Jump to content
UltravioletPhotography

Nikon Z6/Z7 as a UV/IR Conversion: the Bad News


Andrea B.

Recommended Posts

I would not be surprised if the core of the new NX Studio in fact originated from the Silkypix crew, as there are strong similarities in the UI. Silkypix is a little less polished in terms of the user's interaction though. And maybe the Ichikawa labs, maker of the Silkypix software, kept a handful of goodies for their own program.
Link to comment

Anyway, I verified the Z6 can be used for UV video. I still need to work on the adjustment of exposure as the camera doesn't respond in the same manner as the PrimaLuce Z5.

 

If any clip comes out passably, I'll put it up here later.

Link to comment

Another minimalistic Nordic UV Noir .... With Z6, Coastal APO 60/4 APO, Baader U. Straight off the camera.

 

V202103054416.mp4

 

In case anyone wonders about the slight tendency to corner darking, this is due to the wrong lens shade being used. My bad.

 

A frame grab from the video is below,

 

vlcsnap-2021-03-05-16h10m06s245.jpg

 

I cannot detect any striping at all. Curiouser and curiouser.

Link to comment
  • 2 months later...

Hi all, found this post whilst trying to find a solution to my Z5 PDAF banding problem (that I didn't know I was going to have when I had the camera converted :sad: Doh!), and been having a play about with various potential solutions. I tend to use SilverEfex for my B&W conversions, and like higher contrast images, which obviously exacerbates the PDAF issues. Kind of glad I didn't convert the Z7 - the Z5 was a very expensive "cheap" experiment in full spectrum / one camera for everything! and it's fantastic! If it wasn't for the banding!

 

Off the back of this post, I've downloaded Silkypix, but the interface really is... unique! :D What settings should I be changing as the best way to "fix" the banding? I'm guessing it's Noise Reduction, and the suggestion is Neat Noise (I set 63%), but thought it worth learning from other peoples experience. Anyone able to advise??

 

Also been playing around with some other suggested solutions, including surface blur, DXO 4 (which I already had) and Topaz Denoise (demo) - from the *very* limited testing I've done so far, Topaz Denoise AI (at auto, and non RAW settings cos I've not tried anything else yet) and Silkypix seem to give the best results of reduced / removed banding but retained detail / sharpness Can see what I've done here:

 

https://flic.kr/s/aHsmVxrTX1

Link to comment

There is a devious work-around for the Z6, which in some cases might suffice. That is, using the capability of shooting stills in video mode. One does not have to run a video at all, just set the camera to "Video", then push the shutter release as usual. You will get only jpgs, however, they are surprisingly good. With the Z6 set to 4K video mode, the stills are about 8.3 MPix (3840 x 2160).

 

Set the Picture Control to Neutral or Flat, and cast the jpg into 16-bit mode before attempting any further editing. The file white-balances nicely into pretty standard false UV colours either in Photo Ninja or Photoshop.

 

The scene below was shot with the UV-Nikkor at f/4.5, 1/20 sec, ISO 1100 under heavy overcast skies, Apart from the undeniable softening brought about by my shaky hands (years of asthma medication take their toll!!), it is at least the equal of what I could obtain with any of my UV-capable DSLRs.

 

T202105114494.jpg

 

So, is the darned PDAF striping entirely gone? No, not really, but the banding is much subdued from the onset and a quick cleaning with Topaz Denoise using the banding removal parameters set to Horizontal and Width = (5-10) will tidy the scene up to satisfaction.

 

A small drawback however is that this approach prevents using the new Voice Memo feature, highly useful to an experimental photographer, since the Movie Recording button only starts video, not voice recording, when the camera is set to Video mode. You might, however, switch the mode back to Stills, play back the jpg then make your Voice Memo associated with it.

 

A similar approach should work with say the Z5 as well. However, I'm fully aware that this pseudo-solution will not utilise the full potential of the camera.

Link to comment
The same approach works for the Z5 as well, with even less striping initially visible. Plus one gets a "free" 1.7X TC since the Z5 in video mode operates on a cropped sensor.
Link to comment

I am surprised to read that the Z5 has a PDAF banding problem. I thought it was mostly in the clear from Birna's tests.

I wonder why the difference? Was this new or older firmware version?

Or caulk it up to different on sensor filters post conversion.

Link to comment

@dabateman, did you look at my gallery link to see the extent of what I'm seeing? A few examples there (and a few attempted remedies) though I might need to rearrange the gallery and annotate the pics a bit better!

 

I wasn't even aware of the problem before I did the conversion unfortunately, but it was painfully obvious on my first processed IR pic. I'm still trying to decide whether it's an acceptable problem that I can fix well enough and easily enough in post and live with it, or if it's a deal breaker. At the moment, I'm thinking the fix-in-post options (particularly Denoise AI, maybe DXO which I own but don't use, and maybe Silkypix, but that wouldn't fit in my workflow) seem pretty good before I even start looking at adding sharpening, and even then I'm half thinking the benefits of mirrorless far outweigh the downsides (and I replaced my colour D810 with a colour Z7 at the same time, so I'm kind of bought into it now! Though haven't sold my old cameras yet...).

 

It's not on the newest firmware - I've uploaded it, but just not been out with the camera since to see whether there has been an improvement. I wouldn't have thought the on-sensor filter would have made a difference? But it's an interesting question...?

 

It is visible on the RAW, but barely - I do tend to push microcontrast a little, and use higher structure SilverEfex filters routinely on my IR conversions. It's not affected by channel swapping / not channel swapping

Link to comment

Did you see this thread by Birna

https://www.ultravioletphotography.com/content/index.php/topic/4191-primaluce-nikon-z5-full-spectrum-camera-review/page__view__findpost__p__39866

 

Her Z5 looked good, especially compared to the Z6 which has many problems. Maybe Primaluce used a specific filter glass on sensor to avoid the problem or has AA filter or maybe Birna was just really lucky.

 

Where did you get your converted? I don't see it listed oddly on the Kolari website as a Nikon option.

Link to comment
  • 1 month later...
Bill De Jager

Speculation, but from a reliable source. This gives some future hope that the problems with UV conversions of Nikon's Z-mount cameras will eventually be overcome.

 

I keep seeing clues that some future Nikon camera—likely the Z9—will use all photosites for focus and introduce a focus system with better vertical data discrimination.

 

https://www.zsystemu...ovide-a-z9.html

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...