Nikon ZR Lab Test – Rolling Shutter, Dynamic Range, and Exposure Latitude

by Gunther MachuOctober 23^rd, 2025

Subscribe to Lab Tests

The Nikon ZR is generating quite a hype since it was released shortly before IBC 2025: it offers a compact, full-frame body with internal 6K REDCODE RAW NE (Nikon EXPEED) recording and a joint RED / Nikon label screaming “cinema” at the back at a really attractive price! We were eager to run it through our CineD Lab Test procedures to see if the image quality lives up to the hype. Interested to hear more? Then read on …

I have to admit that I wasn’t quite immune to the hype, and when the Nikon ZR finally arrived at our CineD headquarters here in Vienna, I couldn’t wait to grab my colleague, Florian, and together run it through our standard CineD lab torture test. Again, a big “thank you” to Florian for helping me shoot and analyze this test. This time, it was really three lab tests in one, as we wanted to compare the three different RAW codecs available in this camera: REDCODE RAW NE, N-RAW, and ProRes RAW.

Johnnie’s camera review is now available here.

So, let’s start as usual with rolling shutter.

Rolling Shutter of the Nikon ZR

Again, we use our strobe light at 300Hz to determine the rolling shutter via the sequence of black and white bars.

For full frame REDCODE RAW NE, we get the following image, 9.4ms rolling shutter (less is better). For a full-frame 6K readout, this is a really good result and exactly the same as the close sibling Nikon Z6 III, available in our lab database here.

In DX (or APS-C mode), we are getting 6.3ms, thus enabling a 4K120 frames per second readout mode as well.

Nikon ZR – Dynamic Range at ISO800 and ISO6400

If you are not familiar with how we perform our dynamic range testing, then please head over here first.

Let’s start with REDCODE RAW NE in full frame 6K. These are our development settings in DaVinci Resolve 20.2.1 in the camera RAW tab (note that we have disabled the default “Chroma Noise Reduction” – we always test without noise reduction):

Looking at the waveform plot, we see a familiar phenomenon:

I have expanded the RGB curves towards 5600K using the white balance slider to demonstrate a RED-specific phenomenon, which is called “highlight recovery”, built as default into the IPP2 color science (REDWideGamutRGB, Log3G10) – visible for the second patch from the left, which is not clipped, but the red and green channels are recovered and not visible. This is why we can only start to count from the third patch (from the left), which is not clipped – please also have a look at our RED KOMODO X Lab Test here, or the RED V-RAPTOR [X] 8K VV Lab Test here. The third to the fourth patch is the first stop of dynamic range, the fourth to the fifth is the second stop, and so on.

We can see 10 stops of dynamic range above the noise floor. Running this image through IMATEST, we get 10.2 stops at a signal-to-noise ratio (SNR) of 2, and 11.5 stops at SNR = 1.

Looking at the “Noise spectrum” graph on the lower right-hand side, we see good amplitudes at high frequencies, hinting at a very detailed image without much internal noise reduction.

Now, shooting the XYLA21 chart using Nikon RAW (N-RAW), we get 9.7 / 10.9 stops at SNR = 2 / 1 (almost exactly the same as the Nikon Z6 III), and using ProRes RAW, we get 10.2 / 11.2 stops at SNR = 2 / 1. In essence, the “highlight recovery” feature built into RED’s IPP2 color pipeline seems to trick IMATEST into slightly higher scores.

These values are all in the middle to lower ground for a compact hybrid consumer camera. Recently, the Panasonic LUMIX S1 II showed what is possible, especially with their “dynamic range boost” feature turned to “ON” – it shows 2.5 stops more dynamic range than the Nikon ZR for ProRes RAW (when comparing dynamic range values between cameras, you should always compare apples to apples, hence RAW codecs vs RAW codecs and internal compressed codecs vs internal compressed codecs).

Please have a look at the CineD database for various additional results, including ISO6400 (spoiler: we are losing between 0.5 to 1 stop at ISO6400 depending on the codec). Let’s just quickly have a look at the H.265 N-Log ISO800 waveform:

You can see a massive, massive amount of internal noise reduction at work (look at the thin noise floor, and how clean it is). IMATEST shows the following:

Well, we are now getting 12.5 / 13.4 stops at SNR = 2 / 1. But have a look at the “Noise spectrum” graph in the lower right corner: you can see that frequency amplitudes drop very fast, having values of only around 0.1 at the higher frequencies – hence, noise reduction destroys all the high resolution details. REDCODE RAW NE still shows amplitudes of around 0.3 at the same high resolutions.

Again, please have a look at the databases for all the other dynamic range results.

Nikon ZR – exposure latitude at ISO800

Latitude is the capability of a camera to retain details and colors when over- or underexposed and pushed back to base exposure. Some time ago, we chose an arbitrary value of 60% luma value (in the waveform) for our subjects’ faces (actually their forehead) in our standard studio scene. This CineD base exposure should help our readers get a reference point for all the cameras tested, regardless of how they distribute the code values and which LOG mode is used.

REDCODE RAW NE: we used 6K 25fps R3D NE at ISO800, our trusted Zeiss Compact Prime 85mm T1.5, and developed the files using the camera RAW tab again in DaVinci Resolve 20.2.1. Using the exposure and ISO, we normalized the files to base exposure. Then, we brought everything into the REC709 color space by using a Color Space Transform (CST) from R3D to DaVinci intermediate/wide gamut, and then using another CST node to REC709.

Similarly, we used the following settings as a starting point for ProRes RAW and N-RAW in DaVinci Resolve’s RAW tab (finally, we are able to use ProRes RAW in DVR!!), and then again we used color space transforms from the respective color spaces to REC709:

Nikon ZR – comparison of ProRes RAW versus R3D at ISO800, close to clipping highlights

First, we need to talk about a phenomenon we observed when we started with ProRes RAW and then switched over to R3D, using the exact same exposure settings (ISO, lens iris, and shutter speed + studio light setting):

Exposing for ProRes RAW in the usual way, hence having the forehead of Nino as close as possible to clipping, we get the following waveform – resulting in 4 stops above base exposure. You can see, all the color channels are intact, whereas the white paper on the left is partially clipped:

Now, leaving the exposure exactly the same and switching the camera to REDCODE RAW NE, we get the following waveform (ungraded):

Hence, as we are close to clipping on Nino’s forehead, we seem to be already in the region of RED’s highlight recovery, which combines the red and green channels and thus destroys the skin tones (rendering them yellowish). Also, exposure values are mapped to lower code levels in Log3G10 and RED WideGamut RGB for compatibility with the RED workflow.

This is a bit of a strange implementation of the built-in highlight recovery feature. It should only become active if one of the color channels is already clipped! But as can be seen in the ProRes RAW implementation, on Nino’s skin, all color channels should be intact at this exposure level.

Hence, this fact reduces the highlight latitude of REDCODE RAW NE by about 1 stop when compared to ProRes RAW (and also Nikon N-RAW). We need to lower the sensor exposure (studio light levels) to get out of the highlight recovery region and keep the skin tones intact. This will definitely penalize shadow recovery. Hence, I don’t think we’re getting the full potential with this implementation of R3D in the Nikon ZR.

We reached out to Nikon, and they responded in the following way:

“What you have faced is as we have designed. The reason is difference of highlight clipping points between N-Log and R3D NE. Therefore, when you shoot using R3D NE, you need to set the exposure according to the clipping point of R3D NE. Besides, to prioritize reducing noise in dark areas, highlight has been shifted one stop to low light … The actual image sensor sensitivity differs by one stop between R3D NE and other codecs.”

Hence, please be aware of this fact. Looking at the base exposure of all 3 RAW codecs, we can see R3D and N-RAW look very similar, whereas ProRes RAW is a bit more greenish:

In summary, to prevent highlight recovery in action (using the lower light exposure levels) within the RED Log3G10 gamma curve, this will result in 2 stops overexposure above base as the maximum for R3D, and 4 stops above base for ProRes RAW and N-RAW. Within the R3D workflow, this is consistent with other cameras like the KOMODO-X and V-RAPTOR [X] VV 8K.

Don’t confuse this with a loss of 2 stops highlight latitude – code values are remapped to lower levels in the R3D workflow. We are talking about a loss of 1 stop in the highlights compared to ProRes RAW or N-RAW at the same exposure level to prevent highlight recovery for accurate skin tones.

Now, moving to 3 stops of underexposure (i.e., 7 stops of exposure latitude), we get the following images (2 over to 5 under = 7 stops for R3d, 4 over to 3 under = 7 stops for ProRes RAW). We are now utilizing the default “chroma noise reduction” in the camera RAW tab of DaVinci for R3D:

In both images, we are already seeing heavy noise, and the R3D image gets slightly pinkish, whereas the ProRes RAW image gets greenish. Let’s apply noise reduction:

They both clean up nicely. Now let’s look at 8 stops of exposure latitude, namely 4 stops under for PRR and 6 stops under for R3D:

Again, but more pronounced: R3D keeps the colors overall in a better shape, but shadows become more pinkish, whereas ProRes RAW overall loses color accuracy and shadows become greenish. Let’s see if we can clean this up with NR:

Now, it becomes difficult to apply the right amount of noise reduction without losing too much resolution and getting that plasticky look. R3D still keeps the colors somewhat intact (although shadows have this cyan/pinkish color cast), whereas ProRes RAW loses all colors, and skin tones become orangey/greenish.

N-RAW looks like this, applying noise reduction – color representation is a bit better than ProRes RAW, but the image also turns greenish:

We are at 8 stops of exposure latitude, and things are definitely very borderline here. The moving images of all three RAW codecs are actually not really usable anymore, as larger blotches of chroma noise are hovering over the image.

Just for the sake of completion, here are R3D and PRR at 9 stops of exposure latitude:

This cannot be saved in post, have a look at all three RAW flavors here, using heavy noise reduction:

All images are broken beyond repair, but R3D and N-RAW behave better than ProRes RAW. Where the one becomes pinkish, the other becomes greenish. Hence, 8 stops of exposure latitude is the limit! I think there would definitely be additional potential for the R3D implementation, because so far, it penalizes highlight stops as described above.

Putting this into context, the Panasonic LUMIX S1 II managed 10 stops of exposure latitude (Lab Test here) with DR Boost “ON” (9 with DR Boost “OFF”), albeit at the expense of a high rolling shutter. I really urge you to have a look at the LUMIX S1 II results again. You can clearly see how much better the LUMIX S1 II behaves in the exposure latitude discipline! The Sony A9 III (Lab Test here) with the global shutter sensor exhibited 9 stops of exposure latitude, just using the 10-bit internal compressed codec! And of course, the king in terms of latitude is still the ARRI ALEXA 35, exhibiting 12 stops of exposure latitude.

Nikon ZR CineD Lab Test – summary

The Nikon ZR is a very well-rounded package at this price point from an image quality perspective. Very good (low) rolling shutter values, average but OK dynamic range figures, and an average 8 stops of exposure latitude put it in the middle ground of full-frame consumer cameras. You do get a lovely-looking image with REDCODE RAW NE internal recording and RED color science (although we see some potential for improvement here, as mentioned above), and in total, three internal RAW flavor options to choose from, with the bonus of a close match with other RED cameras when using the R3D workflow.

Are you interested in using the Nikon ZR for your work? Have you shot with one already? Let us know in the comments below.