RobGalbraith.com
Go to advertiser website.
     Home
     RSS
     CF/SD/XQD
     About
     Contact
Go
Go to advertiser website.
 
An analysis of EOS-1D Mark III autofocus performance - Continued
December 10, 2007: Assessing the sub-mirror fix and v1.1.3 firmware
 
Canon is now shipping EOS-1D Mark III bodies with a revised sub-mirror mechanism that is meant to correct for poor AI Servo autofocus performance when temperatures are higher than normal. In addition, Canon sales companies worldwide are now or will very soon begin accepting customer cameras to have the sub-mirror fix installed. Canon has also released a new version of firmware that includes changes intended to improve the EOS-1D Mark III's autofocus performance.
 
This article update recaps the autofocus problems we've encountered with the EOS-1D Mark III going back to its May 2007 release, highlights some of the steps Canon USA and Canon in Japan have taken to try and solve the camera's autofocus difficulties and provides our assessment of the camera's AI Servo autofocus performance now, with the sub-mirror fix in place and firmware v1.1.3 installed.
 
lake_louise_action_thumb.jpg
Swiss Air: Canon EOS-1D Mark III (fixed sub-mirror, firmware v1.1.3) + EF 500mm f/4L IS w/Extender EF 1.4x II, ISO 2000, 1/2000, f/5.6, -1 AF Microadjustment (Photo by Rob Galbraith/Little Guy Media)
 
As with most of the bigger installments of this ongoing article, we're rolling out the information in FAQ format. This installment is an especially long one, however, so before we begin the question and answer, here's a summary of our assessment of EOS-1D Mark III autofocus now.
 
Summary
 
Without question, the sub-mirror fix improves EOS-1D Mark III AI Servo autofocus: it eliminates the camera’s big drop in autofocus accuracy when the mirror box gets warm or hot. Our experience is that it helps stabilize the autofocus performance so that the EOS-1D Mark III delivers about the same number of in-focus frames whether it's warm or cold.
 
Firmware v1.1.3 ups the number of in-focus frames the camera captures too and is an important part of improving the camera’s autofocus. It does increase the number of properly focused frames the camera snares when the light is bright and, compared to v1.1.0, noticeably improves the feel of the autofocus system.

But in our view, there’s more work to be done. In bright conditions, the EOS-1D Mark III still doesn’t match the EOS-1D Mark II N; at times the two are close, and at other times the EOS-1D Mark III's autofocus performance is below what we consider acceptable. The EOS-1D Mark III can autofocus much better outdoors on a warm, sunny day now than it could at the time of its introduction, though this is a reflection of how much room for improvement there was. There are also specific situations - mostly involving shooting indoors or in dimmer outdoor conditions - where its AI Servo autofocus is actually superior to the EOS-1D Mark II N.
 
Looking at all of what we've shot with the two camera models in recent weeks, however, the EOS-1D Mark II N is the one that has delivered the most in-focus frames. The culprit that continues to cramp the EOS-1D Mark III's autofocus style is bright sunlight.
 
An expanded explanation follows in the FAQ below.
 
FAQ
 
Q. Have you tested the autofocus of an EOS-1D Mark III with a fixed sub-mirror mechanism?
 
Yes. We've shot with three EOS-1D Mark IIIs with fixed sub-mirror mechanisms; two were selected by Canon in Japan for us to test, the third was obtained independently. Canon also made available a beta release of firmware containing focus tracking improvements that are incorporated in v1.1.3. We've had the opportunity to try the beta firmware in two of the three bodies, and the release version of 1.1.3 in two of the three bodies also. In addition, we’ve looked at several hundred files from other photographers shooting with EOS-1D Mark IIIs that had either the beta or v1.1.3 firmware installed and properly-functioning sub-mirrors.
 
Q. What problems were you hoping to see reduced or eliminated by Canon's change to the camera's sub-mirror mechanism and the new firmware?
 
In the first installment of this article, back on June 19, we identified four distinct autofocus problems when shooting in what we referred to as "certain conditions," where those conditions were a combination of bright sunlight and warm temperatures. Here's what we said:
 
  1. Under certain conditions, the EOS-1D Mark III has difficulty acquiring focus initially. In a multi-frame burst, the camera will sometimes shoot three to five frames before a moving subject comes into focus, and occasionally a moving subject will not actually snap into focus before the burst is completed.

  2. Under certain conditions, the camera is unable to properly track a moving subject. We've shot numerous sequences of 20+ frames where no more than five or six frames are in focus, even when the AF point has been on the subject throughout.

  3. Focus can shift slightly but constantly at times when the subject isn't moving. Under certain conditions, the subject may not actually come into focus through a sequence of frames, even though the point of focus can be seen to be shifting throughout the sequence. This is true whether the camera is set to AI Servo and focus is active throughout the sequence, or when it's set to One Shot and focus is activated between each frame.

  4. When tracking a subject that's moving somewhat erratically, the camera is far too quick to shift focus elsewhere - to the background or, with a field sport like soccer, to a player passing through in the foreground. With the first three problems, autofocus settings changes don't make things better or worse. With this problem, Custom Function III-2, AI Servo Tracking Sensitivity, does have an impact. But regardless of how this Custom Function is set, it's not possible to make the camera's tracking sensitivity be right.
 
Firmware releases up to and including v1.1.1 beta (the version of firmware that Canon seeded to certain photographers starting in late August but never publicly released) went some distance towards tackling problem number 4 above. Through firmware changes alone, the camera's autofocus began to feel less twitchy and overeager to shift focus, even if there was only a modest increase in the number of properly focused frames captured in hot, bright conditions.
 
The sub-mirror change, in conjunction with firmware v1.1.3 (which has been revised further since v1.1.1's private release), needed to address problems one through three especially, and that's what we set out to evaluate.
 
Q. Canon says that the change to the sub-mirror mechanism is meant to improve AI Servo autofocus in higher than normal temperatures. It's not very warm in Calgary, Canada this time of year, so how you were able to test Canon's fix where you are?
 
First, our interpretation of what Canon means by higher than normal temperatures is not the ambient temperature per se, but rather the temperature inside the camera's mirror box. There are several things that can influence this, including of course the ambient temperature, but also whether it's sunny, since the camera's black exterior will absorb the sun's rays and warm the interior, even if the ambient temperature is not hot.
 
Second, our overall EOS-1D Mark III experience, prior to the revised sub-mirror and firmware v.1.1.3, is that bright shooting conditions alone could make problems one through three appear. If it was a really warm day, that seemed to not so much introduce the problems as exacerbate them.
 
Third, Canon’s solution to the camera’s autofocus problems is comprised of both a hardware change (the sub-mirror fix) and tweaks to the AI Servo autofocus algorithm (in firmware v1.1.3). The former is all about stabilizing the autofocus when things get hot, the latter is about bumping up the number of in-focus frames the camera delivers, regardless of whether it’s hot or not.
 
Having said that, we've shot with the EOS-1D Mark III and its revised sub-mirror in temperatures ranging from as cold as about -28° Celsius (-18° Fahrenheit) to as warm as about 36° Celsius (97° Fahrenheit) and full sun. Calgary and the nearby Rocky Mountains provided the chilly weather. To find fall temperatures that toasty, we headed to Phoenix, Arizona.
 
Q. You went to Phoenix, Arizona?
 
Yes, three times. All three trips were in conjunction with Canon, to help assess possible EOS-1D Mark III autofocus solutions. This would be a good time to talk about some of what's been happening behind the scenes involving us and Canon, in the months since Canon USA representatives came to Calgary to look in on our EOS-1D Mark III testing in late July. Here's a brief chronology:
 
July 31 - August 1, 2007 Canon USA sent two representatives to observe our testing of the EOS-1D Mark III with firmware v1.1.0 installed, using bodies and supertelephoto lenses that had been checked and calibrated by Canon USA technicians. They returned to the company's corporate headquarters in Lake Success, New York with a hard drive full of photos showing the camera's poor AI Servo autofocus in bright, moderately warm conditions. The files were passed on to Canon’s global headquarters in Tokyo.
 
At the time, we saw Canon USA's decision to become centrally involved as important in the eventual development of a solution: "[t]he good news is that Canon USA is now tackling this head-on, by dispatching staff to view our testing, by sending carefully-calibrated gear for us to shoot with and by using their influence with Japan to kick-start the investigative process."
 
September 12 - 13, 2007 In early September, Canon USA got in touch to say there was a possible solution to try, something that superseded the v1.1.1 beta firmware from late August. With uncooperative weather on the horizon here in Calgary, the decision was made to head where sun and heat were more certain. That culminated in representatives from both Canon USA and Canon's Tokyo headquarters meeting up with me in Phoenix for two days of shooting athletes in blistering 42° Celsius (108° Fahrenheit) sunshine.
 
The files were taken to Tokyo for analysis, but it was clear even before leaving Phoenix that the possible solution (which wasn't the sub-mirror fix, that was yet to come) was not the solution. It was clear, however, that EOS-1D Mark III autofocus had the full attention of Canon's engineers.
 
October 24 - 25, 2007 Simultaneous with Canon USA giving its staff the green light to talk about the upcoming sub-mirror repair program, a return visit to Phoenix was scheduled. There, I met up with a two-man contingent from Canon USA to try - over back-to-back sunny days and 34-36° Celsius (93-97° Fahrenheit) temperatures - a pair of EOS-1D Mark IIIs equipped with the modified sub-mirror. The regimen included shooting with both bodies plus a combination of v1.1.0 and a beta release of firmware containing autofocus tweaks that would ultimately emerge in v1.1.3. The files from this edition of the Phoenix testing were also sent to Tokyo for analysis.
 
I returned from Phoenix with one of the two bodies, with beta firmware loaded and, separate from Canon, we got a newly-manufactured EOS-1D Mark III with the sub-mirror fix and firmware 1.1.0. Since the second Phoenix trip, we’ve shot with these fixed cameras in Calgary, both outdoors in sunny but cool-to-cold weather, and indoors.
 
November 26 - 28, 2007 To test out the shipping version of firmware v1.1.3, I returned to Phoenix, along with one Canon USA staffer. The temperatures were more moderate, in the 20-24° Celsius (68-75° Fahrenheit) range, and the sun cooperated just enough to allow a full slate of bright light photography. The photos have probably already arrived in Tokyo for engineers at Canon’s camera design centre to see.
 
runner_test1_thumb.jpg
Fast Times: Frame from a series of runner tests, shot on November 28, 2007 in Phoenix with a Canon EOS-1D Mark III (fixed sub-mirror, firmware v1.1.3) + EF 300mm f/2.8L IS, ISO 200, 1/8000, f/2.8, -1 AF Microadjustment (Photo by Rob Galbraith/Little Guy Media)
 
Note: At Canon USA's request, we haven't included the names of the various Canon people we teamed up with in Phoenix, as several of them are in technical roles and prefer to stay out of the public eye.
 
Q. What is the sub-mirror fix designed to do? What's changed in firmware v1.1.3?
 
The sub-mirror fix is meant to ensure that light is being passed down to the autofocus sensor in a consistent manner, even when the mirror box warms up. About the specifics of what’s changed in the sub-mirror mechanism when a camera is sent in for repair, Canon USA isn’t revealing too much. Here's what we know for sure:
  • In all of the repairs, a part in the sub-mirror mechanism is swapped.
  • Once the repair is completed, a mechanical alignment check is performed. It includes a verification that the angle of the sub-mirror, the position of the image sensor and the position of the focusing screen are all correct. This is meant to eliminate misalignment as a source of autofocus (and manual focus) error.

  • That's followed by a check of the body’s autofocus accuracy, using Canon’s standard service department procedure that involves taking photos with a known, good lens, then analysing those photos with proprietary software that automatically determines whether focus calibration is required.
It’s possible to check and adjust the calibration of all 45 of the EOS-1D Mark III's autofocus points; the centre AF point and 18 others that are cross-type can be checked and adjusted for both their horizontal line and vertical line autofocus accuracy. We don’t know, however, if Canon’s autofocus check at the completion of the sub-mirror repair will look at one, some or all 45 points, or whether both the horizontal and vertical orientations of cross-type points are checked.
 
We also don’t know how long the typical sub-mirror fix is going to take, though it’s clear that the procedure is not a while-you-wait affair. Rounding out the list of repair-related things we can’t yet provide any information about is how Canon will be marking cameras to indicate that they’ve been repaired. Answers to some of these questions are probably going to have to come from photographers who have had the repair done. In an apparent goodwill gesture to owners of the EOS-1D Mark III who send it in for the fix, Canon USA is extending the camera's warranty by one year from the date the repair is completed. We can't say if the warranty is being extended in other regions also.
 
Firmware v1.1.3 includes several changes, including two that are autofocus-related. The autofocus changes are intended to improve tracking accuracy in bright conditions and when photographing low-contrast subjects.
 
mens_basketball2_thumb.jpg
Going Up: Canon EOS-1D Mark III (fixed sub-mirror, firmware v1.1.3 beta) + EF 300mm f/2.8L IS, ISO 200, 1/320, f/7.1, -1 AF Microadjustment (Photo by Rob Galbraith/Little Guy Media)
 
Q. Does the sub-mirror fix correct the camera's autofocus problems?
 
Without question, the sub-mirror fix improves EOS-1D Mark III AI Servo autofocus: it eliminates the camera’s big drop in autofocus accuracy when the camera’s mirror box gets warm or hot. If you have a camera that was manufactured before early October 2007 and is therefore eligible for the sub-mirror fix, we strongly encourage you to have the repair done. Our experience is that it helps stabilize the autofocus performance so that the camera delivers about the same number of properly focused frames whether it's warm or cold.
 
Firmware v1.1.3 ups the number of in-focus frames the camera captures too and is an important part of improving the camera’s autofocus. It does increase the number of in-focus frames the camera snares when the light is bright and, compared to v1.1.0, noticeably improves the feel of the autofocus system.
 
Canon had also improved the feel of the autofocus from v1.0.8 to v1.1.0, but with v1.1.3 they've finished the job. With AI Servo Tracking Sensitivity set to the middle (normal) position, the jumpy, premature shifting of focus to a foreground or background object is all but gone. In fact, we now slightly prefer the EOS-1D Mark III to the EOS-1D Mark II N in this regard, as it seems to make good decisions about when to keep tracking a given subject or switch to a new subject that's passing through or has parked in front of the active autofocus point. In other words, Canon has effectively eliminated what we referred to as problem four back on June 19.
 
On the plus side, then, the EOS-1D Mark III's AI Servo autofocus feels better and its performance is better, dramatically so at times when shooting on hot, sunny days, thanks to the sub-mirror fix and firmware v1.1.3. Compared to the results we achieved in both testing and in real-world shooting back in May-August, with bodies that predated the sub-mirror fix and had firmware v1.0.8 through v1.1.0 installed, the biggest difference we notice is the elimination of numerous severely backfocused frames in extended sequences. The improvements are more than just that, however: the entire character of the autofocus of moving subjects has been improved as a result of Canon’s efforts.
 
But in our view, there’s more work to be done. In bright conditions, the EOS-1D Mark III still doesn’t match the EOS-1D Mark II N; at times the two are close, and at other times the EOS-1D Mark III's autofocus performance is below what we consider acceptable. The EOS-1D Mark III can autofocus much better outdoors on a warm, sunny day now than it could at the time of its introduction, though this is a reflection of how much room for improvement there was. There are also specific situations - mostly involving shooting indoors or in dimmer outdoor conditions - where its AI Servo autofocus is actually superior to the EOS-1D Mark II N.
 
Looking at all of what we've shot with the two camera models in recent weeks, however, the EOS-1D Mark II N is the one that has delivered the most in-focus frames. The culprit that continues to cramp the EOS-1D Mark III's autofocus style is bright sunlight.
 
As a result, the EOS-1D Mark III’s ability to pick up focus on a moving subject, then track that subject through a sequence, is still not what we believe it should be. As noted, it's also not up to the level of the EOS-1D Mark II N when making pictures in bright light especially. And in any light - bright, overcast or indoors - the camera's AI Servo focus can still shift about too much when the subject is stationary (though it seems more stable than before).
 
Canon has made considerable progress with the EOS-1D Mark III's autofocus but still hasn't fully solved what we referred to as problems one, two and three in the June 19th section of this article (a description of problems one to four is also towards the top of this page).
 
For us to say that the EOS-1D Mark III autofocus problem is history, then, we need to see the camera match or exceed the AI Servo autofocus performance of the EOS-1D Mark II N in bright sunny light in particular. Since we - like many of you - spend a lot of time shooting outdoors in the sunshine. What the pictures reveal is that the EOS-1D Mark III doesn’t yet do that, even if its autofocus is much more capable now than prior to the sub-mirror fix and firmware v1.1.3.
 
Q. Are there any settings that can help squeeze the best possible AI Servo autofocus performance out of the EOS-1D Mark III?
 
Yes, one: C. FnIII-7 [AF Microadjustment]. This setting's primary purpose is to enable the photographer to compensate for focus calibration errors in the body or combination of the body and attached lens. Its hidden purpose is to improve the EOS-1D Mark III's tracking performance. The Phoenix testing with Canon included shooting sequences with an AF Microadjustment amount of 0, -1 or -5 applied (via the [Same amount for all lenses] option). The inescapable conclusion was that the -1 sequences contained fewer frames that were completely out of focus than the 0 sequences (the -5 sequences were uniformly poorer all around and this amount was excluded from subsequent testing).
 
If we'd rattled off only a handful of 0 to -1 comparisons, this result could have been chalked up to a testing fluke. But because we shot and compared dozens of sequences and could see that the same slightly better performance in the -1 frames is consistently there, we've gone on to shoot almost all real-world peak action with -1 set, and recommend that you try doing the same.
 
To be clear, dialing in -1 for AF Microadjustment in this case is not about compensating for focus calibration error. The bodies and lenses we used to test the impact of AF Microadjustment were all calibrated by Canon USA or Canon in Japan.
 
Nor is it about making all photos be frontfocused slightly (minus side adjustments give the focus a slight frontward push). Instead, it's about making more properly focused or very nearly properly focused frames when the camera is tracking. The weirdest part of all is that the most notable difference between 0 and -1 sequences is that the -1 sequences show fewer instances of the camera focusing too far in front of the moving subject.
 
The tracking improvements brought about by this AF Microadjustment change are slight, but they sure seem to be real. We can't explain why this is so, and we also can't take credit for the idea: it came straight from Canon, as part of the testing regimen specified starting with the October Phoenix trip.
 
The suggestion to dial in a -1 AF Microadjustment raises a few questions:
  • What do you do if you already have this Custom Function set to compensate for focus calibration error with one or more of your lenses?

  • Should the -1 AF Microadjustment be set only when shooting things that move?

  • When photographing static subjects, but with the autofocus mode set to AI Servo, is it better to have the camera set to 0 or -1?

  • Do those who use One Shot autofocus mode choose 0 or -1?

  • In the future, will Canon somehow insert this AF Microadjustment trick directly into the EOS-1D Mark III's autofocus algorithm - perhaps by applying -1 AF Microadjustment compensation automatically when the camera senses subject motion?
The answer to all of these questions is the same: we don't know. What we've chosen to do for now is set -1 when shooting action, then switch over to 0 when shooting portraits. We can see that -1 is the way to go for sports, but we haven't done enough comparing of the difference between 0 and -1 when shooting portraits and other static things to know which setting is better or if in fact it makes any real world difference at all.
 
Q. It sounds like the EOS-1D Mark III's autofocus performance is somewhat variable, depending on the shooting conditions and perhaps what you're shooting. Can you give some examples of where it works well for you, and where it doesn't?
 
If you've been following this article since the first installment back in June, you'll know that we've been judging EOS-1D Mark III autofocus based on the shooting of mostly summer sports like soccer, football, rugby, show jumping and, more recently, indoor sports like basketball, volleyball and hockey. We've also recently covered skiing.
 
These various sports aren't all that myself and site co-editor Mike Sturk shoot, and in my case, sports are actually a small part of the income I derive from photography. But sports are the thing that test the mettle of a camera's autofocus better than anything else we do, and sports are where solid autofocus is an important part of making the best possible peak action pictures. So it has and continues to be the main environment in which we assess EOS-1D Mark III autofocus.
 
With that in mind, below are a few situations in which we've shot with the EOS-1D Mark II N and the EOS-1D Mark III, the latter of course containing the fixed sub-mirror and either the beta firmware or v1.1.3. For each situation we've graded the cameras' autofocus capabilities, from A+ to F, if applicable. We opted to go with a grading system for three reasons:
  • To give you an idea of how we think the EOS-1D Mark III compares to the EOS-1D Mark II N for that specific application. For example, a grade of A+ for one and a grade of B for the other means that one camera's autofocus is noticeably better.

  • To rate the cameras in absolute terms.

For example, an A rating means we've found the camera to be an excellent choice for that application because it can deliver many, many properly focused frames and almost all of the key moments are perfectly focused or nearly so.

A B rating means the camera is usable, but it's not ideal - the camera captures some out of focus frames when it ought not to, as well as a number of frames that aren't perfectly focused but they are close enough to still be usable for all but bigger enlargements or severe crops.

A C rating means the camera is usable, but barely - the camera is shooting more than a few frames that are out of focus, and more than a few key moments aren't usable from the take because they are simply too out of focus.

A D rating means we'd avoid using the camera if possible, though it does make some properly focused pictures.

An F rating means we'd avoid using the camera altogether because the focus is just really, really bad.

  • To avoid the use of percentages. While no rating system is perfect, since there is an element of subjectivity in deciding what's perfectly in focus, what's usably in focus and so on, we think percentage ratings are about the worst of all ratings options. That's because they bring the illusion of precision and objectivity to the assessment process, which in turn leads to apples-and-oranges comparisons of one photographer's assessment with another's. When, as often as not, two shooters will have a different idea of what constitutes a well-focused picture and their ratings will be skewed accordingly.
The grading system, then, is all about us applying our subjective assessment of how the EOS-1D Mark III's autofocus performs, and how it stacks up vs the EOS-1D Mark II N. That assessment is influenced by the fact that a growing part of the way I personally make money from photography is through the selling of custom prints. And not stacks of 5 x 7s; it's rare that I make a print smaller than about 22-23 inches wide. This means that perfectly focused or very nearly perfectly focused pictures are not a luxury but a necessity. An autofocus system delivering less than a full B in our own grading system is one we'd prefer not to use for that purpose.
 
If, by comparison, your photographs typically end up three or four columns wide in a newspaper, or a few hundred pixels wide on the web, you've got more wiggle room, and therefore a B-, C or even a C- autofocus system might prove acceptable to you.
 
We've tried to bring as much accuracy as possible to this process, by looking at every one of the tens of thousands of frames we've shot during the last several months in which we've been writing about EOS-1D Mark III autofocus. But it's important to remember that our view on this is through our own prism of what we need a camera to do and what we expect a camera to do. The collection of downloadable sequences at the end of this page are there to help you decide whether the EOS-1D Mark III is performing well enough to meet your expectations (assuming also that what you shoot is somewhat similar to what we shoot).
 
The EOS-1D Mark III's autofocus is no longer the poor bright light and warm day performer it was before the sub-mirror fix and firmware changes culminating in v1.1.3. At the same time, the camera hasn't been transformed into an autofocus star by Canon's changes, so the onus is squarely on you to read what we've written, to look at some of the full-resolution sequences and come to your own conclusion about whether its autofocus has improved enough to meet your standards of acceptability.
 
If you never shoot pictures outdoors in bright sunlight, your overall opinion of the camera is likely to be rosier than ours. Conversely, if you shoot in nothing but bright sunlight, your view of the EOS-1D Mark III's autofocus might be more negative.
 
While we've managed a fair bit of indoor and outdoor photography with the EOS-1D Mark III since taking possession of fixed bodies and the new firmware in recent weeks, because so much of the testing has been done at actual sports events, and some seasonal sports are already done for the year, the list below includes only what we feel we've shot enough of to confidently offer an opinion. And, since the camera's autofocus has been tripped up primarily by bright light and warm temperatures in the past, our efforts when outdoors have been concentrated on shooting the camera in the sunshine.
 
Our EOS-1D Mark III and EOS-1D Mark II N report card:
 
Runner tests, bright sunlight (EOS-1D Mark III: B-, EOS-1D Mark II N: A) All three Phoenix trips have included the shooting of multiple sequences of an athlete running straight towards the camera in bright sunlight and warm or hot temperatures. Because the subject is moving at a fairly constant speed, there are no refs or other obstructions passing through the frame and with a little practice it's possible keep the active AF point trained on the athlete's torso throughout, this is the sort of thing that a pro camera's autofocus should be able to nail.
 
The EOS-1D Mark II N comes closest to doing so, delivering multiple sequences where just about every frame is properly focused, and gets a solid A for its efforts. The EOS-1D Mark III is a B-; its autofocus is somewhat more successful when the runner is moving reasonably fast and is filling the frame, and somewhat less so when the runner is moving at a jogging pace or is looser in the frame.
 
Towards the bottom of this page there are full-resolution sequences shot with both cameras available for download. The sequences show that while there is some run-to-run variation, at their best we think each camera operates at the grade level we've assigned here. In other words, the EOS-1D Mark II N performs noticeably better in this test, and is just plain good.
 
All Phoenix runner testing was done with focus-calibrated EF 300mm f/2.8L IS lenses.
 
Incoming: Click to play a brief movie showing a runner sequence being photographed during the late-October trip to Phoenix. The photographer in the foreground is Rob Galbraith (Movie courtesy Canon USA)
 
Soccer player tests, bright sunlight (EOS-1D Mark III: B-, EOS-1D Mark II N: A) All three Phoenix trips have also incorporated shooting multiple sequences of a soccer player repeating a drill in warm to hot temperatures (plus we've shot the same drill at home in bright but frigid conditions).
 
The rate of speed is less constant, but this is still a test that should be easy for a pro camera to handle. The EOS-1D Mark II N once again earns an A; as long as the autofocus point isn't jumping around the athlete's body too much, this camera gets most frames in focus in this test. The EOS-1D Mark III's tendency to imagine or overestimate the degree of forward motion gets it a B- overall.
 
The downloadable sequences section has example batches from both cameras.
 
All Phoenix soccer player testing was done with focus-calibrated EF 300mm f/2.8L IS lenses.
 
Soccer, bright sunlight (EOS-1D Mark III: B-, EOS-1D Mark II N: A-) The EOS-1D Mark II N is the best camera we've ever used to shoot the game of soccer, but there's room for improvement too: it struggles a bit with tracking on white uniforms in bright light, and it sometimes misfocuses slightly when players are jumping straight up (it sees forward motion that isn't there) or when players are running quickly across the frame. Despite quirks like these, it does turn in an impressively high number of properly focused frames overall, and for that it gets an A-.
 
The EOS-1D Mark III, by comparison, captures some frames that are perfectly focused and quite a few more that fall just short of the mark but are usable on the web or for smaller or lower-quality print uses.
 
This web gallery (the link will open and resize a new browser window) contains 46 soccer game photos shot with fixed EOS-1D Mark III bodies and a beta version of what would become firmware v1.1.3 (the photos were taken during the October Phoenix trip). There are a number of frames whose focus is slightly out, and yet the gallery images look fine for the most part. But there's no getting around the fact that the EOS-1D Mark III's take of properly focused soccer frames trails that of the EOS-1D Mark II N.
 
The EOS-1D Mark III's grade is B-.
 
The downloadable sequences section includes five folders of pictures from the EOS-1D Mark III, each showing characteristics of the camera's autofocus, including the fact that it hangs onto the focus much better now when the AF point starts to drift off the subject, but also that it tends to overestimate frontward subject speed when the light is bright. It is, however, very difficult to try and sum up the EOS-1D Mark III's autofocus performance for a field sport like soccer by showing just a few dozen frames. The downloadable sequences will give you some idea of how the camera performs, but not the whole picture.
 
Lenses used to shoot soccer were the EF 70-200mm f/2.8L IS, EF 300mm f/2.8L IS and EF 500mm f/4L IS, all focus-calibrated.
 
Competitive skiing, finish area, overcast light (EOS-1D Mark III: C-, EOS-1D Mark II N: C) This one is a surprise, because we had expected that the EOS-1D Mark III - whose autofocus generally performs better when it's quite cloudy - would do pretty well and probably eclipse the EOS-1D Mark II N.

lake_louise_jubo_thumb.jpg
Yippee: Canon EOS-1D Mark II (firmware v1.2.6) + EF 300mm f/2.8L, ISO 400, 1/8000, f/4.5 (Photo by Mike Sturk/EPA)

Both cameras performed similarly, however: each had some trouble tracking skiers slowing down as they got deep into the finish area, and more trouble holding focus on the skiers when they stopped to see their result, wave and occasionally celebrate. Shooting side-by-side with Mike Sturk, it was clear that the EOS-1D Mark III produced more first frames that were completely out of focus, and its shifting of focus on stationery skiers was more pronounced.
 
The EOS-1D Mark II N rates a C, and the EOS-1D Mark III a C-.
 
We shot on the course as well, and looked at hundreds of files from other photographers too, but this hasn't added up to enough experience to assign a grade to the EOS-1D Mark III for downhill action (and we didn't shoot at all with the EOS-1D Mark II N for this; both it and an EOS-1D Mark II stayed in the finish area).
 
That said, in overcast light, I did shoot several four frame sequences where the EOS-1D Mark III's focus was just about perfect on the first frame of the skier coming around the flag, and it got the focus close for the remaining three.
 
All finish area shooting was done with focus-calibrated EF 300mm f/2.8L IS lenses. On the course, a focus-calibrated EF 500mm f/4L IS was used (with Extender EF 1.4x II attached).
 
Pole vaulting, backlit in fading light (EOS-1D Mark III: A, EOS-1D Mark II N: B-) The EOS-1D Mark II N rates a B- when the subjects are predominantly backlit. The EOS-1D Mark III is an A- thanks to its ability to snap the subject into focus quickly at or just before the peak of the vault routine, even if the run up shows a mix of crisp and blurry frames. This is true when things are frontlit, but performance climbs to an A when the vaulters are predominantly backlit and the light is fading overall. The EOS-1D Mark III is happiest outdoors when the light levels are down from full brightness by at least several stops.
 
Though the EOS-1D Mark III's autofocus generally doesn't match the EOS-1D Mark II N in bright sunlight, when the sun begins to fade at the end of the day, and the light is behind or to the side of the subject, the tables turn. Then, the EOS-1D Mark III does a very good job of picking up focus quickly and accurately, and does a fairly good job of tracking the subject through the run up to the vault too.
 
The downloadable sequences section includes two short bursts from a pole vault practice, shot with the EOS-1D Mark III.
 
The lens used was a focus-calibrated EF 300mm f/2.8L IS.
 
Strobed volleyball (EOS-1D Mark III: A+, EOS-1D Mark II N: B) This sport is all about the speed and accuracy of focus acquisition before the first frame is captured. Or, in our case when shooting with strobes that recycle about every 0.6 seconds, the first and only frame - there is no sequence shooting and almost no tracking per se.
 
For strobed volleyball, the EOS-1D Mark III earns a grade of A+, because when we get the AF point directed at the right player and press the shutter button at the right time (which is no small feat in this fast-moving sport), the camera rarely produces an out of focus picture. This is true when shooting with the EF 300mm f/2.8L IS from a higher angle, and it's true with this lens or the EF 70-200mm f/2.8L IS from the floor.
 
By comparison, the EOS-1D Mark II N rates a B. It performs reasonably well with the same 300mm lens, but the speed of focus is noticeably slower and sometimes too slow with the 70-200mm from the floor.
 
The EOS-1D Mark III is a killer camera for strobed volleyball photography.
 
Strobed basketball (EOS-1D Mark III: B, EOS-1D Mark II N: B) We've had decent results with both the EOS-1D Mark III and EOS-1D Mark II N, but neither camera tracks quite as well as needed in moderately bright gymnasium light. Both cameras earn a B.
 
womens_basketball_thumb.jpg
Peekaboo: Canon EOS-1D Mark III (fixed sub-mirror, firmware v1.1.3 beta) + EF 300mm f/2.8L IS, ISO 200, 1/320, f/7.1, -1 AF Microadjustment (Photo by Rob Galbraith/Little Guy Media)
 
Lenses used to shoot strobed basketball were the EF 70-200mm f/2.8L IS and EF 300mm f/2.8L IS, both focus-calibrated, and the EF 135mm f/2L, which wasn't focus-calibrated.
 
Q. It sounds like you've mostly been trying AI Servo. How well does the camera work in One Shot autofocus mode?
 
It's true, in the lion's share of testing, and in all real-world shooting, our Canon cameras are set to AI Servo, even when photographing stationary subjects. We're died-in-the-wool rear button focus users, and "lock" the focus when needed by releasing the AF-ON button (as was the case for the black-and-white photo below).
 
AI Servo focus stability in bright light and warm temperatures has definitely been improved. While the focus distance still shifts about too much, it does with the EOS-1D Mark II N too.
 
With either camera, the steadier the photographer the more stable the focus, particularly with telephoto lenses; conversely, the focus distance can fluctuate a fair bit when the lens starts to bob up and down some. So, neither camera's AI Servo focus holds as steady as it should on static objects. But, as we noted back on June 19, we've never used a camera with autofocus that is as stable as we'd like when set to continuously focus and the subject isn't moving.
 
womens_volleyball2_thumb.jpg
Focused: Canon EOS-1D Mark III (fixed sub-mirror, firmware v1.1.3 beta) + EF 300mm f/2.8L IS, ISO 200, 1/320, f/7.1 (Photo by Rob Galbraith/Little Guy Media)
 
But the question was about One Shot. And the best answer we can give is that while we haven't used One Shot much at all, we did one batch of tests where we shot the Phoenix runner standing still, using both the EOS-1D Mark III and EOS-1D Mark II N, in both AI Servo and One Shot modes.
 
For the AI Servo frames we had the cameras firing at their fastest frame rates, with the autofocus point on the runner's torso throughout. For the One Shot frames we focused on the background, then back to the runner, between each frame, taking care to steady the camera before activating the autofocus.
 
The AI Servo sequence shows some focus fluctuation, as expected, but the frame-to-frame focus consistency is markedly better than the comparable One Shot photos. This is true of the EOS-1D Mark II N, and true of the EOS-1D Mark III.
 
Make of this what you will. While we don't have any reason to doubt the results of this test (which you'll find in the downloadable sequences section), we also don't have a whack of real-world One Shot experience to accompany it.
 
Q. I don't shoot the same things you do, how well will the EOS-1D Mark III's autofocus perform for me?
 
Since first publishing this article, over 3000 photographers have sent questions and comments about EOS-1D Mark III autofocus. Several hundred of those have been photographers who own the camera and shoot the same sports we do with it, but several hundred more either shoot different types of moving subjects - birds, dogs, surfers, race cars and more - or they shoot exclusively subjects that are comparatively static - wedding formals on location or senior-year portraits in the studio, for example.
 
All these messages have been a great reminder of how varied the applications are for a camera like the EOS-1D Mark III. But all these uses add up to a real challenge when putting together a piece like this, since there's no way we can either test the camera's autofocus performance in every situation or make good predictions about shooting environments that are way different than what we know about. We hope that you'll be able to map at least some of our experiences with the camera to your own needs, and therefore the information and pictures in this article will be helpful to you.
 
Q. Is Canon working to further improve EOS-1D Mark III autofocus?
 
We don’t know. Canon in Tokyo, via Canon USA, has everything I shot in Phoenix this fall. This includes comparisons between the EOS-1D Mark III and EOS-1D Mark II N that show the EOS-1D Mark II N performing better when the sunlight is bright. Plus they’re undoubtedly soliciting files and feedback from shooters globally.
 
Presumably, they’ll take all this information and decide whether more improvement is required and if so, how that can be achieved. But this is all speculation on our part; only some key individuals at Canon in Tokyo know what the company’s plans are going forward.
 
Q. Do all EOS-1D Mark IIIs built before early October 2007 have a faulty sub-mirror?
 
Canon has indicated that at least some of the EOS-1D Mark III's built before early October 2007 contain a sub-mirror mechanism that is in good working order, and that if your camera's autofocus is meeting your requirements then you don't necessarily need to send it in for the sub-mirror fix.
 
While we have no reason to doubt what Canon is saying, at the same time our own experience is this: of the nine EOS-1D Mark IIIs we used before the introduction of the sub-mirror fix, bodies whose build dates varied but were made before early October 2007, all performed especially poorly in warmer temperatures. It therefore stands to reason that all nine contained sub-mirrors that were being affected by the heat.
 
To have this many affected bodies pass through here might mean we're just unlucky, or the number of units built before early October 2007 with properly functioning sub-mirror mechanisms might be small. It's impossible to know for sure.
 
Canon has promised to do the repair on any camera made before early October 2007, and it may be trickier to sell the camera down the road if it isn't confirmed to have a fixed sub-mirror, so our suggestion is that if your EOS-1D Mark III's autofocus is doing what you need it to do right now, then send it in eventually, once the crush of the first few weeks or months has subsided and at a time you can manage to be without the camera for the duration of the repair. It's your camera, however, so ignore this advice as you see fit.
 
More information about Canon's sub-mirror repair program is in the October 18, October 29, October 31, November 1 and November 16 updates to this article.
 
Q. I’m in the queue for the sub-mirror fix, but my repair date isn't for awhile. Should I install firmware v1.1.3 now or wait until my camera has been repaired?
 
Install firmware v.1.1.3 now. The sub-mirror fix and the firmware update are related only in that both are meant to help drive up the number of in-focus pictures the camera captures. But they're not interdependent; one doesn't require the other for you to see an improvement.
 
Q. How does the EOS-1Ds Mark III's autofocus compare to the EOS-1D Mark III?
 
The preproduction EOS-1Ds Mark III body we've had for a few weeks doesn't contain the revised sub-mirror mechanism - though all production bodies are promised to have it - and its prerelease firmware's autofocus algorithm may not be and probably isn't the same as shipping cameras. So, it wouldn't make sense to talk about the camera's autofocus based on our time with this particular unit.
 
In addition, we've only just taken delivery of a production EOS-1Ds Mark III and so its autofocus hasn't been checked out yet.
 
Q. Has Canon USA given you an official reaction to your conclusions about EOS-1D Mark III autofocus now?
 
No.
 
Q. When will you be writing about EOS-1D Mark III autofocus again?
 
The timing of any future addition to this article will be driven by Canon. If they choose to further develop the EOS-1D Mark III's autofocus algorithm, we'll definitely want to see if that improves the camera's autofocus and then write about it.
 
The three EOS-1D Mark IIIs we've tested for this update were manufactured with a properly functioning sub-mirror (so-called "blue dot" cameras), they were not earlier bodies that had been through the sub-mirror repair process. We've had requests to test a repaired camera as well, and the suggestion is a good one. But because of a backlog of particularly non-website work, we're not able to properly test a repaired body at this time and can't commit to doing so anytime soon.
 
Downloadable full-resolution sequences
 
Below you will find links to sequences of full resolution pictures shot with the EOS-1D Mark III and the EOS-1D Mark II N. The cameras were set to capture CR2s. The CR2s were processed to JPEGs in Canon Digital Photo Professional with Sharpening set to 3, then run through a Photoshop script that superimposed the active AF point on the frame, so you can see where the point of focus should be.
 
All of the EOS-1D Mark III photos were taken with cameras that had properly functioning sub-mirrors. All were shot with the release version of firmware v1.1.3 installed, except for the soccer game photos, which were shot with beta firmware prior to v1.1.3's official release. The photos' Caption field contains key shooting information, including the lens model, AF Microadjustment setting, focus mode, firmware version plus standard info like shutter speed, aperture and ISO.
 
Important viewing tips:
  • As before, the sequences have been selected not because they're portfolio-quality photos, but because they're representative of the autofocus traits of the EOS-1D Mark III and EOS-1D Mark II N as described in this article update. Note that these pictures haven't been adjusted in any way other than sharpening, and in some cases the colour and brightness would have benefited from some editing work. In other words, please judge focus in these pictures, but try not to get hung up on image quality if you see colour or contrast that isn't quite right.

  • To have a proper look at the photos, view them in a browser like Photo Mechanic that enables you to display them rapidly at full resolution. Key shooting information is in the Caption field. Also, consider printing some of the slightly out of focus frames, at sizes that are typical in your work, to see if your eyes notice the focus error when ink hits paper.

  • We had considered marking each photo in the downloadable sequences as either properly focused, very slightly out of focus or unusably out of focus, based on our notion of what would fall into each category. This idea was eventually rejected, because really the sequences are here to help you reach your own conclusions about the EOS-1D Mark III's autofocus system and its usability for your photography.

  • Comparing the autofocus of two digital SLRs whose sensors have different pixel pitches can put the camera with the smaller pixel pitch at a disadvantage. An explanation of this can run a bit long, so suffice it to say that if you have two cameras with the same size sensors but different resolutions, files from the higher resolution camera can show slight focus error that might not be apparent in files from the lower resolution one.

    Our experience, however, is that one camera's pixel pitch has to be pretty far removed from another's before this is a factor. It's also our experience that the EOS-1D Mark III, with a pixel pitch of 7.2µm, is not at a disadvantage in this way when being compared to the EOS-1D Mark II N and its pixel pitch of 8.2µm. Starting in May, and through the months since, we've made a point of resampling certain sequences of EOS-1D Mark III files down to the pixel dimensions of EOS-1D Mark II N files, which very roughly approximates making the EOS-1D Mark III into an 8.2µm camera. Examining the downsampled files, it's just as simple to identify the in-focus, slightly out of focus and unusably out of focus frames as it is at the camera's full resolution.

    To be fair, there is another factor that does come into play when comparing files from the EOS-1D Mark II N and EOS-1D Mark III: pixel quality. There is a crispness and clarity to EOS-1D Mark III CRs, particularly when processed through Digital Photo Professional, that comes from improvements Canon has made to the camera's sensor and processing and not from its modest resolution bump (relative to its predecessor). The EOS-1D Mark II N CR2 files, when converted, show almost the same crispness and almost the same clarity, but not quite. This has the effect of making it slightly easier to spot the point of focus in EOS-1D Mark III files. When analysing pictures from the two cameras, then, this means you have to study EOS-1D Mark II N files a bit more carefully to spot the point of focus. Do that, and the EOS-1D Mark III's slight disadvantage - though it's strange to call better image quality a disadvantage - is negated and a fair judging of the files from each model can be done.

  • Most of the downloadable EOS-1D Mark III sequences were shot with AF Microadjustment set to -1, though there are a handful where AF Microadjustment is set to 0 instead. To get an overall sense of the EOS-1D Mark III's tracking capabilities - which is the main reason we've posted full-res photos for download - it honestly doesn't make a difference whether you're looking at 0 or -1 sequences. When AF Microadjustment is set to -1, extended sequences tend to show fewer out of focus frames, but the improvement is slight.

  • In testing Canon's hardware and firmware changes in the EOS-1D Mark III, we had the camera set to the centre AF point exclusively, and varied only the frame rate (10fps, 8fps) and AF Microadjustment settings. As noted earlier in the article, the AF Microadjustment setting does make a small difference, but the frame rate doesn't.

    Otherwise, our experience has been that - like other 1-series Canons before it - the EOS-1D Mark III will deliver the strongest raw autofocus performance in configurations that are more basic. Assuming that the autofocus point can be kept on the subject almost all the time, manual selection of the centre AF point is the way to go. Our confidence about this was bolstered by the fact that Canon's testing regimen for the Phoenix trips called for only centre AF point shooting.

    It's not that the camera's other 44 points and other configurations don't serve a purpose, since they obviously do. We'll manually choose a point outside the centre if that's best for the composition, for example, as well as expand the number of active points in situations where it's difficult to keep the AF point right on the subject.

    But to see the best of what both the EOS-1D Mark III and EOS-1D Mark II N could do, we used the centre AF point only, and have included only centre AF point sequences below.

  • If you're not eager to download the entire 1.4GB of photos here, but you do want to see something that shows the bright light autofocus differences between the EOS-1D Mark III and EOS-1D Mark II N, then grab just the runner test and soccer player test sequences. These sequences are linked to the right of the first four thumbnails.
These photos are for personal viewing and printing only. They may not be republished in any form without the permission of the copyright holder. This includes the posting of these photos onto another server.

EOS-1D_Mark_III_Fast1_thumb.jpg
Canon EOS-1D Mark III

Notes: Each downloadable sequence contains one run. In the sequences called Fast1 and Fast2, the runner is moving quickly towards the camera, though her speed is well short of a full sprint. The run in Slow is a brisk jogging pace. Each sequence is a bit differerent, so if you download all three, then compare them to their matching EOS-1D Mark II N sequences, you'll have a pretty good idea of how the two cameras stack up in this test.

Download full-resolution sequence(s)
• Fast1: EOS-1D_Mark_III_Fast1 (36 Pictures, 75.1MB)
• Fast2: EOS-1D_Mark_III_Fast2 (36 Pictures, 73.2MB)
• Slow: EOS-1D_Mark_III_Slow ( 37 Pictures, 76.9MB)
EOS-1D_Mark_IIN_Fast1_thumb.jpg
Canon EOS-1D Mark II N

Notes: The downloadable sequences Fast1 and Fast2 each contain one run. The runner is moving quickly towards the camera, though her speed is well short of a full sprint. The run in Slow is a brisk jogging pace. D ownload all three and you can then compare to similar sequences from the EOS-1D Mark III. Be cause of the EOS-1D Mark II N's smaller buffer, the shooting time for each sequence isn't quite as long as that of the EOS-1D Mark III, and this made the jogging-speed run a little short. To work around that, we've combined two runs into one extended sequence in Slow, so you're better able to compare this camera shooting this runner speed to the EOS-1D Mark III version.

Download full-resolution sequence(s)
• Fast1: EOS-1D_Mark_IIN_Fast1 (23 Pictures, 43.1MB)
• Fast2: EOS-1D_Mark_IIN_Fast2 (23 Pictures, 40.8MB)
• Slow: EOS-1D_Mark_IIN_Slow (39 Pictures, 72.7MB)
EOS-1D_Mark_III_Player1_thumb.jpg
Canon EOS-1D Mark III

Notes: The sequences Player1 and Player2 can be compared to sequences of the same athlete shot with the EOS-1D Mark II N. These sequences were shot in Phoenix; Player3 is a different athlete doing the same drill here in Calgary (and her several layers of clothing hint at the fact it was in much cooler weather).

Download full-resolution sequence(s)
• Player1: EOS-1D_Mark_III_Player1 (36 Pictures, 77MB)
• Player2: EOS-1D_Mark_III_Player2 (36 Pictures, 76.5MB)
• Player3: EOS-1D_Mark_III_Player3 (35 Pictures, 85.7MB)
EOS-1D_Mark_IIN_Player1_thumb.jpg
Canon EOS-1D Mark II N

Notes: These sequences can be compared to ones of the same athlete shot with the EOS-1D Mark III.

Download full-resolution sequence(s)
• Player1: EOS-1D_Mark_IIN_Player1 (23 Pictures, 44MB)
• Player2: EOS-1D_Mark_IIN_Player2 (23 Pictures, 44.5MB)
EOS-1D_Mark_III_Hurdles1_thumb.jpg
Canon EOS-1D Mark III

Notes: This sequence can be compared to one of the same athlete shot with the EOS-1D Mark II N.

Download full-resolution sequence(s)
• Hurdles1: EOS-1D_Mark_III_Hurdles1 (22 Pictures, 55.7MB)
EOS-1D_Mark_IIN_Hurdles1_thumb.jpg
Canon EOS-1D Mark II N

Notes: This sequence can be compared to one of the same athlete shot with the EOS-1D Mark III.

Download full-resolution sequence(s)
• Hurdles1: EOS-1D_Mark_IIN_Hurdles1 (22 Pictures, 44.4MB)
EOS-1D_Mark_III_Hurdles2_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Hurdles2: EOS-1D_Mark_III_Hurdles2 (24 Pictures, 55.3MB)
EOS-1D_Mark_III_Track_thumb.jpg
Canon EOS-1D Mark III

Notes: This sequence can be compared to one of the same athletes shot with the EOS-1D Mark II N.

Download full-resolution sequence(s)
• Track: EOS-1D_Mark_III_Track (28 Pictures, 67.9MB)
EOS-1D_Mark_IIN_Track_thumb.jpg
Canon EOS-1D Mark II N

Notes: This sequence can be compared to one of the same athletes shot with the EOS-1D Mark III.

Download full-resolution sequence(s)
• Track: EOS-1D_Mark_IIN_Track (25 Pictures, 61.4MB)
EOS-1D_Mark_III_AI_SERVO_thumb.jpg
Canon EOS-1D Mark III

Notes: Both the AI Servo sequence and the series of One Shot frames can be compared to ones of the same athlete shot with the EOS-1D Mark II N.

Download full-resolution sequence(s)
• AI Servo: EOS-1D_Mark_III_AI_SERVO (36 Pictures, 70.8MB)
• One Shot: EOS-1D_Mark_III_ONE_SHOT (14 Pictures, 27.6MB)
EOS-1D_Mark_IIN_AI_SERVO_thumb.jpg
Canon EOS-1D Mark II N

Notes: Both the AI Servo sequence and the series of One Shot frames can be compared to ones of the same athlete shot with the EOS-1D Mark III.

Download full-resolution sequence(s)
• AI Servo: EOS-1D_Mark_IIN_AI_SERVO (23 Pictures, 40.3MB)
• One Shot: EOS-1D_Mark_IIN_ONE_SHOT (12 Pictures, 20.7MB)
EOS-1D_Mark_III_Vault1_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Vault1: EOS-1D_Mark_III_Vault1 (5 Pictures, 16.6MB)
EOS-1D_Mark_III_Vault2_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Vault2: EOS-1D_Mark_III_Vault2 (7 Pictures, 14.7MB)
EOS-1D_Mark_III_Wave_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Wave: EOS-1D_Mark_III_Wave (11 Pictures, 31.9MB)
EOS-1D_Mark_III_Soccer1_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Soccer1: EOS-1D_Mark_III_Soccer1 (25 Pictures, 80.1MB)
EOS-1D_Mark_III_Soccer2_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Soccer2: EOS-1D_Mark_III_Soccer2 (9 Pictures, 21.9MB)
EOS-1D_Mark_III_Soccer3_thumb.jpg
Canon EOS-1D Mark III

Notes: The EOS-1D Mark III tends to autofocus better outdoors when the light is down from full brightness, as it is in this sequence.

Download full-resolution sequence(s)
• Soccer3: EOS-1D_Mark_III_Soccer3 (11 Pictures, 33.4MB)
EOS-1D_Mark_III_Soccer4_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Soccer4: EOS-1D_Mark_III_Soccer4 (19 Pictures, 53.2MB)
EOS-1D_Mark_III_Soccer5_thumb.jpg
Canon EOS-1D Mark III

Notes: None.

Download full-resolution sequence(s)
• Soccer5: EOS-1D_Mark_III_Soccer5 (10 Pictures, 25.1MB)
EOS-1D_Mark_IIN_Soccer_thumb.jpg
Canon EOS-1D Mark II N

Notes: None.

Download full-resolution sequence(s)
• Soccer: EOS-1D_Mark_IIN_Soccer (17 Pictures, 39.3MB)

Thanks to Canon USA and Canon Inc. for their efforts to improve EOS-1D Mark III autofocus and to assist us in the testing of this camera. Also, thanks to Nick Didlick, Mike Blake, Andy Clark and the many other photographers who've sent files, feedback and helpful information over the past several months.
 
Revision History
December 11, 2007: Added information about Canon USA's extension of the EOS-1D Mark III's warranty
Next Page: December 11, 2007 update
Go to Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 
Go to advertiser website.
©2000-2013 Little Guy Media. Not to be reproduced without written permission.