Nikon Z9 ISO performance

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

A series of images just post to the facebook Nikon D5 / D6 - that compares several back to back same scene photos with the D6 and Z9. Street night lighting, both using the 24-120 lenses. Interesting to see. He is a professional news photography and is at a demonstration.


what was the outcome.
 
I have both a D850 and Z9, so I am interested in the differences in sensors. I went to PhotonsToPhotos.net to look at their Dynamic Range and Read noise curves. Their data shows the D850 has a greater DR and better noise performance from ISO 64 to 400, after that they are pretty much identical.


That makes sence
 
The Bill Claff data show that the Photographic DR of the D850 at base USI (ISO 64) is slightly higher than that of the Z9.

To investigate noise differences I photographed a ColorChecker with both cameras using 1/20s @f/8 and evaluated the noise of the second square of the bottom row, noting the G1 and G2 σ (standard deviation) and μ (mean) G1 and G2 pixel values in this square using RawDigger. Since Rawdigger does not support Z9 NEF, I converted all of the files to DNG using the latest DNG converter (14.2.0.1028).

Here is a photo of the chart:

View attachment 33907

Here is the Rawdigger evaluation of the Z9 G2 raw channel with the second square of the lowest row selected. The stats are shown second from right. I used the green channels since green constitutes the majority of the human visual response.

View attachment 33908


I then evaluated the green channels for the D850, the green channels of the Z9 using both lossless compressed and HF*. The results are shown in this table. σ is standard deviation and μ is the mean. With lossless compressed, the noise with the Z9 is slightly better than with the D850. With lossless in the Z9 noise is slightly less than with the D850. I don't think that these differences are visually significant.

View attachment 33909

Bill

So would you feel the Z9 is not as good for landscape or its not noticeable using base ISO.
 
So would you feel the Z9 is not as good for landscape or its not noticeable using base ISO.
So would you feel the Z9 is not as good for landscape or its not noticeable using base ISO.

As far as landscape photography goes, I think they are equivalent. The small diffrence in photographic dynamic range at base iso is not perceptually different.

Bill
 
what was the outcome.
D6 was cleaner and better color - The person used a Black Rapid twin strap, same settings and lens on both cameras and exposed scene one after the other. I linked it in my original post but the link did not stick or was removed. On FB find the: Nikon D5 / D6 and it is in the Z9 wall, page, what ever they call it.
 
As far as landscape photography goes, I think they are equivalent. The small diffrence in photographic dynamic range at base iso is not perceptually different.

Bill
I've been shooting my D850 and Z9 (currently a rental, but in theory my Z9 should be here within a week or so) side-by-side for both landscape and wildlife. From a quality perspective I see no real difference. One advantage I've found from the Z9 for landscapes which I didn't expect was the benefit of 20FPS. I shoot 5 shot brackets most of the time and then in post I either pick the best of the bunch and work it up or I take several of them and merge into an HDR. With the 20FPS, the merge gets a little better/easier because there's less movement in the scene between shots.
 
D6 was cleaner and better color - The person used a Black Rapid twin strap, same settings and lens on both cameras and exposed scene one after the other. I linked it in my original post but the link did not stick or was removed. On FB find the: Nikon D5 / D6 and it is in the Z9 wall, page, what ever they call it.

My bet is you are seeing the difference in the amount of light gathered by a FF 20.9mp sensor pixel vs that of a 45.7mp FF sensor pixel.
 
My bet is you are seeing the difference in the amount of light gathered by a FF 20.9mp sensor pixel vs that of a 45.7mp FF sensor pixel.

If you downsize the 45.7 mp FF to 20.9 mp FF the results should be similar. This is known as normalizing and is the essence of the print and screen data on Dxomark. Bill Claff also does an normalization procedure for his tests.

Bi
 
If you downsize the 45.7 mp FF to 20.9 mp FF the results should be similar. This is known as normalizing and is the essence of the print and screen data on Dxomark. Bill Claff also does an normalization procedure for his tests.

Bi
Normalization is an accepted method of comparing apples to oranges, but it does have it's flaws. Another way is to put the high mp cameras in DX mode so you are comparing similar pixel densities. The larger sensil size of the lower resolution sensors will win every time, but that test has it's flaws as it ignores the advantages of higher mp sensors to capture more detail. Another test might be to res up the lower mp sensor image to the same size as the higher mp sensor image for comparison, but again that method has it's flaws. You just have to know what the flaws are in any mismatch comparison testing. My personal preference is to pick the sensor size you want for your particular genre of photography and compare it other cameras with similar sensor sizes. To me, it doesn't make sense to compare, say, an 24mp R3 to a 45mp R5 as they were designed for different types of photography.
 
I've been shooting my D850 and Z9 (currently a rental, but in theory my Z9 should be here within a week or so) side-by-side for both landscape and wildlife. From a quality perspective I see no real difference. One advantage I've found from the Z9 for landscapes which I didn't expect was the benefit of 20FPS. I shoot 5 shot brackets most of the time and then in post I either pick the best of the bunch and work it up or I take several of them and merge into an HDR. With the 20FPS, the merge gets a little better/easier because there's less movement in the scene between shots.

Clever ...................i will have a look at that Thankyou
 
Covering several points.............

A cats eye has a iris that opens and closes massively, hence they see things in the dark we cant, ie !!! they have huge pixels in their eyes, we don't LOL.

There are so many different ways and factors to access things with noise and pixels on different sensors, and the geeks love playing with the science, hay its fun and interesting at times, the bottom line simply for me is, what do you see on your screen or your print that you like or you don't that matters.

For me From the coal face street perspective............Large pixels pick up lots of light, easier faster in far less time and in more challenging conditions and by far making that style of camera far more forgiving.

Example D4, D4s, D5, D6 versus D800 D810.............along comes heaps of new technology say like Mirror Less, with the shutter removed, stacking introduced, juiced up sensitivity and heaps of horsepower and bingo you can now up scale resolution to 45 mp with improved ISO performance, with insane speeds, its as simple as that...........you can now do on a 45 mp sensor want you could do on a D4s D5 D6.......in the ISO department especially.

To answer a question D850 versus Z9 ISO or Z9 versus D6 ISO

Take the cameras in question, take the same images side by side, that's the only way to do it.

So far i haven't seen with the eye any noticeable difference between the D850 and the Z9 at high iso 12800 doing a Soccer match at night recently, or image quality, making the D850 the best value quality camera on the market, period.

D850 versus Z9

ISO, no visible difference seen with eye so far.

Tracking Z9 wins
Speed Z9 wins
Focus attack Z9 wins
Actuation life Z9 wins
Video Z9 wins

Image quality, no visible difference seen with eye so far.

D6 versus Z9 Tracking and ISO in the field so far.......

ISO in the field, the D6 is still just ahead......looking at the end results, no editing,

Tracking in the field of fast sports action in challenging conditions, the coal face experience to date for me and i am no expert just a photographer, is that the D6 is defiantly sticker based on myself and a few mates experiences, now this may change with more Z9 updates or more user experience over time with the Z9 but so far the D6 has the edge over the Z9, again this is in extreme conditions and may vary at times dependent on specific situations, however side by side to date the D6 is ahead. Others may disagree and that's fine. Myself the Z9 is great but the tracking is still an area to improve, as is the mirror less market as a whole.

The Z9 like all new Nikon's, it takes time to grow into it.
 
My bet is you are seeing the difference in the amount of light gathered by a FF 20.9mp sensor pixel vs that of a 45.7mp FF sensor pixel.

Assuming that the quantum efficiency of the two sensors is equal, the amount of light collected by the two sensors would be equal. The question is how do you compare the results? If you look at the two images at 100% in Photoshop, the 45.7 mp image will be much larger and noise will be more apparent with the larger image. To make things equal, you have to view the images at the same size. You could resize the 20.9 mp image to 45.7 so that the two images would be viewed at the same size. Alternatively, you could downsize the 45.7 mp image to 20.9 mp.

The 20.9 mp sensor collects more light per pixel than the 45.7 pixel. If you downsize the 45.7 mp image to 20.9 mp you have 4 read noises which add in quadrature, while with the 20.9 mp image there is only one read noise. If read noise is relatively high with both sensors, the 45.7 mp image will have a worse SNR, while if read noise is small, the differences will be smaller. In general, the larger pixel sensor will have an advantage. With the very low read noise of current sensors, the differences will be smaller.

For an excellent technical analysis, see Emil Martinec's essay here.

Bill
 
Assuming that the quantum efficiency of the two sensors is equal, the amount of light collected by the two sensors would be equal. The question is how do you compare the results? If you look at the two images at 100% in Photoshop, the 45.7 mp image will be much larger and noise will be more apparent with the larger image. To make things equal, you have to view the images at the same size. You could resize the 20.9 mp image to 45.7 so that the two images would be viewed at the same size. Alternatively, you could downsize the 45.7 mp image to 20.9 mp.

The 20.9 mp sensor collects more light per pixel than the 45.7 pixel. If you downsize the 45.7 mp image to 20.9 mp you have 4 read noises which add in quadrature, while with the 20.9 mp image there is only one read noise. If read noise is relatively high with both sensors, the 45.7 mp image will have a worse SNR, while if read noise is small, the differences will be smaller. In general, the larger pixel sensor will have an advantage. With the very low read noise of current sensors, the differences will be smaller.

For an excellent technical analysis, see Emil Martinec's essay here.

Bill

That is a cool article. I've read and reread it over the years and each time have a little different takeaway message. This time I focus on this quotation:

"Rather than having strong dependence on the pixel size, the noise performance instead depends quite strongly on sensor size -- bigger sensors yield higher quality images, by capturing more signal (photons)."

Which makes a lot of sense since more total light is gathered even though the light per unit area stays the same, if I interpreted correctly.
 
Nikon Z9 Vs Z7 Dynamic Range Comparison Test | Which camera is the king of landscape photography?

This is an interesting series of tests. Since you are using the same shutter speed and aperture for each test with the same lens, photographic exposure in terms of lx/sec is the same. At the relatively low magnifications, I don't see that much difference between the two sensors in terms of noise. However, many of the Z9 images appear darker to me. It would be good to use Rawdigger or a similar program which looks at the raw values which are proportional to the number of photoelectrons collected. Also, if you are using Adobe converters there may be a differing baseline exposure offset and the ISO calibration of the two sensors could be different as could the quantum efficiencies. Rawdigger would clarify many of these possibilities since it looks directly at the raw data. ISO determines brightness of the image with a given exposure, while exposure is not really a component of the exposure triangle. This a slide of a presentation I gave at my local camera club after thorough research.

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


Bill
 
Last edited:
Back
Top