Focus Focus Focus


I’ll be posting reviews of several “fast focusing” mirrorless cameras over the next few weeks. As I’ve played with these cameras, watched others shoot with them, and looked at what gets posted on fora around the Internet about them, I’ve noticed a couple of things that need to be discussed. 

First, a lot of people are saying “it works for me.” They may even offer a shot or two to prove their contention. Frankly, a lot of those proof shots aren’t exactly nailed on focus, but even when they are, I’m noticing that it isn’t usually a long telephoto lens nor a complete burst sequence that we’re seeing, nor does it tend to be a low contrast subject or low light work. 

This is the old “good enough” argument in a different form. First, image quality got to a good enough plateau even with smartphones for the majority of people, and that put a damper on more sophisticated camera sales. Now we’re nearing the “good enough” stage for focus, and it very well may have the same impact long term. 

But “good enough” is not necessarily “DSLR level.” So be careful from drawing conclusions from casual observations. Heck, even some of the so-called “controlled tests” that are getting heralded don’t exactly show what the testers claim. Pitting a m4/3 camera against a full frame DSLR both shooting at f/4 means that the m4/3 camera has a deeper depth of field, for example. The full frame DSLR has to be incredibly more accurate to match or exceed the m4/3 results, especially as you move up in focal length. Yet it does ;~). 

As you all know, I took an X-T1, E-M1, and D7100 to the Galapagos in April and shot with them side-by-side. Generally, I had one of the mirrored cameras plus the D7100 with me on most shore excursions, so I was able to do real life scenario testing on some of the toughest subjects there are: small birds flying close to me and somewhat randomly. It’s a real stress test of focus systems and how well they handle constant and random change in subject position.

Here are a few of my general conclusions from that and other testing I’ve done:

  • The more focal length you’re using—especially if it’s fast glass—the more the DSLRs shine compared to the mirrorless cameras. Some of this simply may be that the right mirrorless lenses haven’t yet appeared, but with the E-M1, for example, it uses a contrast detection step after phase detection in C-AF, and I believe that’s where the small misses and more rare big misses are coming from. The more constant the motion speed and direction of the subject, the better the mirrorless cameras do in continuous autofocus with burst shooting. So the big birds like the albatross, which fly fast but at a relatively constant speed and direction, can be captured in flight with the mirrorless systems pretty well. The small birds that are constantly changing direction and speed didn’t fare nearly as well as with the DSLR, and even the DSLR would sometimes struggle to keep up with them.
  • Most of the mirrorless continuous autofocus systems shot in bursts have some discontinuity to the burst (e.g. they slow down), even when Release Priority is selected. This is doubly problematic, because one of the reasons to use bursts for action is to be able to pick “the moment.” But if the camera was set at 8 fps and only achieves a random 4 fps (X-T1, for example), you’re less likely to have the moment than you would with a DSLR that kept the frame rate where you set it (or Nikon 1, by the way). 
  • Be careful of interpreting initial acquisition speed with follow focus speed. They’re not the same thing. I’d say that virtually all of the mirrorless cameras are getting into the vicinity of or equalling initial focus acquisition speed of the DSLRs. This is great if you’re a one-shot person dealing with mostly static subjects, subjects that are moving at a constant distance from the camera, and even a moving subject that has high contrast in good light and the autofocus sensor clearly positioned on it. That indeed describes a lot of most folk’s shooting. Where things start to change is when you want to shoot more than one photo in sequence (either via a burst or by pressing the shutter release in quick succession), where the subject is moving randomly and at/away from you, and when you can’t keep a single autofocus sensor on the subject with reliability. Then, DSLRs (and the Nikon 1 with native lenses) start to show off their advantage. 

Mirrorless cameras have come a long way in a very short time in terms of focus performance. Could I live with some of the models that have appeared? Yes, I probably could. Indeed, a few are already better than the original autofocus film SLRs I used back in the late 80’s. But it’s wishful thinking at the moment to think that they’ve completely matched what a decent DSLR in skilled hands can do. 

I suspect, however, that we’re nearing the “good enough” mark for focus performance with some of these mirrorless cameras. The ones that particularly to mind in that respect are the Fujifilm X-T1, the Nikon 1 series, the Olympus E-M1, the Panasonic GH4, and the Sony A6000. These cameras very well may be good enough for the types of shooting you do. So take a good long look at them if they attract you in some way.   

Looking for gear-specific information? Check out our other Web sites:
DSLRS: | general:| Z System: | film SLR:

sansmirror: all text and original images © 2024 Thom Hogan
portions Copyright 1999-2023 Thom Hogan
All Rights Reserved — the contents of this site, including but not limited to its text, illustrations, and concepts, 
may not be utilized, directly or indirectly, to inform, train, or improve any artificial intelligence program or system.