Canon 1D MKIII Autofocus- Interactive Review

Canon 1D MKIII Autofocus Problems- Interactive Review

Click here to read the entire article

Thanks, Mark and everyone. I have resisted becoming a main provider of content as I view this site as primarily about what the community has to say. I also did not have as much time before. I have enjoyed writing up the review.

Day 8 Added. Canon 1D MKIII Autofocus Study Awards now being offered.

i too...a thanks to drew for the testing....you must like to run !!

perhaps look at the tracking sensativity especially for the stationary shots in servo AF. i would curious to know how the CF that allows you to expand the number of AF points around the selected one helps/hurts this problem. i know with 300+mm lenses it grows even larger, up to 13pts for those, but only to 7points for shorter lenses.

perhaps since canon added more cross sensors and added them to the outside, all the cross sensors from all over the AF oval are competing with another and all those signals coming into the the AF computer...causing something to go weird. i've been with the 45pt AF since the 1v and they way to use it is to start your subject in the center, (on the cross sensors), lay on the shutter and let the other AF points track it. now it seems it doesnt know who should lead and who should follow.

i think what they were trying to do is to make the single line AF points more sensative and better trackers.

IMHO, seems like the AF system needs to be rethought alittle. perhaps all the AF points would be cross type just for sensativity, but ignore that if it causes problems, but it seems the AF points need a leader and a bunch of other follower AF points. its been this way since the 1n and its worked well over the years. and i say that as a loyal fan of canon of over a decade with them.

It looks like the camera will not be available until the end of this week. If it is available then. On the plus side, they now have two camera bodies in rental. Ideally, I may be able to test the "newer" body.

If you have your own MKIII and want some quick pocket cash (up to $150) please feel free to contribute to this report by doing your own study as outlined on the Day 8 entry.

Alternately, if anyone is within an hour of the Atlanta, GA area I will be happy to run some of the tests with your camera.

Hello Everybody!

This is my first ever post op PF. I've had a Mark III since 2 weeks and have noticed that there might be more problems than with just servo autofocus doesn't seen to be as impressive as with the Mark II or Mark IIN, both of which I also have.

I have something else to report which is just as disturbing is a weird hiccup with the exposure (aperature priority mode). I've had several instances where a whole series (of one moment over several seconds) grossly over exposed is, and I'm talking so overexposed that you can't see any detail anymore in most of the image. Yesterday I had the opposite where 4-5 consecutive exposures were grossly underexposed. I have never had anything like this with the Mark II/Mark IIN. And I still don't as I use the Mark IIN together with the Mark III. It's the Mark III that's giving me trouble. BTW, I own my own wedding photography business and have 17 years shooting experience with everything from 35mm to large format 4x5". I know how to meter.

And something else has to do with formatting memory cards in the Mark III. If I do this then I'm having problems using the cards in my Mark IIN. Is this normal? Do I need to keep separate sets of cards for each camera?? I will be contacting Canon Pro Service with this concern, but since I'm over here in The Netherlands I'm not expecting that they have very much up-to-date information. I'll try just the same, you never know. I thought I would just throw it out there to see if anyone else has had the same experience with this camera.

I was one of the first in my area to receive the 1DMkIII. I received my two MkIII's in late May. Here's a general summary of my experiences to date.

One of my first MkIII's had an out-of-the-box manufacturing defect. The vertical shutter release button would not trip the shutter. It would do half-press just fine, but no amount of pressure on the shutter release would cause the shutter to trip. My local dealer was able to expedite a replacement for me, and I had a brand new one within a week.

I do a mix of photojournalism, general assignment (architectural), and studio photography. In the past month I have shot with the MkIII in all three roles. Assignments have included two rounds of NCAA baseball in 90+ degree heat, house/apartment fires, press conferences, day and night homicide scenes, interior shots of a six-figure condo project, photographs of glass artists working in hotshops, glass sculptures in the studio, product photography in-studio, and various other general assignment shoots. Outside of the studio I use the Canon 70-200/2.8 IS, 24-70/2.8, 16-35/2.8 Mk I, and the 300/2.8 IS. In studio I end up using a mix of primes.

Since Rob first published his report on the AF issues, I've tried to find fault with the AF system in the MkIII. I have not done the specific tests he outlined, but I have shot real-world in similar conditions - most notably hot-weather baseball with the 300/2.8. I have not been able to find a flaw in the AF system. It works as designed and as specified.

BUT - the AF system in the MkIII is much tighter than the MkII/IIN. Specifically, the AF boxes in the viewfinder of the MkIII are exactly dead-on to what you see in the viewfinder. This is a lot different than the MkII. In the MkII, there's a lot of slop around the AF boxes. I always found that you could be a little bit sloppy with lining up your focus point on the box and still get a lock.

Sometimes this was a good thing...i.e. when tracking moving objects with long glass. Other times it was a bad thing...the slop has a tendency to cause the focus to not be exactly accurate when you're in an area with multiple potential focus points. For example...if you're shooting portraits with a 1.8 lens, the MkII will lock on a lot of different points along the curve of the face somewhere around the AF box. It's very unpredictable which point it will focus on, and with each bump of the shutter release you'll get something slightly different...which means you pretty much have to manual focus.

Not so with the MkIII. The MkIII AF locks to whatever is dead center of the AF box, period.

Is this a flaw? No. It's technically working exactly as it is designed. But it is a lot different from the MkII, and that difference will show up mostly with long glass.

Another thing that will exacerbate the situation is the MkIII AF is really really fast. If you drift even for a fraction of a second, the AF is off to the races locking to whatever is now in the box. The MkII was slower - small excursions were forgiven.

So with these two differences (tight AF boxes, very fast AF reaction time), I can see where people might think the AF is flawed - but it's not. It's technically working exactly as specified - and it's working the way the MkII *should* have worked. So one way of looking at it is the older MkII AF system is actually a "happy accident" - it is flawed in a favorable way for long-glass shooters. The MkIII is technically much more accurate - yet it doesn't "feel" as good to the long glass folks.

My personal experience? The MkIII was as good as I was shooting baseball, which was the closest approximation to the Rob Galbraith scenarios (hot temperatures, long glass). If I nailed the AF point on a good location with contrast and kept it there, I got in-focus shots. If I drifted, I got OOF shots. This performance is consistent with the examples posted on Rob's site. If you look at his examples, the OOF shots are ones where there is no contrast in the selected AF box...or the OOF shots are maybe one frame away from a no-contrast frame, indicating a focus hunt. You can't blame the camera for that.

In other long-glass situations I shot, the AF performed better than the MkII.

Once you get used to the sensitivity of this camera, it's a dream to use...even with long glass. It is VERY responsive and locks very quickly. Spot a hawk swooping down from the treeline? If you can put the AF box on it, you WILL get the shot.

I could go on and on about all of the other great things about this camera (insanely good high ISO, etc.), but this is an autofocus thread so I'll limit my comments to those issues.

I am not a Canon fanboy. I have not sold my MkII's yet (just haven't decided how many I want to keep for remotes). If I was having problems I would pull out the old MkII's in an instant. But the MkIII is a really good platform if you get your head around the idea that it's a NEW camera with a new feel. You have to get used to how THIS camera performs and stop trying to make it be a MkII. It's not - it's better.

If there was one suggestion I could make to Canon, it would be to add a real AF sensitivity custom function setting. Function III-2 only adjusts obstacle sensitivity...i.e. how quickly the system reacts when the AF detects a major excursion from the existing focus point. What would be nice is to have a way to adjust how often the AF system samples the focus point, so that minor drifts are more-or-less ignored. If they added this function, you could make the MkIII feel like a MkII and make everyone happy.

Here are some pictures I shot at Bowdoin College during the Region 1 soccer finals a week or so back. There are the unprocessed jpegs right from the card. If you download, all the exif data is there. I set the expansion to just the left and right points with the center point as the designated AF point. Release timing was set to focus priority for the first and subsequent frames. Lens was the 100-400L wide open, speed was 1/500 or faster for most of the sequence. The girl was dribbling diagonally an was about 2-6 yards outside the penalty area, while I was behind the goal line.


http:/www.firstgm.com/m3/