Some Initial Z7 Impressions

I’ve been shooting extensively with the Z7 this week as I work on reviewing the camera. I know many of you are waiting for the specific information that usually appears in the performance sections of my reviews, so I thought I’d give you some initial impressions and data points. I’m continuing to shoot with the camera and work through learning its ins and outs, so what I write next is subject to change as I develop more experience.

Battery: The first full set of shooting to exhaustion was 1375 shots (6% battery life left). There appears to be a problem, though. If you really shoot all the way until you get the message that the camera can’t continue shooting, when you replace the battery with a new one the top display will show ERR. You have to press the shutter release and take a dummy shot to release that error message. This is one of those things I need to check with other copies of the camera, as it could be a quality issue with my unit.

Buffer: I’ve noticed some oddities about the buffer performance when you’re pushing the camera constantly, as I am on the animal shoots I’m doing. Very similar to the Sony A7 models, if you start reviewing images while the buffer is clearing, there can be delays and the write-to-card-light stays lit for a long time. The camera appears to give preference to write over read, but I need to do some measurements of actual write performance, as it seems slightly sluggish to what I’d expect with fast XQD cards.

Autofocus: In the continuous autofocus modes the Z7 isn’t a D850. Or even a D7500. This is particularly true with lenses on the FTZ adapter, which forces the camera to always use phase detect. In my initial opinion, the Z7 is substantively behind where the Sony A7Rm3 is for continuous autofocus sequences of objects moving towards you (even at a diagonal). The faster and closer the movement is, the more likely the camera is to get a low hit rate on continuous focus. I’m still exploring the variables here, though. There may be techniques or settings that can improve that. 

Don’t get me wrong, it’s not that the camera can’t focus on fast-moving subjects in continuous autofocus. Witness:

bythom US MT Kalispell TripleD Z7 57607

Z7 with 70-300mm f/4.5-5.6E AF-P lens in continuous autofocus (in this case, Dynamic 9 point). And yes, that cat's quite close to me.

But the entire sequence of the tiger running across the water to me was not as good as I’d have gotten if I had used my D850. Ditto other subjects I’ve tried, and I’ve got plenty of subjects that are moving at speed and erratically this week; it’s a solid test of the system. My initial assessment is that the Z7 isn’t as bad as some suggest, but it isn’t as good as others suggest, either. Moreover, given the choice of the Sony A7Rm3 and the Z7 on these subjects, I’d likely pick the Sony. (Actually, I’d pick my D5, which has no trouble at all keeping focus.)

That said, I need to do more testing with many other lenses and many other types of moving subjects to form a full opinion.

Image Quality: I think those that think they are trying to squeeze the last drop of toothpaste out of the tube probably should be ignored at this point. Is there evidence of non-random noise in the FPN? Yes. Can you promote that to visible banding? Sure, but I can do that easier on my Sony cameras, despite the fact that they measure slightly better. Have I ever encountered an image where I’ve done that? No. 

Generally, I want blacks to stay black, and that’s where the fixed pattern noise lives. I’d have to have missed exposure so badly that I’d have to blame myself before the camera to see any evidence of “the problem” that’s getting written about so much. 

Does this type of noise reduce dynamic range? Well, it might if you used some statistical method of measuring dynamic range—the so-called engineering dynamic range, for example—but no it doesn’t if you’re leaving the blacks down near black. As many of you know from my other writing and books, I use a different methodology for determining dynamic range based upon visibility of unwanted blowouts or noise in fairly large print sizes. I don’t currently see any difference between the D850 and the Z7 that’s meaningful.

You may remember that I mentioned that there was some interesting new tags in the NEF files for the new Nikon mirrorless cameras. Indeed, they’re there. If you pull a Z7 NEF into even the current non-optimized version of ACR, you’re going to see that a number of settings are not where Adobe usually sets them. The easiest one to see is the sharpening settings: Nikon clearly is putting values into the NEF that ACR is picking up. What this all means and whether there’s been any real optimization of this remains to be seen (I still see the classic Nikon/Adobe overabundance of “orange” luminance, among other things; in other words, a Z7 NEF is very much like a D850 NEF in terms of the underlying data. I haven’t yet found a nuance that’s meaningfully different there, but this could just be that Adobe is processing Z7 images as if they were D850 images).

Overall: The Z7 arguably shoots and handles like a recent Nikon. That’s a good thing, as Nikon has a really good UI and ergonomic set at this point. And yes, handling the Z7 is better than handling a Sony A7Rm3 in my opinion. Clearly so. One morning with the temps in the low 30’s and me using gloves just put that ship to sailing. But I had that impression even before that. 

That said, nothing’s really changed in my base opinion: the D850 is still the best all-around ILC you can buy at this moment. The Z7 is somewhere below that. Exactly where I don’t know yet. It’s far too early in my use of the camera to make any call on that other than it’s not going to best the D850. 

text and images © 2019 Thom Hogan
portions Copyright 1999-2018 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #sansmirror