does anyone out there own a D810 that is in the unaffected serial number range?

Started Aug 19, 2014 | Discussions thread
Tzvi R New Member • Posts: 2
Re: does anyone out there own a D810 that is in the unaffected serial number range?


In past advisories, Nikon usually posted ranges of serial numbers. I suspect that this strategy where one has to type their serial is used to cover that all cameras are affected, that this is a result of some hardware design issue. (not necessarily a problem but perhaps a conscious design decision that backfired unexpectedly) It is also unclear as to what "firmware" is being upgraded? If it's a regular camera firmware then why didn't they just post it online? This may be some firmware belonging to an internal part, to which the user has no direct access.

There is always some degree of noise in every sensor - and this manifests itself especially in raw images when you push the exposure quite a bit, making the noise clearly apparent.

I don't have the "black dot" marking (I suspect that only the most recent batches may have it) and my D810 shows as affected on Nikon's web site. That said, I have taken several long exposures with my camera, starting from pure black (cap on, viewfinder covered) to brighter and brighter scenes by manipulating the aperture while shooting 30 seconds exposures at a white wall (since the sample images I saw posted on the web manifested noise only in partly bright areas, not in the extremely dark or extremely bright ones - but a whole flurry of sprinkled white dust).

I took RAW images of a white wall (perhaps that's a bad subject? Perhaps I should have chosen a background with some distinct color) all in 30 seconds exposures with variable apertures. Opening them all in LR5.6 they show the average gray level at 68%, 47%, 30% and 0.3% (pure dark).

I zoomed in at 1:1 and scanned for white spots, while playing with the exposure slider pushing it both up and down a few stops (while avoiding blowing highlights).

My conclusions:

None of the bright images (that contained some light in them, that weren't completely dark) had demonstrated any white spots. It seems like even the small amount of light had completely overcome the noise levels.

The completely dark scene on the other hand, had started showing randomly distributed white spots when the exposure slider went past 2 stops, however, this is to be expected since noise is a dominant factor in such a dark image, and pushing 3 stops I can see the noisy spots at 1.5-2% brightness (using the white balance picker at 11:1 magnification). Increasing the exposure, clarity and reducing contrast will make the effect worse.

So I can't say I see the results reported by others. Either I'm not replicating the conditions correctly or there's some variance between the cameras and I got lucky (or they got unlucky). All I see is standard, expected noise levels.

Here's my 30sec fully dark image through various processing steps. This is the only image that exhibited visible noise. The brighter images did not show any noticeable noise.

Dark frame, 30 sec exposure f/5.6 with the cap on, taken from a 14bit lossless compressed raw file

Perhaps trying to shoot at the sky during a dark night may be more revealing although I suspect that the stars may interfere with test

-- hide signature --


 Tzvi R's gear list:Tzvi R's gear list
Nikon D300 Nikon D810 Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF Nikkor 35mm f/2D Nikon AF Nikkor 50mm f/1.8D +7 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow