Nikon Z8/URSA 12K - Why we need to move past 4K & 6K

[Zack Birlew]

Hi, everybody, I just saw this Youtube camera test and it's really incredible nobody else has done a nice comparison like this:

This test shows a great detail difference between the 8K of the Nikon Z8 and the URSA 12K at 12K. Granted, the 12K has been delivered as 8K but the detail difference is clear with a very noticeable pop of detail to everything! While everyone is clamoring for new 4K and 6K options, the incredible amount of detail and visual impact on display with affordable 8K and 12K really shows why higher resolution cinematography is the future. The infinity landscapes show that at 8K there is a great amount of detail but on the mountaintops the trees are still blurred but can clearly be defined at 12K by comparison. Imagine tracing the image of that mountain line and how much easier it would be at 12K. I know there was that MTF test that was made not too long ago where the URSA 12K placed last against even the ARRI LF but clearly something had to be off with the test or it simply doesn't apply to the URSA 12K's new sensor design, I don't know, but the difference here is very clear.

Nikon is already doing a great job at 8K and I know that the Z8 and Z9 are on my list of cameras to get but if there's going to be a leap in video resolution it will probably be coming from the RED side of things. With Blackmagic already about to release the Cine 12K FF and the Cine 17K is not too far off from now, I'm eager to see if there will be a major response from RED or even other competing manufacturers.

 

[Phil Holland]

The more important question is to ask why the shot from the Z8 at around 01:35 matches the LoD of the 12K more than the other shots.

 

[Zack Birlew]

There is the odd shot in this test that doesn't have a 12K comparison shot but in any case, the switch to 12K is very noticeable. The other settings like F-stops, ISO, etc. aren't listed so that may have an affect as well. I wouldn't mind seeing more tests like this, a lot of URSA 12K stuff is just slow motion 8K and/or delivered in Youtube 4K so the effect at 12K is just lost.

 

[Thomas B.]

Agreed, Zack, a high(er) resolution image is better suited for big screen productions.
However, I find it very difficult to compare the image in this video.
Unfortunately, he used (or showed us mostly) the UWA-zoom on the 8k camera and the normal-length to tele zooms on the 12k camera.
Wouldn't it have made more sense to use the same lenses (contrast-/resolutionwise) on both cameras (and crop the ursa image)
or at least to use the ursa for wide angle shots and the Z8 for normal/tele shots (and maybe crop the ursa image for detail comparison)?

 

[Phil Holland]

There's several potential answers why the one Z8 shots looks better than the rest, but that should be the only shot that is ideally worth comparing.

I've tested both cameras extensively, in particular, with the same lenses. This isn't indicative of expected results.

 

[Zack Birlew]

Well, maybe a nice cinematic 8K vs. 12K comparison video could be in the cards? The base rules would be max resolution at all times with the only exception to be an 8K delivery due to streaming limits. Other than that, no lower resolution slow-mo or overuse of super close macro shots, just straight what can be done at max resolution frame rates from normal filming distances and compositions. Lastly, forget being wide open, limited to F2.8 or slower, the test is for resolution not blurred out bokeh collages.

 

[Phil Holland]

Well, maybe a nice cinematic 8K vs. 12K comparison video could be in the cards? The base rules would be max resolution at all times with the only exception to be an 8K delivery due to streaming limits. Other than that, no lower resolution slow-mo or overuse of super close macro shots, just straight what can be done at max resolution frame rates from normal filming distances and compositions. Lastly, forget being wide open, limited to F2.8 or slower, the test is for resolution not blurred out bokeh collages.

I haven't published my last 9 camera tests, mainly due to time. I don't wrap production until late November and will be in post after that for a few months. The new Cine 12K and 17K I'll test more closely after wrap, but I also currently have footage from the Cine 12K I was tinkering with prior to start of filming.

Without controlled tests, most observations are just throwing wildly darts and not knowing where the targets are.

I'm for higher resolution acquisition obviously, but there's a lot that needs to be looked at beyond simple A/B comparisons. I might dive into those topics perhaps when the time is right.

 

[Jon Pais]

Hi, everybody, I just saw this Youtube camera test and it's really incredible nobody else has done a nice comparison like this:

This test shows a great detail difference between the 8K of the Nikon Z8 and the URSA 12K at 12K. Granted, the 12K has been delivered as 8K but the detail difference is clear with a very noticeable pop of detail to everything! While everyone is clamoring for new 4K and 6K options, the incredible amount of detail and visual impact on display with affordable 8K and 12K really shows why higher resolution cinematography is the future. The infinity landscapes show that at 8K there is a great amount of detail but on the mountaintops the trees are still blurred but can clearly be defined at 12K by comparison. Imagine tracing the image of that mountain line and how much easier it would be at 12K. I know there was that MTF test that was made not too long ago where the URSA 12K placed last against even the ARRI LF but clearly something had to be off with the test or it simply doesn't apply to the URSA 12K's new sensor design, I don't know, but the difference here is very clear.

Nikon is already doing a great job at 8K and I know that the Z8 and Z9 are on my list of cameras to get but if there's going to be a leap in video resolution it will probably be coming from the RED side of things. With Blackmagic already about to release the Cine 12K FF and the Cine 17K is not too far off from now, I'm eager to see if there will be a major response from RED or even other competing manufacturers.

I'd prepared a longer response, but I feel physically ill after watching that video sitting 1.5 picture heights from my 65" QD-OLED. All of the footage is of extremely poor quality, much of which looks out of focus. The number of Ks alone has got little to do with perceived sharpness.

 

[Tom Roper]

There's several potential answers why the one Z8 shots looks better than the rest, but that should be the only shot that is ideally worth comparing.

I've tested both cameras extensively, in particular, with the same lenses. This isn't indicative of expected results.

Hi Phil,

I shot the 8K/60 fps video Zack referenced here. Although it is not represented as a controlled test, I own the cameras and the land they were shot from. Both cameras have similar LoD at 7680 x 4320, the U12K scaled from 12288 x 6480, The Z8 scaled from 8256 x 4644. As mentioned in the YouTube comments, several lenses were used on the U12K but the Z8 was shot only with the 24-120 kit lens; it was all I had for it at the time.

In order to obtain a similar FOV for the frame grab that follows, I used an EF 85mm F1.4L IS USM on the U12K and shot the Z8 at 120mm focal length of the kit lens. The U12K pixels are only 2.2 micron. In order to avoid diffraction limited resolution in accordance with Rayleigh's Criterion, I opened the iris on the 85mm lens to f/2.8. The Z8 kit lens doesn't open past f/4 but its larger pixels are not diffraction limited. The result of the side by side 500% crop from Resolve is shown below.

I don't see a significant difference, both seem good to me, Z8 on the left, Ursa 12K on the right.

Z8_vs_U12K_500%.JPG

drive.google.com

 

[Tom Roper]

I'd prepared a longer response, but I feel physically ill after watching that video sitting 1.5 picture heights from my 65" QD-OLED. All of the footage is of extremely poor quality, much of which looks out of focus. The number of Ks alone has got little to do with perceived sharpness.

The footage is bad I agree, particularly the last clip, I don't know how I let it go out looking like that. But that said, people talk the same bullshit when 4K oversampled from 6K is said to be appreciably sharper than the native 4K itself. They see THAT difference, but when it comes to actual 8K, oh no..not a factor, K's don't have anything to do with sharpness.