With Suggestions of Pattern/Banding Noise in the New Nikon Z7, It Might Make Sense to RENT Before Buying
Update Oct 10: field use shows no pattern noise issues.
I’ll have the Nikon Z7 system for review in a couple of days.
With dpreview.com reporting on pattern/banding noise in the Z7 due to the addition of PDAF pixels, I’m 'scared': a very large part of my work involves contrast control, e.g., bumping up the shadows and/or high dynamic range images. To date, all Nikon cameras have been superlative for years in this regard: Nikon D800, Nikon D800E, Nikon D810, Nikon D850.
If Nikon has screwed the pooch with ugly pattern noise like the troubled Sony A9, then the Z7 is not something I’d consider for anything but testing. Problem is, I can’t afford to dump $4K into a camera I won’t want to use myself for other stuff.
Accordingly, a key focus of my initial testing will be Nikon D850 vs Nikon Z7 in terms of image integrity under demanding field conditions. I'll be picky and my reference standard will be Nikon D850 at ISO 64.
One option for many users: rent a camera system before buying.
Nick C writes:
I saw your post, as well as DPR’s post, and also the discussion on their forums. I cannot detect any banding in the NEF files from my Z7. 14bit, lossless compression. I am pushing in ACR as much as it allows me and cannot say that I can detect a trace. There are others who claim they don’t see banding either, so who knows what gives.
My apologies. I do indeed see the banding, but you have really, really, really want to push in an unrealistic fashion. Like in a test where I have a white, sunlit billboard, and part of the frame is covered by an overhang. If pushed +5, the dark area does indeed show banding. However, this is in no way a realistic shooting scenario.
DIGLLOYD: (not speaking to Nick C in particular)—my experience with every major camera over a decade tells me that stuff exists that people don’t see for various reasons. Cognitive commitments* are a huge reason, but also just shots that don't show it except faintly. See my Sony A9 examples.
An issue that is not seen “most of the time” does not make it a non-issue. In particular, my outdoor shooting regularly demands all of the dynamic range available. I’ll be applying all my years of practical experience to proving or disproving that an issue exists or does not with the Nikon Z7, Canon EOS R and Fujifilm GFX-100S (which also has PDAF pixels).
* “most of the time” might mean 50% or even 100% of the time in some outdoor shooting scenarios. Or 1% in some general “average”—how does that mean anything if you’re in the 50% shooting situation?. It’s a straw man statement to say “most of the time”.
“Realistic shooting scenarios” always involve high dynamic range, so I doubt that a +5 push is the only circumstance of relevance. Consider Shadows=+100 in Photoshop Adobe Camera Raw— I do that fairly often. Doing so is almost the definition of dynamic range—the ability to capture and use a range of dark to light tones.
“Like in a test where I have a white, sunlit billboard, and part of the frame is covered by an overhang” ==> like in just about every landscape shot of a certain genre, say, like this one?