Going out to shoot this morning I feel like I had a major breakthrough in understanding what is going on here. There's nothing in what I'm going to say here that I didn't already know as individual elements, but for a few different reasons I hadn't put it all together yet.
One this is that stopping down definitely makes a huge difference. I'd already sorted that out, but today I became convinced it's definitive. 6.3 is just not that sharp on my copy - BUT with everything else coming together I was able to figure out that just as Dinusaur said stopping down even just to 7.1 is making an enormous difference. I still wish it was as sharp wide open as my 200-500, but having to be at 7.1 vs having to be at 9 makes an enormous qualitative difference as to whether I'm willing to accept the limitation.
The more important factor, though, is the VR. I may go out to the pond this afternoon to try to confirm, but based on this morning I feel extremely confident that having the VR on at higher shutter speeds was doing a number on the sharpness. I know this is a common idea that has been discussed at length with Nikon VR, but you have to realize that my 200-500 did not have this problem. I had at one point tested it a fair bit and found that having it on or off at high shutter speeds didn't make a difference, so I was used to that and assumed that newer technology would perform at least as well in this regard, but this 180-600 absolutely seems to get much, much sharper when either shooting at low shutter speeds with VR on or at high speeds with it OFF. I spent a few hundred shots this morning of critters in the backyard bouncing around between 1/500 with VR set to sport vs off and at 1/2500 or so with VR set to sport vs off and the difference is dramatic and pretty consistent.
That doesn't mean I'm finding amazing sharpness to be consistent. It's definitely
much more consistent, but I am certainly getting a fair share of softer looking images - mainly when using high shutter speeds. With low shutter speeds, even with VR on, I expect a fair amount of the shots from my bursts to me off based on motion, either of the subject or even motion from me that the VR just didn't quite get. Yet even working at very high speeds like 1/3200 I am still seeing more that are off than I expect, with of course the key point that if I put VR on almost all of them are slightly off.
Here are some samples from after I figured this out. The critical thing to understand about these is that unlike every other photo I've posted in this thread, I have applied no sharpening to these at all beyond whatever LR imports with. For all of the others, I applied LR's sharpening slider (either globally or with masking), increased the texture and/or clarity using a mask, and then went into Photoshop to apply sharpening with a high pass layer. For the ones posted here, I have only cropped and applied basic exposure and color corrections but left anything related to sharpening alone. I suppose the one exception is that I ran a denoise on a few of the higher ISO ones which can affect sharpness a bit, but I consider this something different.
Something else I have to mention: in looking at these I decided to see what NX studio did with them as what I see in LR looks slightly worse than what I saw on the camera LCD and I would say that NX studio's rendering is noticeably sharper. Now I have always found NX studio's overall rendering of texture/sharpness to be slightly better than LR's, but this feels a bit more pronounced to me. Alas, I've spent many an hour trying to find a away to get LR to render in the same way that the cameras/NX studio does but can't get it to do so. Importing with camera profiles, for instance, is quite different from Nikon's own rendering for the Z8 and I haven't found any tweaks that quite match Nikon's. With LR's denoising and powerful editing features, though, I wind up feeling like I have to use LR anyways and give up on that lovely texture that Nikon produces.
View attachment 84782
View attachment 84795
View attachment 84783View attachment 84784
I will now post two copies of the same photo, one from NX studio and the other from as close as I can get it in LR. First is NX Studio.
View attachment 84785
Now LR:
View attachment 84786
I honestly don't know if the difference will come across in a forum sized image, but looking on my computer the Nikon version just has a certain something in the texture that really makes it look much more critically sharp. (By the way, to match NX studio's look as closely as possible, I have texture at +8 [this is from LR's camera settings profile] clarity at +9 [the profile wants +4], saturation at -10, and the red channel calibration saturation at -11).