Z9 Release Thread

[chambeshi]

Early days...pre-production firmware and subject to rigorous comparisons

"Is a Z9 as good as a D6 at high ISO? ....Given the above answer—that the Z7 II and Z9 are likely to be hard to tell apart in image quality, even while pixel peeping, I think it safe to predict the following: up through about ISO 2400 the Z9 usable dynamic range is going to exceed the D6's, while above that ISO setting the D6's usable dynamic range will be slightly better (a half stop would be a tentative guess). That, however, isn't likely to dictate the use of the camera. Many of those photographing sports and using high ISO values, for instance, aren't seeing their work output at 19" and 300 dpi. More like JPEGs that are 2mp to 4mp in use size on the Internet, actually. Unfortunately, we're going to see a continuation of this question basically forever, as people get deep down in the trenches and try to evaluate per pixel noise in ways that aren't always relevant to the way we output images. "

https://www.zsystemuser.com/z-mount-cameras/z-camera-articles/the-unanswered-z9-questions.html

[iustin]

Hmm, ISO performance is with the current pre production firmware a little behind the Z7II and miles behind the Z6II:
https://youtu.be/_im4K3yBoz4

Not really, see https://www.youtube.com/post/Ugkx_flJxNF9_Z-6lIZQkvX035qwsXLToxE-

[Birna Rørslett]

As the video was removed, we really wouldn't know

[Jan Anne]

Saw some comments about using Adobe software which doesn’t fully support the Z9 raw files yet, others mentioned that the color profiles were different, etc.

[Akira]

According to Adobe, Camera Raw ver. 14.0 supports Z9 but "preliminarily".

[richardHaw]

all shops here are on back-order!!!

[chambeshi]

Interview in chinese. The 2nd section discusses the 100-400 f4.5/5.6S

I did not try and count how many answers repeat "There is no answer to future products, but we will listen to customer feedback and plan." :-)

https://info.xitek.com/allpage/attitudes/202111/12-357400.html?_x_tr_sl=auto

[Ilkka Nissilä]

Has anyone seen a review or evaluation of the Z9 (e-shutter and EVF) in different indoor lighting environments?

I noticed something a bit disturbing yesterday as I walked around in Helsinki and took pictures with the decorative Christmas lighting contributing to the images, sometimes they were in the picture and sometimes they just provided some of the light. The EVF was flickering wildly, to the point where I stopped using it. This doesn't usually happen with the Nikon EVF and I'm wondering if there is some specific frequency in the lighting which the camera is unaware of and is unable to compensate for. I tried with flicker reduction on and off. The pictures came out fine. I'm wondering if the LED manufacturers use different frequencies and thus it might be difficult for the camera manufacturer to offer a generic fix to the issue.  I've also noticed at home that when I first lift the camera to my eye the viewfinder can flicker a bit until it settles in. This could be the camera "learning" the flickering behavior and optimizing a correction for it. I should check which lights cause it and maybe replace them with something else.

I am wondering if the Z9's faster sensor read time and EVF can offer a fix for this. Mostly the Z9 material I have seen has been in natural daylight or in environments with high-quality artificial lighting. But this is more a question of the variety of decorative lighting that is used by stores and on streets (shop windows can sometimes have this issue, though most don't).  If automated flicker reduction can't work with the variety of lighting then perhaps a system where the user can adjust the EVF update frequency and dial in manually frequencies which correspond to the lighting for the camera to be able to correct for flicker at those frequencies.  Correction could also be important for the pictures as there is no mechanical shutter.

[Ilkka Nissilä]

Interview in chinese. The 2nd section discusses the 100-400 f4.5/5.6S

This also mentions the flickering lights issue, though not for the EVF but for the electronic shutter.

"Question: In the case of electronic shutter, it is easy to cause stroboscopic problems due to indoor lighting when shooting indoors. How did Nikon solve this problem? Answer: Set the flicker reduction in the photo shooting menu to "ON" to reduce the stroboscopic problem caused by the flickering of the lighting under fluorescent lamps or mercury lamps and other light sources. When the light frequency is 50 Hz, the flicker period of 100 Hz is detected, and when the light frequency is 60 Hz, the flicker period of 120 Hz is detected. However, if the flicker period of the light source is variable during continuous shooting, the effect of flicker cannot be reduced. "

It seems the flicker reduction is designed to handle the normal electrical grid frequencies but LED lights can flicker at other frequencies. "Striped flickers may still occur under some LED lighting. For this we plan to launch a solution in 2022." Hmm. interesting, so they are working on the problem. I hope they address EVF flicker also and not only banding in the images taken with electronic shutter, and also I would like to see firmware updates to include less expensive models as well, not just the Z9.

With regards to the 100-400, it's interesting that they discuss all the troubles of the previous lenses though not specifically referring to them as flaws of the 80-400. ;-) They talk about quality control and design to minimize problems due to manufacturing tolerances at the extended zoom setting as well as lens balance and the resistance of the zoom when shooting from angles. It really sounds like they are getting the right feedback and are trying to correct problems of earlier lenses.

[iustin]

This also mentions the flickering lights issue, though not for the EVF but for the electronic shutter.

"Question: In the case of electronic shutter, it is easy to cause stroboscopic problems due to indoor lighting when shooting indoors. How did Nikon solve this problem? Answer: Set the flicker reduction in the photo shooting menu to "ON" to reduce the stroboscopic problem caused by the flickering of the lighting under fluorescent lamps or mercury lamps and other light sources. When the light frequency is 50 Hz, the flicker period of 100 Hz is detected, and when the light frequency is 60 Hz, the flicker period of 120 Hz is detected. However, if the flicker period of the light source is variable during continuous shooting, the effect of flicker cannot be reduced. "

It seems the flicker reduction is designed to handle the normal electrical grid frequencies but LED lights can flicker at other frequencies. "Striped flickers may still occur under some LED lighting. For this we plan to launch a solution in 2022." Hmm. interesting, so they are working on the problem. I hope they address EVF flicker also and not only banding in the images taken with electronic shutter, and also I would like to see firmware updates to include less expensive models as well, not just the Z9.

Thanks for the interesting extract (I have not watched the video). However, I have one question: since the e-shutter for the Z9 has readout speed on-par with a mechanical shutter, wouldn't such special LED lighting also cause problems for any other camera? Am I missing something?

[Ilkka Nissilä]

Thanks for the interesting extract (I have not watched the video). However, I have one question: since the e-shutter for the Z9 has readout speed on-par with a mechanical shutter, wouldn't such special LED lighting also cause problems for any other camera? Am I missing something?

The e-shutter at least in the A9 works by reading the data in blocks of lines whereas the mechanical shutter curtains move smoothly across the sensor surface (and the curtains are out of focus). This creates a different effect in flickering lights: sharp bands in the e-shutter vs. softer bands with the mechanical. The softer bands are less conspicuous. I don't know how the Z9 sensor is read; it might be that it is read one line at a time instead of blocks in which case it would be closer to how mechanical works.

The image taking can be slightly delayed to take place during the peak of the oscillatory cycle, this has been implemented in newer Nikons (since 2016). However, this technique isn't applied for the EVF.

The EVF data feed and updates are not necessarily happening at the same frequency as the main image data is read, the Z9 has a parallel pathway for the viewfinder data. I think we'll just have to see how the Z9 EVF works in practice in these kinds of (non-photographic) LED-lighting situations.

[Jack Dahlgren]

The e-shutter at least in the A9 works by reading the data in blocks of lines whereas the mechanical shutter curtains move smoothly across the sensor surface (and the curtains are out of focus). This creates a different effect in flickering lights: sharp bands in the e-shutter vs. softer bands with the mechanical. The softer bands are less conspicuous. I don't know how the Z9 sensor is read; it might be that it is read one line at a time instead of blocks in which case it would be closer to how mechanical works.

The image taking can be slightly delayed to take place during the peak of the oscillatory cycle, this has been implemented in newer Nikons (since 2016). However, this technique isn't applied for the EVF.

The EVF data feed and updates are not necessarily happening at the same frequency as the main image data is read, the Z9 has a parallel pathway for the viewfinder data. I think we'll just have to see how the Z9 EVF works in practice in these kinds of (non-photographic) LED-lighting situations.

Fluorescent and other lights flicker at the frequency of the mains power supply (typically 50/60 hz). This is usually longer or shorter than typical shutter transit times. LEDs flicker at what ever frequency their own power supply runs at. There are ways to make an LED power supply with no flicker, or in the case of cheap decorative lights, the power supply is based on the cheapest possible circuit which often induces high frequency flicker. I think that making the shutter transit time such that it does not often correspond to the common frequencies is the best that can be done by camera makers. This is a problem that can be mitigated, but not completely solved as the camera is capturing the reality of the situation, even if our desire is to want the lights to appear to be always on. Perhaps an eventual solution is multiple exposures for very high frequency flicker.

[Ilkka Nissilä]

Fluorescent and other lights flicker at the frequency of the mains power supply (typically 50/60 hz). This is usually longer or shorter than typical shutter transit times. LEDs flicker at what ever frequency their own power supply runs at. There are ways to make an LED power supply with no flicker, or in the case of cheap decorative lights, the power supply is based on the cheapest possible circuit which often induces high frequency flicker. I think that making the shutter transit time such that it does not often correspond to the common frequencies is the best that can be done by camera makers. This is a problem that can be mitigated, but not completely solved as the camera is capturing the reality of the situation, even if our desire is to want the lights to appear to be always on. Perhaps an eventual solution is multiple exposures for very high frequency flicker.

Well the intensity of the light flickers at twice the frequency of the mains voltage (in conventional lights such as fluorescent, halogen, tungsten). LEDs run on DC current and I suppose the quality of the DC generated from AC may cause some of these issues, and the other thing is the dimming technology, where the LED is switched on and off at high frequency. Another technique for dimming would be to just reduce the current and maintain a DC current at a different level. I suppose I could take a photodiode and oscilloscope to study what the frequencies are in these decorative lights and then figure out how the EVF flicker might be mitigated.

[Jack Dahlgren]

You can use your camera. Set a 1 second exposure and pan the camera so that you keep the light in frame. Then count the number of times the light appears. 

[pluton]

I hope they address EVF flicker also and not only banding in the images taken with electronic shutter, and also I would like to see firmware updates to include less expensive models as well, not just the Z9.

Whether Nikon, or any camera maker, will implement a useable and useful anti-flicker feature, I cannot say.
Video cameras of mid-level and higher quality have had an adjustable flicker/shutter feature for many years.  It first became necessary in the mid-1980's. when personal computer CRT screens became common, and simultaneously, the sensors in the video cameras changed from plumbicon tubes, which were largely 'blind' to the flicker,  to the new, modern, flicker-sensitive CCDs.