(commentary)
So far I’ve documented over 50 differences between the D800 and D810. And that’s before we get to image quality, focus performance, battery performance, and a few other things that have to be measured and compared carefully.
I’ve finally gotten a chance to spend a few hours shooting with the camera, so I do have some initial impressions that you might be interested in. A full review will come in September.
My overall first impression: Nikon spent a lot of time attending to details and made an excellent camera better. Cheese did get moved, but not a lot and not cheese I eat much ;~). (I’m sure that many new-to-the-site readers won’t get that reference, so I’ll explain the joke: for almost 20 years now I’ve been writing about how Nikon constantly makes small changes of moving buttons, menu items, renaming things, and other annoying little things that we have to deal with as users. I’ve been loosely making references to the book Who Moved My Cheese [affiliate link] in this regard ever since it appeared in 1998.)
The interesting thing in the early “reviews” that are appearing on the Internet is that there appears to be split decision: minor update versus significant update. I’m not sure how you come to such conclusions so fast without doing a lot of shooting or having the right tools to evaluate the product, though. It doesn’t feel like a minor update, though. Way too many things changed that impact the high-end shooter: viewfinder, shutter, frame rate, buffer, focus system, and so on.
Which brings me to a further point: it isn’t good enough to say that the camera is a “minor update” or a “significant update.” For example, the D2h was a significant update of the D1h but that update failed to deliver much usefulness to shooters. The D3 was a significant update of the D2h and changed the way we shoot. There’s a difference between “significant changes” and “significant change to experience.” That’s what I’ll be looking for as I review the D810. Does it truly change the shooting experience enough to justify over the D800/D800E?
With that said, some preliminary comments about specifics:
Visible Body Changes
A number of people seem to be making highly positive comments about the hand grip, but frankly, I don’t feel a big change there. The slight notch for the middle finger of the right hand and the more pronounced thumb ridge on the back are the most defining elements for me in the changes. My hand position feels a little more locked in on the D810 than the D800, but I didn’t really have any issues with the previous grip.
That said, there’s a drawback to the new grip, too: the shutter release height is higher than before (the right camera top on the D810 is a bit taller than on the D800). People with big hands will like this more, while people with smaller hands will like that less, as it’ll feel like they’re reaching for the shutter release.
You can’t really win the “fits all hands” game. My sense is that both grips are acceptable but people will react to them slightly differently depending upon their hand size.
Some of the body changes are cosmetic/functional, others are cosmetic, and others are functional. The primary functional changes other than grip position are these:
- Three doors for the connectors on the side instead of one big door. You can actually have headphones and an external microphone plugged in (which means the top door is open) and keep most of your left hand grip on the camera intact. (I should point out that the orientation of the USB and HDMI connectors is still set up for engineering ease, not user convenience, though.) The three doors make the RRS L-bracket really tight: you can’t really open the top door with the bracket attached because the three doors are slightly taller overall than the single door on the D800.
- The metering has moved to a button on the top left cluster, which forced the bracketing button to move to a location above the Flash Options button. I’m less thrilled by this, as now I have a harder time locating the Flash Options button by feel. (And the Flash Options button has always had the problem of being buried behind the 10-pin connector if you’re using anything plugged into it, which continues on the D810.) What did we gain? A fourth metering option: highlight preservation. I’m not sure why this isn’t a programmable function (e.g. Fn button option, and it’s entirely unclear how the selection works from Nikon’s documentation; the icon used for it suggests it is spot metering, but it is and it isn't).
- Slight button position changes. The record video, exposure compensation, and Info buttons have moved ever so slightly. Not a big deal.
- Feel changes. The card slot door has textured material over it now, instead of bare plastic. But the big change is the AF Mode button on the front of the camera now has a pattern of nobs on it that make it easier to find. Finally!
- Additions. We get a QC (quiet continuous) mode on the frame advance dial, and an extra i button to match the most recent Nikon designs. Silly, really. Now when you press the INFO button twice, it turns on and then off. Before, it turned on and then went into quick settings mode. Now to get to quick settings mode you do it directly via the separate i button. Seems like one of those changes of mind that we get from Nikon in user interface every now and again. Before: button overload. Now: individual buttons. Tomorrow: something else.
- Technology improvements. The rear LCD changed a bit in specs. The big question when the camera was introduced was whether magnified Live View assessment improved. Well, yes, some. Enough to be clearly visible under most circumstances. Whether that’s enough to satisfy those that complained about the original, I’m not sure yet.
Overall, switching back and forth between a D800 and D810 in shooting isn’t going to cause any big aches and pains, which is good for those of you who were considering keeping your D800 as a backup when you get the new body.
Hidden Body Changes
Funny thing is, Nikon really hasn’t talked about either the Df or the D810 viewfinder changes. Like the Df, the D810 viewfinder seems brighter than the older FX viewfinders. They’ve also changed the color of the camera info LEDs at the bottom of the viewfinder view to white instead of yellow, which also seems brighter. I’m going to have to use the cameras more side by side to see if this is a change worth truly considering, but my initial impression is that it’s a nice, subtle, and welcome touch.
The big thing that’s hidden is the change in shutter. Nikon does make claims about the performance of the new shutter, but the thing that everyone will almost immediately notice is that the shutter is much more quiet and discrete. Less slappy and lower pitched in terms of the noise it produces. If you’re shooting events with the camera, this is a big deal. For me out in the woods, if a shutter fires and no one is around to hear it, did it make a noise? Well, for that there’s the electronic front curtain shutter to take out any shutter slap vibration using MUP, another nice change.
The slightly increased frame rate and the big increase in available memory (which creates a larger buffer) coupled with the EXPEED4 chip make the D810 much closer to the camera that a lot of folk wanted the D700 followup to be. Yes, I know some of you will still be complaining about file sizes, but I’m talking more about the ability to be a responsive camera in situations where you have to fire a lot of shots, especially in longer bursts. Like the D7100, which doesn’t quite match the D300s, the D810 doesn’t match the D4s, but it’s not what I’d characterize as a slow camera. The D810 didn’t quite cross over into sports/action camera land, but neither is it the slow D3x type of beast. The D800 wasn’t either, but the D810 moves clearly further away from the slow camera marker. I’ll have much more to say about this in September.
EXPEED4 also seems to have sped up the focus system a bit. The addition of the D4s’s Group Autofocus mode caught a lot of folks' eye, but I’d say the thing that should catch their eye about focus is that it’s more D4s-like now than before. This is a bit like the frame rate/buffer change. The D800 wasn’t exactly slow at focus. But if you went back and forth between a D800 and a D4 you’d notice that the D800 sometimes would lag the D4 a bit in finding focus and following complex movement. Now, that doesn’t seem to be the case. If there’s a lag, it’s small enough that I don’t perceive it and will have to discover it in extended use and measuring of performance.
Put together, the hidden changes really do push the D810 forward very nicely over the D800. I like that Nikon hasn’t seemed to have broken anything that wasn’t broken, but has goosed so many things in a positive direction that the D810 does feel a bit like a new camera again. It’s too early in my experience with the camera to substantiate this, but for two years I’ve written that the best all-around DSLR made is the Nikon D800. I think that’s going to change from D800 to D810 in my statements going forward.
Please note the “all around” in my statement. No DSLR excels at everything. The question really is how many things you might attempt for which a given model might give you sub-optimal results that cause you some grief. I try to spend time with every DSLR that comes out. Right now, the D800 (and soon D810, I think) is at the top of the class when it comes to being well-rounded. Based upon my recent experience with it, I’d probably put the Canon 1Dx right behind it. Others that are certainly not too far behind would be the D7100, D4s, and Canon 5DMk3 and 7D. But each of those others isn’t quite as “well rounded” in my opinion.
Put another way, it’s hard to imagine a better DSLR than the D810 without pushing into areas that just haven’t happened yet or are technologically not possible yet at “reasonable" prices. A real close second for the 1Dx, though it’s got a little more "special purpose" in it.
Update: I’ve heard from a few people having problems with Lexar 1000x cards. This appears to be the same problem that hit with the D4s and the D800 updates, and Lexar seems to be replacing them. If you look around on the net you’ll also find some folk saying that they’re having problems with Transcend and Kingston cards, as well. Do test your cards before trying to rely upon them.
Image Quality
As I wrote last week, it’s too early to tell a lot about D810 image quality, as the workflow tools aren’t all there yet. While I have software tools that can help me see what’s happening in the raw data, it doesn’t make any difference if the changes never make it into visibility using the mainstream products, such as the Adobe converters. Even Nikon’s own Capture NX-D doesn’t seem to be optimized fully for the D810. Either that or Nikon thinks that “optimum” has moved its mark slightly ;~).
That said, there are some things I can say about image quality so far.
ISO 64 seems like a real addition, though is really probably ISO 50 or so in reality. When I compare the D800 and D810 data with both set at ISO 100, though, any dynamic range improvement seems to disappear, though. I know that DxO just reported their tests and show a fairly even match of the two models from ISO 100 upwards, with most differences in their charts probably within their testing variation errors. That said, I don’t test cameras the same way DxO does, and it takes some trial and error testing on my part to come up with what I tend to think are optimal results. So wait for the review.
The break-over point in the way the sensor handles ISO seems to have changed from 1600 to 3200. Sensors can and do use slightly different approaches to boosting ISO these days. In the old days, gain was fixed and you simply multiplied the data received to get higher ISO values. These days, variable gain and other even more subtle techniques are applied at the sensor, so typically we see something slightly different in low ISO values than in data at very high ISO values. With the D800 something was changing at ISO 1600; with the D810 that appears to be ISO 3200.
What that might mean I’m not fully prepared to discuss at the moment. With the D800, for instance, I would tend to say that I’d shoot with it from base ISO to ISO 800 before picking up the D4s (or Df these days). Some of that was probably due to how data was handled at the sensor. Am I prepared to say the crossover point where I’d pick up the better high ISO camera is now ISO 1600? Not at all. I need to do a lot more testing here, and I need raw converters that are reasonably optimized to the camera and my workflow to see if the underlying difference changes how I shoot.
Still, that something changed in the high ISO data stream, coupled with the new lower base ISO, does show that this isn’t just the old D800 sensor accessed exactly the same way. Frankly, any improvement to the D800’s image quality would be unexpected but quite welcome. I’m pleased to see that Nikon continues to tinker with sensors in attempts to extract better data from them. Given that I’m an “optimal” shooter rather than a “good enough” shooter, such continued progression is welcome.
I can say that long exposures are a bit better on my D810 than D800E. On really long exposures (measured in minutes) and especially if the camera is hot or if you’re boosting ISO values, the D800 models tended to show some amp noise and substantial hot pixels. I can’t quantify it yet, but a quick check shows that my D810 is better behaved in this respect than my D800E. Certainly not perfect, but better.
I wish I could say the same about workflow of images, though. Given that I don’t like installing beta or release candidate software on my working computers, I’m going to be scrambling to figure out how to convert D810 images during August when I go offline for the month. (Yes, I do go pretty much completely offline. That’s the point of such breaks: if you start making exceptions, then you really don’t get any relief from the constant chatter on the Internet.)
This is one of the things wrong with the “proprietary” systems the Japanese camera makers tend to make (especially Canon and Nikon): because there’s nothing shared with other companies before announcement and very little shared after announcement, it just sets off a large wave of reverse engineering activity as everyone tries to figure things out that should have just been disseminated by the maker. It’s not as if the secret sauce stays secret very long, anyway. If Nikon doesn’t think Canon isn’t doing a full teardown and analysis of every new Nikon camera, then I want some of the saki they’re drinking. Likewise, if Nikon isn’t doing a full teardown and analysis of every new Canon camera, they’re drinking too much saki.
In short, the proprietary nature of things stays proprietary for about one to three months. You can’t change engineering course at a competitor that fast, so there’s really no reason to strive for that level of secretiveness. Had Nikon worked with the top three lens makers and the top five raw converter makers to make sure those companies had all the information they needed prior to the D810’s launch, Nikon would sell more D810’s faster. Why? Because right now I’d have to say that if your’e a raw shooter, you don’t need to be in a hurry to buy a D810, as the software support isn’t there yet. (Yes, I know Adobe has a “release candidate” that does, but it clearly looks to be sub-optimal and not yet fully profiled against the camera.) Likewise, we haven’t even begun to check whether third party lenses work or don’t work fully on the camera, so you’d be relying upon Internet fora to direct you there.
So I think that you can all probably wait until my review in September before worrying about spending you hard earned money on a D810. Oh, and with Photokina in September, that’ll give you some time to consider whether new technologies or products might change your mind.
If you really wish to ignore that last paragraph, support this site by purchasing from this advertiser ;~):
