*The expanded buffer is long overdue. The D3 is a speed camera, and many of those using it don't ever want to be slowed down. The original buffer was good, but with NEF shooting in particular, it was quite possible to get "didn't catch that action" due to the hiccup that occurs when the buffer fills and the frame rate drops from max to one or two images a second (depending upon card speed). My opinion: making the optional buffer expansion standard is the right move.
* Adding sensor cleaning plugs a hole. These days, all cameras should have some basic dust management capability, something that gets basic dust off the sensor automatically. The D3 was notoriously lacking that. But there's another issue here: the primary problem with sensors that D3 shooters have had is the shutter throwing lubricant at the sensor. The addition of dust shake off isn't going to address that issue. My opinion: welcome addition, but we'll still be cleaning sensors.
* Adding video makes the D3s join the fad club. There certainly is a smallish group of photographers that are demanding that they have some video capability in their still camera. I'm not one of them. And the emphasis is on "some video capability," because the D3s doesn't go very far with video. Indeed, we can say that the D3s is an exceptional still camera and a marginal video camera. Hmm. Anyone else see the problem with that conjunction? If the D3s is a pro tool, it should be a pro tool, period, no matter what it's doing at the moment. That means 1080P/24/30/60, manual audio recording capabilities, and headphone monitoring at the least. Indeed, I can think of about a dozen ways that the video side of the D3s still falls short of "pro." Nikon's positioning of video is somewhat revealing: press users can get a 1280x720 pixel still on the fly while shooting video. The basic claim is that newspapers don't need high resolution stills. Of course, shutter speeds might be a bit of a problem for action stills taken out of the video sequence. My opinion: a poor choice by Nikon. Either they should have added better video capabilities, or none at all. The mismatch of top-of-the-line still features with bottom-of-the-line video features is clear and wreaks of design-by-fad instead of design-by-use.
* Added in-camera cropping flexibility. The 1.2x choice is strange to me. Moreover, why only the single addition? Where's 16:9, 4:3, 1:1, 1.3x, and 2x? My opinion: nice to have something additional, but it really seems like Nikon's engineers have no idea what photographers might want. To make it clear: size flexibility (1x, 1.3x, 1.5x, 2x) and format flexibility (3:2, 4:3, 5:4, 1:1, 16:9). And we want both together (e.g. 1.3x + 16:9 crop). This feature has the smell of "maybe something close to Canon's 1.3x will attract some more Canon shooters" to it. That's bad product management to me. I wouldn't accept such a proposition from someone working from me in PM.
* Quiet mode makes it all the way from the lowest end camera to the top, finally. Shouldn't that progression have worked the other way round? Unfortunately, "quiet" doesn't actually mean quiet, it means "less noisy." My opinion: not nearly good enough.
* HI3 ISO value feels like just another stretch. The improved sensor dynamics are nice. But Nikon now thinks that they can produce a useful ISO 102,400. My opinion: Useful to whom? They used to think they could provide a useful ISO 64,800, and I think that was a stretch, too. The real benefit here is in the ISO 3200-12,800 range, I think. Basically, the best high ISO camera got better. But don't be misled by the super high numbers.
* The Live View button and other minor changes are part of what we've asked for. Both are areas where quite a bit of user comments have collected over time, especially the fact that putting self timer, mirror up, and live view all on the same dial meant that some desired combinations were impossible. My opinion: the changes just show that Nikon does sometimes hear user requests. It seems silly that there's no mechanism by which such desires can be directly communicated to Nikon, nor that Nikon ever acknowledges that they did something specifically to user requests. But that's a different story. We'll take real improvements any way we get them, and these are.
What's the bottom line on the D3s? A modest but competent update, with consumer quality video thrown in for good measure. There wasn't anything really wrong with the D3 for its primary target user, and now there's even less wrong with it. The D3s also manages to do something that nobody expected: push high ISO results to even higher levels of quality, almost enough to make D3 users jealous.
But Video enthusiasts now have strong evidence that Nikon doesn't fully understand their needs.