Henries $3400, K-1ii $2400.

Is that the cost of going mirrorless these days? $1000 for practically identical images but brownie points in some places for being on the mirrorless band wagon? Those brownie points are darn expensive.

How insane do you have to be to go for the Z7?

[Ian Stuart Forsyth]

Well, I don't think Nikon makes the sensor, I think Sony probably does and we know their focus has been on getting faster read out speeds and better video. 6K video, here we come!!

So, still photography has been pretty steady unless you count the accelerator as a bit of an improvement.

One thing Nikon has done is tweaked how they are able to use the sensors from Sony






While Nikon has not really done anything to increase the DR they have lowered the noise for a good part of that DR If we look at 10% grey they have increased the SNR from 36 dB to 37.4 dB this is where in the shadows i do most of my pushing from and not the deepest.
You can have a large DR but the noise throughout that range can limit the how pleasing the image will look

While the D7200 has the same DR, how clean even at 10% grey the D7200 is nearly 2 stops behind the K1 as for how noise appears

---------- Post added 12-29-2020 at 09:11 PM ----------

So Ian, why is the Z7, a four year newer camera a point behind the K-1 on DxO? And that's without using pixel shift on the K-1.

Behind how ?

[luftfluss]

Thanks, somehow I got that wrong. But none the less, past experience has taught us, a difference to 3 isn't visible, a difference of 5 is, so you still have to ask... 3-4 years later, what's the issue?

I wonder about the effect of on-sensor capabilities wrt to AF and video performance. How negatively does that impact the image quality of a sensor?

Meanwhile Nikon Z7 at Henries $3400, K-1ii $2400.

Is that the cost of going mirrorless these days? $1000 for practically identical images but brownie points in some places for being on the mirrorless band wagon? Those brownie points are darn expensive.

How insane do you have to be to go for the Z7?

People like what they like... or what they think they like. Each camera has specific features unavailable on the other. Alternatively, you can purchase the FF 24MP Nikon Z5 for $1197 USD.

[LeeRunge]

I think the main reason to go with a Z6/7 is video. Mirrorless makes way more sense for they hybrid photographer/videographers out there for weddings, instagram, youtube etc.

[biz-engineer]

So Ian, why is the Z7, a four year newer camera a point behind the K-1 on DxO? And that's without using pixel shift on the K-1.

On the Z7, the native ISO is around 64, due to less light going through each lens that cover each RGB pixel square (which increase pixel to pixel contrast, and so make lens charts resolve more).
And Nikon also did the trick of the Pentax K-5 ISO80 setting, i.e overexpose +0.3ev (ISO32 on the Z7/D850). So if you dial +0.3ev exposure compensation on you K1 at ISO100, you get the same as the Nikon at ISO32.

One thing Nikon has done is tweaked how they are able to use the sensors from Sony

In addition to having smaller pixel aperture, the Nikon D850 and Z7 also have fake ISO settings (underexpose for high light protection, or overexpose for better SNR), like Pentax did on the K5 with ISO80 (overexpose). Pentax K1 users can do the same fake ISO by playing with exposure compensation depending on the scene.

---------- Post added 30-12-20 at 07:52 ----------

DXO mark ("print") measurement figures:

Dynamic range:

K1 at ISO100 = 14.6ev;

Z7 at ISO64 = 14.61ev (Z7 is 0.01ev better, practically a measurement error)


SNR 18%:

K1 at ISO100 = 45dB;

Z7 at ISO64 = 45.3dB (Z7 gets 0.3dB better, but real ISO of Z7 is 44 instead of 64 => Z7 overexpose to get that figure)
Z7 at ISO32 = 47.1dB (real ISO32 = 44 for Z7 they overexpose, real ISO100 = 95 for the K1 no overexposure)

[surfar]

I think the main reason to go with a Z6/7 is video. Mirrorless makes way more sense for they hybrid photographer/videographers out there for weddings, instagram, youtube etc.

Definitely,as well as a growing range of modern lenses, all WR.Access via adapter to Nikon and 3rd party Dslr glass.With another adapter access to EF/s 120+miliion lenses.

[normhead]

One thing Nikon has done is tweaked how they are able to use the sensors from Sony






While Nikon has not really done anything to increase the DR they have lowered the noise for a good part of that DR If we look at 10% grey they have increased the SNR from 36 dB to 37.4 dB this is where in the shadows i do most of my pushing from and not the deepest.
You can have a large DR but the noise throughout that range can limit the how pleasing the image will look

While the D7200 has the same DR, how clean even at 10% grey the D7200 is nearly 2 stops behind the K1 as for how noise appears

---------- Post added 12-29-2020 at 09:11 PM ----------



Behind how ?

So what you're saying is I should happily pay $1000 more to get the same curves?

You can have a large DR but the noise throughout that range can limit the how pleasing the image will look

And yes the K-1 is rated better than the Z7 for high ISO performance. You're trotting out some pretty subjective evidence here offering nothing to support you claim that except a technical number where in some areas the Z7 is 5% better in noise. 5% is hardly ever visible comparing digital images. This is marketing hype.

[beholder3]

> Disadvantage to Nikon's approach: lot of artifacts and false colors (much more than Pentax K1), evidence can be found in a side by side comparison between K1 and Z7 with DPR studio comparison tool: Pentax K1 show less false colors despite having 36Mp instead of 45Mp.

That is again an area to prove how foolish it is to stare at diagrams.

Any noob staring at a diagram might think, a Z7 was only a little bit behind on on max dynamic range versus older K-1 (even if trying ISO 64 versus ISO 100, much more lagging at same ISO), so you could pull similar details out of the shadows (the area, where Canon a long time was lagging).

Now if you use your own eyes instead of charts, you'll be reminded about the banding issues Nikon Z has:

Image comparison: Digital Photography Review

You loose 2 (!) full stops of dynamic range to the Z's banding problem and it is quite visibly noisier as well, much more than the charts express the 2016's K-1 advantage.

[normhead]

That is again an area to prove how foolish it is to stare at diagrams.

Any noob staring at a diagram might think, a Z7 was only a little bit behind on on max dynamic range versus older K-1 (even if trying ISO 64 versus ISO 100, much more lagging at same ISO), so you could pull similar details out of the shadows (the area, where Canon a long time was lagging).

Now if you use your own eyes instead of charts, you'll be reminded about the banding issues Nikon Z has:

Image comparison: Digital Photography Review

You loose 2 (!) full stops of dynamic range to the Z's banding problem and it is quite visibly noisier as well, much more than the charts express the 2016's K-1 advantage.

After years of doing this, you realize most of the chart stats mean little to nothing on digital images, it's really only of use for internet bull sessions. You have to look at the images.

Truth be known, at 4k, my K-5 will have just as good images as the Z7 for a fraction of the price, this based not on charts, but on the direct observation that it's very difficult to tell APS-c from FF images 95% of the time and my K-5 images are pretty much identical to my K-5 images on my 4k display. People always say "well if you print big." Yet I can't tell which of my big prints is taken with which camera.

You do these things for a while, you clue into these things. Not if all you do is look at test charts.

With that in mind, we've ordered a second hand k-5ii, and there are places where I will find using it preferable to both the K-1 or a theoretical Z7.

I find I had to believe given the price difference that intelligent people would y a z7 and nothin instead of a K-1ii and D FA* 50 1.4. A truly top of class lens.

There are a few photographers for whom some Z7 features will be awesome. I'm glad they have what they want.

[Smolk]

The difference you observe between Nikon lens tests and Pentax lens tests are fully explained below:
- Nikon designed their sensor micro-lenses smaller on their Z7 45Mp sensor in order to increase micro-contrast (~15% boost of contrast between pixels), see here (The Nikon Z7’s Insane Sharpness | Strolls with my Dog)

- smaller microlenses on the Z7 sensor let less light flow in per pixel, so is the base ISO on the Z7 lower than 100. That why Z7 need to be at ISO64 to reach the same DR as the K1 at ISO100.

- high ISO noise on the Z7 is compensated by better quantum efficiency of the BSI tech + dual gain enabled at high ISO settings, which compensate for the smaller micro-lenses

> Advantage to Nikon: sharper images (and lens test charts looking superior...), without sacrificing high ISO noise

> Disadvantage to Nikon's approach: lot of artifacts and false colors (much more than Pentax K1), evidence can be found in a side by side comparison between K1 and Z7 with DPR studio comparison tool: Pentax K1 show less false colors despite having 36Mp instead of 45Mp.

Thank you, that's helpful. It also shows how hard it is to actually compare the cameras. And that extends to the images they produce, since they are not neutral in most cases either.
But the discussion does give me a better idea of the real differences between the two systems.
And digitally, I haven't invested in either yet.

[Ian Stuart Forsyth]

In addition to having smaller pixel aperture, the Nikon D850 and Z7 also have fake ISO settings (underexpose for high light protection, or overexpose for better SNR), like Pentax did on the K5 with ISO80 (overexpose). Pentax K1 users can do the same fake ISO by playing with exposure compensation depending on the scene.

what is it? do they over or underexpose? You just told me they are doing both

Is Pentax underexposing their raw file be cause they needing a BLE -0.56?

There is no fake iso as there is no iso for raw and how they decide to use the sensor has nothing to do with iso or the standard.
What would happen if I told you that 36mp sensor is able to capture a larger exposure that pentax is allowing. Is Pentax using the sensor at base iso with no gain or offset?

Go rent a camera that allows the user to set the gain and the offset and you will see that there is no iso setting with that interfaces with the sensor
Like this camera, same sensor as the one found in the K1 and D810


This is the interface with the 36mp sensor and there is no iso setting.
Using this camera with no gain and no offset ( as you can program the sensor to be used in this manner and with pentax you cannot) this camera can capture nearly 3 stops larger exposure than the D800 at iso 100 this is using the exactly same lens on the very same scene as I had done while trying the camera along side the D800.

The reason why this interface is there is so that the user can better decided how to program the sensor to work and pick the best gain and offset for the exposure they are working with. More gain less noise along with less DR, Less gain the larger the FWC but with more noise.



This tells me that Pentax at iso 100 is still using gain and or offset, this is further supported by the fact the K1 still needs a BLE to offset the camera data in the raw file so that converters know how to process the raw file.

So if you dial +0.3ev exposure compensation on you K1 at ISO100, you get the same as the Nikon at ISO32.

But you can not

As you can see the k1 is up against the right side of the raw histogram and if you are to push the K1 down to a shutter speed of 1/5 like in the D580 then you would clip

As you can see with the D850 you still have the same highlight headroom but they have used a shutter speed of 1/5


And if you look closely to the BLE for both cameras this tell us that they are nearly using the same highlight heard room

---------- Post added 01-01-2021 at 04:10 AM ----------

Z7 at ISO32 = 47.1dB (real ISO32 = 44 for Z7 they overexpose, real ISO100 = 95 for the K1 no overexposure)

Again overexposed as to what ? If they have the same amount of head room and they both are not clipping anymore than the other what is overexposed?

---------- Post added 01-01-2021 at 04:13 AM ----------

So if you dial +0.3ev exposure compensation on you K1 at ISO100, you get the same as the Nikon at ISO32.

that a little more than +0.3 going from iso 64 down to 32

[biz-engineer]

There is no fake iso as there is no iso for raw and how they decide to use the sensor has nothing to do with iso or the standard.

There is no fake ISO, under you need to let the shutter open for a longer time to collect the same amount of light on the sensor!
People don't check the shutter speed when they compare lens test charts. Except, in the real world, shutter speed does matter...

if you need 1/100th for the photographic matter, and the camera meter says 1/50th, you need to increase the ISO setting to avoid subject motion blur!
Therefore, fake ISO beat the charts, but it is a drawback in real world photography. I personally prefer manufacturers who don't cheat on ISO settings.

---------- Post added 01-01-21 at 13:16 ----------

Originally posted by biz-engineer QuoteSo if you dial +0.3ev exposure compensation on you K1 at ISO100, you get the same as the Nikon at ISO32.But you can not

I can. I can ETTR K1, like I can do it on any brand of camera that offer a manual exposure mode, so I fully use the 14bits range of the sensor.

[normhead]

Looking at the posted images....which display practically no dynamic range, that chart was concocted bak when they were still testing 6 MP cameras with 11 or 12 stops of dynamic range... I'll buy the cheaper one. Practically the same result for less money, despite the circles and arrows and commentary.

The circles, arrows and commentary don't change the fact, the images are practically identical.

There is one use for this type of commentary. Useless internet "my camera is better than your camera" bull sessions.

In my experience, 99% of the time, neither of these cameras is better than my K-5 on a 4k TV. If time has taught me anything, arguing camera to camera for practically identical images is pointless. At some point, you go with the cheapest, and leave the "my camera is better" guys to wallow in their self delusion. Of course some guys go for the most expensive and then stick their nose in your face. I hate those guys.

One of my favourite tactics after someone does the out in the park. "Here's what I got, let me see what you have?' It's amazing how many are too embarrassed to even show me. They want to brag about having the "best gear" but they can't brag they are getting better pictures.

As for 4k video, I have. ZS100 that does 4k video and I can buy both the a K-1 and a ZS100 for less than a Z7. And take video of myself taking still images.... how cool is that? Try that with your Z7. The ZS fits in my pocket or a small corner of my camera bag. There's absolutely no problem taking it where ever I take my camera. For me, a much superior solution. Portable, functional, and I don't lose the use of my still camera when shooting video.

These guys who want their best still camera to do video, I don't know who they're talking to, but they aren't talking to me.

After coming to the thread to find out what kind of person might buy a Z7, I'm disappointed. I always like to think there might be something better (for my work flow) out there for cheaper. What a disappointment. Just another expensive wedding photographer camera competing in a niche market with few customers trying to pass itself off as camera for everyone, so folks like me will pay for capacity they'll likely never use. I prefer to not be one of the ones subsidizing the R&D costs of the camera, so wedding guys can have what they need.