In the earlier portions of this process, I have:
After some preliminary screening I had decided to investigate several full-frame and several APS-C sensor cameras:
I've spent about 3 weeks now doing the things that are most important (other than counting the money hit) -- actually taking pictures with the various cameras.
I've also spent a lot of time reading reviews, just like everyone else would. Not to mention I took to heart some of the very good comments made in the previous posts (there are really good, non-Fanboy suggestions that are worth your while to look at).
In case you haven't realized it, I'm not reviewing these cameras and making recommendations, I'm just sharing how I figure out the best system for me and how I went about deciding what to buy. There won't be lots of technical information, pretty pictures, or even numerical system ratings in this post.
Let's face it, it doesn't matter how great the pictures are, if you don't find the camera comfortable to use, you probably won't be using it very long.
Fitting My Hands
The first thing I did was spend a day with each camera and a standard-range zoom, just to get a feel for them. Why putting a camera in my hands is vitally important became quickly apparent. The Nikon D5200 just doesn't fit my hands well. This surprised me a bit, since I'd been shooting an Olympus OM-D with grip for a while, and the D5200 is actually larger than that camera. I used it for a day, thinking I would adjust, but I didn't.
All of the other cameras on the list fit me comfortably and within an hour or two I was able to find the commonly needed buttons and wheels by feel.
Other Ergonomic Factors
I've shot so long with Canon and Nikon cameras that any comments I have on their layout are minor. While I'd shot a few times with the Alpha and Pentax systems, my experience with each has been fairly limited. Exploring features and menus was first order of the day for both systems.
One thing that I didn't like was the Sony joystick button. Compared to many other cameras I've shot with, the Sony button didn't have a lot of tactile feedback and I found myself having trouble selecting menus and other things with the joystick. I've talked to Alpha shooters who love the joystick so this probably has to do with my overly large and clumsy hands - but it didn't work well for me.
I had heard and read complaints about the Sony menu system. Perhaps because I have shot with NEX cameras a lot, which are similar, I didn't struggle with the menus, although I can't say I find them intuitive.
The Pentax system reaffirmed what I'd already thought during limited use. The menus and features must have been designed by photographers, not programmers, and they were completely intuitive for me. There were little things with the Pentax system that I absolutely loved, like the ability to map out hot or dead pixels from the menu. There are some mirrorless systems that let you do that, too, and I find it a bit insulting that the other SLRs require a trip to the factory for that, especially since hot pixels are something likely to occur outside of the warranty period.
A lot of people stress a camera's autofocus system a lot more than I do. I didn't find anything with the AF system of any of these cameras that made me want to cast it aside, but I wasn't using Servo mode for action shots, either. For a lot of you who do that kind of thing, the AF system would have been a more important part of their choice.
I felt the 5D Mk III was the most accurate in low light, but there wasn't a huge difference. The Nikon and new SLT Sony bodies were on a par for me. They may not be quite as good as the Canon 5D Mk III in low light, or maybe they were. All were perfectly fine for my needs.
I had heard some questions about the Pentax AF system, particularly in tungsten light, but the K-5 II bodies were supposed to be better. I stressed it in single-shot AF and poor lighting and found it was pretty accurate. AF speed varied a lot depending upon which lenses I was using, and was perhaps a bit slower than the others. Still what I would consider acceptable, though, no question.
Viewfinder and Live View
I was interested in taking a good look at the Sony EVFs and found they were quite nice and I really enjoyed some of the heads-up display features. The horizon level features are particularly well done and I've gotten quite attached to those using mirrorless systems as much as I have. (Again, I'm not doing a lot of rapid-fire shooting of fast moving subjects. The shortcomings of even a great EVF might be more important to you.)
I also use the LCD for live view focusing quite often. The D800's LCD interpolation artifacts during magnifion are distracting somtimes but didn't prevent me from getting accurate manual focus shots very often. I really liked the articulated LCDs on the Sony cameras during live-view focusing and I've always loved the focus peaking options. My eyesight isn't what it once was and that allows me to manually focus even through the viewfinder. I liked the Sony's focus-range feature, too, and I'd use that fairly frequently, I think.
I know there are some specification differences between the LCDs, but there was nothing I could really notice just using them. They were all excellent, but none of them are going to work for me in bright summer sunlight.
400mm Equivalent Images
As I'd written earlier, one of the biggest motivators for me making this choice was being unhappy with the telephoto images I was able to get from a Micro 4/3rds system, so a big part of my comparison was image quality of a 16" x 20" telephoto print.
I did some fairly appropriate and fairly simple tests using the hand-holdable lenses I would be interested in. I shot still subjects at 100 yards distance, resized the resulting image to print at 16" x 20", cropped the centers, and compared. I looked at 100% crops to see exactly how much detail was possibly visible, too. I'm sure there are 692 arguments people will make about the validity of this, but hey, they're my prints and it reproduces what I want to do.
Within each system I experimented with lenses a bit to see what I thought gave me the best images and ended up with this:
- Nikon: 300mm f/4 with a 1.4X teleconverter (better than the Sigma 50-500 OS, but it was close)
- Canon: 400mm f/5.6 (a tiny bit better than the 100-400, but I'd probably buy the 100-400 for the IS and zoom)
- Pentax: 300mm f/4 (450mm equivalent, but hey, give the little guy a break, OK?)
- Sony: Sigma 50-500 OS (better than the Sony 70-400, and yes I tried a couple of copies of the 70-400)
- Olympus: Panasonic 100-300, but I also tried the Olympus 75-300 and really couldn't tell a difference
For the Olympus zooms I even shot at 200mm, 250mm and 300mm, to see how much the increased magnification might translate into resolved detail, even realizing I wouldn't be able to zoom that far on most of what I shoot.
I'm not going to present you a review, I'm not a reviewer. But after looking at all the images myself and then asking 6 photographers who work here to look at them without knowing which image was which, some things were very apparent (all 7 people had the same conclusion).
The D800, when pixel peeping, resolved more detail than anything else and it wasn't even close. Resized to this print size (the D800 is actually downrezzing here) the difference became less apparent, but still 7 of 7 picked it as the sharpest image.
The Canon system was considered very close in a 16" x 20" print, but when pixel-peeping on a monitor the difference was more apparent (so it probably would be more apparent in a larger print, too).
The micro 4/3 images, even taken at 300mm (600mm equivalent) didn't resolve as much detail as the other systems. This is simply about the lenses. One interesting point was that when I zoomed either lens from 250-300mm I got a larger image, but really didn't resolve any more detail. In other words, I could have done that in Photoshop.
The Sony A99 with 70-400 at 400 was weaker than the other SLR systems. The A77, using the 70-400 at just under 300mm resolved better, although it still wasn't as strong as the others. I repeated the tests with the Sigma 50-500 OS at the same focal lengths and it was better, so again this seems to be about the lens.
The Pentax, despite only being 16 megapixels, surprised me positively. It didn't quite resolve as well as the Canon, but was very close. I think the excellent 300mm lens and no AA filter gave it a higher resolution than I expected. I will mention that I tried, and ruled out, the 60-250mm f/4 lens, which I thought would be most useful. It wasn't nearly as sharp as the 300mm prime.
So my conclusions were that moving to a new system would definitely be worthwhile for my 400mm shots. The D800 provided the best resolution, the Canon and Pentax systems were both acceptable, but the Sony systems lagged a bit. I know there are new lenses on the way for the Sony system, but as I mentioned earlier, I'm not considering anything that's not available now.
Macro is another area of importance. I found the Canon 100 f/2.8 IS L macro's image stabilization really did make a bit of a difference for me here, probably because it also stabilizes angular shake. I'm the first to admit I use it as a crutch and I should be setting up my tripod and macro rails more often . . . but hey, I like a crutch as much as the next guy.
The Nikon D800e with the 105 f/2.8 VR did show a bit more detail than the Canon did, but I couldn't cheat my handholding exposure quite as much with it. The Pentax didn't quite resolve as well as the others, but it was certainly acceptable, as was the OM-D system I'd been shooting with.
Dynamic Range and ISO
I had mentioned earlier that ISO 1600 performance was really important to me and ISO 3200 would be useful, but I rarely go past that. Shooting with these cameras it quickly became apparent than none of them particularly limited me at ISO 1600. I would be willing to print jpgs from the camera with all of them. Similarly, there was some noise that I'd want to work on in post, generally, at ISO 3200 but all were usable.
While there were on paper dynamic range differences, at my usual ISO (200 to 400) none of the cameras limited me greatly and I could rescue shadows and highlights reasonably with all of them. The Canon had the known problems with pulling out shadows, but I could pull back highlights beautifully.
From everything I've read it seems D800e files allow a bit more rescue leeway than the other cameras, and the Canon does have a problem pulling out shadows sometimes. But in what I shot, using my usual 'Zeiss, the jet-black dog' dynamic range tests, I'd be happy with any of the cameras.
The Wide End
There were no suprises here. Then D800 with the 14-24 f/2.8 AF-S was simply better than any other zoom combination. With prime lenses like the 14mm f/2.8 or 17mm TS-E, the Canon could hold it's own against the Nikon, but with the Canon 16-35 f/2.8 the difference was quickly apparent.
The Sony 16-35 f/2.8 lens was much like the Canon 16-35 f/2.8. OK, but clearly not competitive with the Nikon system. If I were to go with the Sony system, I'd really consider a RokiBowYang 14mm f/2.8 lens. It's sharper and wider than the zoom, way less expensive, and at 14mm manual focusing wouldn't really be an issue, especially with Sony's peaking filters. For that matter, I might well consider it an option with the other systems, too.
I didn't have access to a Sigma 10-20mm f/3.5 in Pentax mount so I didn't do a hand's on with this system and ultra-wides. The 15mm f/4 was quite nice, but I'd want something wider.
Where I Stand Now
Nikon D800e system: Meets or exceeds all of my needs.
Canon full-frame system (I'm still doing the 5D Mk III vs 6D tap dance): Meets all of my needs.
Pentax K5 IIs system: Meets my needs, but not quite as easily as the first two.
Sony Alpha system: While I love some features, and lust after the 135mm f/1.8 lens, it's a bit weaker at 400mm which is a major motivating factor for me.
At this point I'm leaning towards either the Canon or Nikon full-frame systems, although I'm still considering the Pentax. I like some features of the Alpha system, and some of the lenses are awesome, but for what I wanted most to do, it wasn't quite as good as the others.
I'm going to do some system exploration, as in what lenses I'd actually buy with each system (the initial cost lists were simply trying to put things in perspective from the m4/3 system). I'm also going to consider one of the earlier suggestions about keeping my Micro 4/3rds system and simply buying a dedicated camera - telephoto lens combination. I'll make my final decision in a few days, but there were and are plenty of options that will fill my needs perfectly.
I think the takeaway point is that the process I used made me look at a system (Pentax) that I hadn't really considered on the front end. The process also ruled out (for me) some cameras that I thought would be good choices on the front end. My goal with every major purchase is to avoid buyer's remorse and I think this has helped me do that.
I'll put a brief post up in a couple of days with my final choices. Uwe Steinmueller of The Digital Outback Photo has volunteered to write a counterpoint post, about what he would choose for his own system to publish at the same time. I think this will be a nice contrast and perspective: one gearhead hobby photographer and a full-time working professional photographer doing a similar exercise. We'll publish them simultaneously on Friday.