Autofocus Torture Test Updated (Canon 1D MKIV & Nikon D3S added)

Background:

Autofocus is one of the most important aspects of any camera. There has been much debate over the past few years about how well autofocus performs on the various camera models. We originally developed this AF torture test as part of an ongoing article about the autofocus problems that surrounded the original introduction of the Canon 1D MKIII. Using this test we were able to discern that there was most likely a hardware issue at the root of these AF problems. And, so there was. Anyway, people seemed highly interested in this particular autofocus test, so we continued to conduct this test during various camera reviews.

How we test:


The tests are fairly simple and straightforward. A camera is placed on a tripod and a runner runs directly at the camera at a steady pace. The tests are all performed using servo focus mode at f/2.8 using the best 300mm f/2.8 lenses available from the respective camera manufacturers. The camera exposure is set to ensure that a very high shutter speed (usually 1/2000 or higher) is achieved throughout the run. Normally, at least 5 test runs are conducted and then coded for the number of in focus images in each sequence. A percentage is then taken for each test run. These 5 percentages are then averaged to get an overall result. These tests are all done using the default camera settings and default custom function settings. Tests are done using the center focus point only.

This is a specific Auto Focus torture test, not an overall Auto-Focus score:

Please note that these results do not necessarily reflect the overall autofocus system performance of any particular camera. They do provide a lot of insight into how well the AF systems stand up to one of the toughest tasks they can possibly face.

Overall winner is denoted in red.

Results of Canon 1D MKIV and Nikon D3s

Canon 1D MKIV Results

Trial 1

45 in focus out of 50= 90% in focus

Trial 2

56 in focus out of 60= 93% in focus

Trial 3

48 in focus out of 55= 87% in focus

Trial 4

49 in focus out of 55= 89% in focus

Trial 5

52 in focus out of 58= 90% in focus

Canon 1D MKIV Overall= 89.4% in focus


Nikon D3s Results

Trial 1

29 in focus out of 46= 63% in focus

Trial 2

37 in focus out of 47= 79% in focus

Trial 3

46 in focus out of 53= 87% in focus

Trial 4

40 in focus out of 52= 77% in focus

Trial 5

38 in focus out of 47= 81% in focus

Nikon D3s Overall= 77.4% in focus


There has been lots of talk about whether the new Canon 1D MKIV auto-focus will outperform the poorly received Canon 1D MKIII auto-focus system. In actuality, our tests of a good Canon 1D MKIII body showed very good results on this test. However, the Canon 1D MKIII bodies that had hardware, and possibly other quality control issues, did not perform well at all. So, while the 1D MKIII was certainly capable of performing well on this test, some examples of the camera did not. As a result of these inconsistencies, the recalls by Canon, and a lot of internet discussion of poor auto-focus performance, the camera quickly gave Canon a black eye.

In the end, this test shows that the Canon 1D MKIV does indeed perform better than the best Canon 1DMKIII we tested. In fact, it turns in the best results we have ever seen from any camera. The surprising thing is really how well the Nikon D3s does on this test. While it does not produce the tack-sharp images of the 1D MKIV, it does perform very respectfully and has improved significantly over the prior Nikon performance levels.

What about other tests of auto-focus that suggest the MKIV may perform well on a running test, but perform poorly when shooting certain sports? Well, here is our current response to this argument.

You may view more detailed results for the previously tested cameras in this and other sections of the site.

Drew Strickland

51 Responses to “Autofocus Torture Test Updated (Canon 1D MKIV & Nikon D3S added)”

Read below or add a comment...

  1. Dan says:

    Extrange 1D Mar III vs D3 the winner is Canon 69% vs 52% sorry but i cant beilive this test

  2. Al says:

    Excited to see the improvement in the 1D MKIV as I prefer my Canon lenses for this type of sports shooting. But I’m most surprised by the results delivered from the 7D. Think its time to take that body for a test ride…

  3. PAUL HAMES says:

    If this is the case, Why have lots of photographers who were canon have now migrated to Nikon!? I am a Canon pro photographer. I believe that Canon have the edge over Nikon on colour, being more organic and less digital. The two things that canon need to do is sort out the flash control on the Eos 1 range and the focus. So if this is correct then Canon are back in front. I will not go back to Nikon I have invested to much in glass. Also if the camera was perfect in every way then selling a new model would be much harder. Sticking with canon.

  4. Sergio says:

    How can you state that an 1D MkIII outperforms the D3 autofocus (moving subject) ?
    I have regretfully bought am MkIII and afterward did test the fantastic D3.
    For flying birds I stepped back to the old beloved 1D Mk II and leave home the Mk III
    ciao
    sergio

  5. Michael says:

    No kidding about the D3 result. I am glad to see some good results with the 1D4, but I have not seen anything to indicate that the D3 autofocuses this poorly. Just the delta betweemn the D3 and the D3s is confusing.

  6. Jim says:

    This test is based on one type of shooting and I doubt the results. I am a Canon user but have shot with the D3 and it outperformed the autofocus on a Mark III. In fact I was considering the switch after using the D3 it was that much better than the Mark III. You do not shoot anything but a single person running at the camera, not really a good sports scenario unless all you shoot is track. You need to test it on Football, Basketball, Hockey and Baseball as well as other sports, where the movement is erratic. Poor test.

  7. Jim says:

    One more thing, Your quote “They do provide a lot of insight into how well the AF systems stand up to one of the toughest tasks they can possibly face.” WRONG. This is one of the easiest scenarios for autofocus to track, not the hardest. Even the inferior and horrible autofocus of the D1 and D2H could focus a single object like your test.

  8. Thys says:

    A randomly moving runner will probably yield more equal results between the 2. Good test nonetheless. Must say am VERY skeptical about RG agenda – judging by the ads on his site, he’s a serious Nikon cheerleader nowadays.

  9. Anthony says:

    Note the use of default camera settings. That alone is sufficient to limit the real-world value of this test.

  10. Giuseppe says:

    You will be sure of this one I’m saying: all the Canon cameras (either 40D, 50D, 5D, 5D II, 7D, 1Ds, Mark IV), has and will have focusing problems !!!
    Nikon is another world !!!

    Ciao from Giuseppe. – Italy

  11. inTempus says:

    Honestly, these tests are biased towards the 1D Mark IV. By default the D3/D3S is set to ignore focus lock and to give priority to continuous shutter speed. In essence, when set to default settings, the D3 will fire as quickly as it possibly can without giving any consideration to whether or not it actually has AF lock before firing the next frame.

    By default the 1D Mark IV gives AF lock priority. That means the 1D Mark IV will not fire until it thinks it has AF lock on the subject.

    In essence you have one camera firing wildly as fast as it can with no regard for AF lock (D3/D3S) and the other only firing if it has AF lock (1D4).

    If you wanted to make a truly representative test, you wouldn’t use default factory settings but would instead set all cameras up with the same settings, where possible.

    This test is completely invalid IMHO. Although it’s interesting to note the 1DsIII and 1DIII are bested by the 1DIV in this particular test. But then it’s well known that the 1DIII didn’t fair well in tests where the subject was approaching the camera directly.

  12. Sean says:

    I wholly agree with Jim’s comments above, try it again with the various sports and compare results.
    I will say though that I am happy that somebody is doing this good work, keep it up and thank you for your great efforts!

  13. ShutterDog says:

    What are the error margins of these numbers? Without the error margin information, the comparison is meaningless.

  14. Tony says:

    This confirms my suspicions that the mkIII was by no means as bad as many believed. I had considered switching to Nikon but after talking to many renowned photogs I went with with a mkIII.
    This test is no more flawed than the RG tests which many took as gospel. It is also worth noting how much stress is put on correct custom function set up on the mkIV -amost unmentioned in mkIII – & a reason why the more thinking photographers attributed dissatisfaction to user error. Something many users could not come to terms with.
    We are free to believe in whatever we choose & most of what we choose is due to bias or predisposition. I suppose my bias is because of the price of Nikon superteles.
    Congrats to Canon for the mkIV.
    Happy mkIII owner.

  15. Russ says:

    I would have loved to see a comparison to the 1D Mk II. The Mk III produces a better image but the Mk II focusing is far superior. Is the Mk IV better than the Mk II?

  16. inTempus says:

    Tony,

    I am a 1D Mark III owner (and a very happy one) that purchased one of the first 1D Mark IV’s to hit our shores. I do believe the AISERVO tracking of the 1D4 is superior to the 1D3′s. However, in my testing, the 1D3 locks focus faster than the 1D4 and also is better in low light AF with a single AF point. It seems Canon traded performance in a couple of areas on the 1D4… and I’m not sure I like it.

    I am not a Nikon shooter, but I am also not a Canon fanboy. When I see a test that I believe is faulty, even when it favors my brand of choice, I want to point it out. This test is, IMHO, faulty and doesn’t give an accurate representation of the D3′s true AF performance under this particular circumstance.

    I liken it to setting up a drag race between a ZR1 Corvette and a Porsche 911, but forcing the ZR1 driver to keep his Vette in first gear. What would be the point of such a “test” of speed?

  17. Guy Collier says:

    Drew – the point made about about not setting the Nikon to focus priority is right on the money. As you’ve tested it the Nikon doesn’t care whether it’s acquired AF, it simply fires.

    For that reason alone you need to perform the test again. I also think it’s HIGHLY dubious to call this a torture test when anyone using these cameras will set them up as appropriate for their uses. I also think we both know that if a D3s could only achieve 50% accuracy we’d be hearing lots about it.

    As it stands this is useless, because one simple setting on the D3s to set to focus priority will mean staggeringly different results. I’m more than a little surprised that you’re lambasting Rob Galbraith so publicly whilst making such an obvious, elementary mistake in your test procedure. Of course, you did this before and look who turned out to be right….

  18. admin says:

    @inTempus and Guy Collier

    Thanks for the feedback.

    However, I don’t think this is much of a reason to dismiss the findings.

    I have gone back and added some more details regarding each of the five trial runs.

    As you can now see, the Canon 1D MKIV fired more overall images during the trials. Bear in mind these two cameras were tested at the same time, one right after another using the same operator, same runner, same track, same starting point, etc.

    So, I guess your argument would go something like this. Since the Canon can not only decide on a focus point, lock it in and take more shots than the D3s supposedly firing indiscriminately then that somehow makes the D3s look better in terms of auto-focus.

    You are right, however, the Nikon only showed the ability to focus lock on about 3 or 4 of the frames in each running trial sequence.

  19. admin says:

    @Jim

    Regarding the auto-focus torture test being an easy task for an auto-focus system. In the past I would have disagreed entirely.

    Now, however, I would only disagree as regards Nikon. Canon seems to have finally “nailed it,” as others have also noted.

    We may have to come up with a new test for the pro level Canon models as they appear to have just about maxed out this one. Nikon still has a ways to go, it seems.

    Drew

  20. Guy Collier says:

    Then repeat the test with the correct ‘focus priority’ settings.

    It seems obvious to me. The Nikon, as you tested it, doesn’t care whether it’s acquired focus. Therefore it’s not locking.

  21. matt says:

    This is the stupidest test I’ve seen so far. I’ve been shooting Canon for 14 years and switched to Nikon D3 after the Canon Mark III fiasco. Really man, get a grip.

  22. alpha says:

    rubbish test
    do not know how to say it better ………..

  23. Aussie pics says:

    Just as losing face can destroy a career, the reputation of a tester can be totaled when obvious errors are made. Prophotohome.com’s future isn’t looking too bright. You can’t seriously have the Nikon set to release priority and then tell the world it’s AF performance was so poor. Have the courage to do the test again and at least balance the test. Given the Canon can’t be set to release priority but the NIkon can, is your judgement so poor you truly believe it’s the right thing to do ? Is Canon paying you to make their camera look better ?

  24. Aussie pics says:

    As an aside, try the test again in the lower light at dusk. Dial up the ISO’s to try to freeze the subjects, or even throw a bit of flash in there with the AF assist disabled. Let’s see how they both work then, I bet I know which one falls to below 10%.

  25. Peter G says:

    I re-read your report today, and am still not impressed. Yes, I know you used the camera makers default” settings, and I disagree with NIKONS default settings. I changed my D3 to ” FOCUS” priority, and that solved a lot of the problems that I experienced the first time I used my D3. However, I feel your test does not go into the detail that Rob Galbraith did with his testing.
    You merely had a runner going up a track running towards a camera. There is a lot more to action shooting than simply running at a camera. Tracking left to right etc.
    It seems that you have a” beef” with Rob Galbraith. To me, his testing seems a lot more scientific and balanced that your report. I re-read his report today, and it was fair and balanced .
    I am a NIKON man for many years, so, I may be biased, but, there is a lot more to a camera than just the ” bragging rights” to how many shots out of 10 for example.
    Reliability comes into it, and , in my opinion, along with others, NIKON wins hand down in that area.
    In fact , a few photographer friends I have know for 20 odd years, went from NIKON to CANON, and are now coming back to NIKON. They got tired of the promises and reliability issues associated with CANON.
    Yes , a good idea to have a new test for CANON….How long do they go before failing ?

  26. inTempus says:

    Admin,

    Thanks for the additional info you posted. I find it interesting that the 1D4 still fired substantially more shots than the D3S did. That’s very useful information. That means if you had put the D3S into AF priority mode two things likely would have happened.

    1) More in focus shots would likely be taken, perhaps rivaling the results posted by the 1D4. This is an assumption on my part of course.

    2) The D3S would have posted even fewer shots, which honestly is quite alarming. It would appear your findings indicate that the D3S is not shooting anywhere near 9fps even under ideal conditions.

    Again, thank you for sharing your findings.

  27. Joe Twin says:

    I find it interesting, for sure, but somewhat incomplete.

    An auto-focus system could actually be wildly inaccurate, yet still fair quite well on a full-speed motion test such as this. It would be interesting to see the ‘tracking’ abilities of the various AF modules alongside the the simple ‘accuracy’ (and while we are at it, alongside the accuracy/tracking in the much much darker environments (0-2 EV?) that these new high-ISO powerhouses are often employed)

  28. Bob says:

    “So, I guess your argument would go something like this. Since the Canon can not only decide on a focus point, lock it in and take more shots than the D3s supposedly firing indiscriminately then that somehow makes the D3s look better in terms of auto-focus.”

    he means you should set the Nikon to not shoot unless it thinks it’s in focus. then re-do your test.

    then learn how to read.

  29. admin says:

    @inTempus “1) More in focus shots would likely be taken, perhaps rivaling the results posted by the 1D4. This is an assumption on my part of course.

    2) The D3S would have posted even fewer shots, which honestly is quite alarming. It would appear your findings indicate that the D3S is not shooting anywhere near 9fps even under ideal conditions.”

    You got it. My assumptions exactly. Of course, we wouldn’t know for sure until we tested that non-default test condition.

  30. admin says:

    @Bob and others

    So, let me get this straight. You would have me redo the test such that we do an apples to oranges comparison in terms of af speed to try and see if Nikon can up its percentage of in focus frames? Is this right. Doesn’t hardly seem a fair test.

    Right now the Nikon while firing “indiscrimately”, as someone earlier stated, can only muster a little over 8fps in real world shooting. The MKIV is shooting the full advertised 10 fps in real world shooting and nailing the auto-focus.

    So, my guess is we would then be comparing percentages of a 10 fps Canon machine with about a 6fps Nikon machine.

    Is this really providing an equitable test? Not that I mind possibly redoing the test at some point.

    There are many lobs that could be thrown at the overall design of this study, but I don’t think this is one of them. Start thinking more along the lines of sample sizes, n and other potential areas for improvement of the study.

    Drew

  31. Thanh says:

    You must be joking! Is this the “scientific test” that you use to bash Rob Galbraith’s test? Rob has almost 2GB of samples to back up his claims. What are yours? And since when a simple straight forward runner test become a “torture” for camera’s AF systems?

    And how on earth that such flag-ship cameras as the D3, 1D4, 1D3 can get only 52%, 69%, and 89% of shots in-focus, respectively? And what is your definition of an “in-focus shot”? Let’s take a look at this small test of Per L, a D300 user: http://dslr-video.com/blogmag/?p=1408 . Under much more difficult situations (distracted background, subject almost totally obscured, variable colours, variable lightings etc), his/her D300 was able to get almost 100% in-focus shots according to his/her samples. That high percentage of in-focus shots is very different from the surprisingly low figure that you reported here (55%). Unlike you, he/she provides detailed photos as evidence so guess whom will I believe now?

    Finally, assume that the test results that you reported are correct (which I really doubt), what is the point of testing the cameras using the default camera settings and default custom function settings? You think sport-photography pros use their equipments as point-and-shot camera, just point and shot without fine-tune any settings?

    And what you answered others’ comments also do not make sense to me. Sorry, but the more you say/post, the less I believe in your “scientific” test. Let me be honest. Your current tests are just not in the same level as those from Rob Galbraith. In fact, I don’t think they are of any practical value. Sorry

  32. Bob says:

    Drew,

    Do I understand this? you ran an autofocus tracking test with one of the cameras set at focus priority and the other NOT set at focus priority? And you think that’s apples to apples ?

  33. inTempus says:

    Drew,

    I would like to see a test where the D3S is configured to give priority to AF like the default setting of the 1D4. Sure, it will slow the frames per second down (which would be interesting to see how much) but it would give us a better picture of how the D3S stacks up against the 1D4. It’s odd that Nikon would set the default setting the way they have, but I’ve yet to meet a D3 shooter that didn’t change that before firing their first shot.

    If the cameras are configured to perform basically the same way, I don’t see this as being an apples to oranges comparison. If anything, having them configured to behave two different ways is more of an apples to oranges comparison in my view.

    Thanks again for your testing, it’s appreciated.

  34. admin says:

    @Bob

    Not sure where people got the idea that AI Servo mode on a Canon requires a focus lock to fire. It does not. One shot mode does. Perhaps that is part of the confusion here.

    All of these tests were done on AI Servo for Canon and Continuous for Nikon.

    What “inTempus” is suggesting, I believe, is that he believes the Nikon does not focus well in the default continuous mode (called “Release” by Nikon). There are two other settings available for the af-c mode on the higher end Nikon focusing systems. One is Release + Focus and the other is Focus. I don’t have a Nikon D3s with me on the road. However, I did a very quick test on the Nikon D300s we use at the trade shows, and these alternative continuous shooting modes slowed things down considerably.

    Nikon probably ships it in the mode it does because this is the only mode that has a chance of reaching the advertised frames per second across most shooting situations. Besides, it doesn’t really perform that poorly. The % returned by the D3s in this test is actually quite good. Not up to the MKIV, but still very good.

    Drew

  35. admin says:

    @Thanh

    Rob appears to have spent a lot of time on his tests. I can see how one would think that counts for a lot. After all, effort usually equals results. 2 TB sure does sound large, I guess.

    But where you see a long time spent as equalling iron clad results, I see thousands upon thousands of uncontrolled, unaccounted for, and unexplained variables. I see variance and error running wider than the mighty Mississippi.

    At best, I see a case study that might provide some insight into developing a number of different experiments. One or two of which, if they are well-designed, might actually provide some true, repeatable insight into this supposed af issue.

    In regards to his results, he may be truly on to something. But, at this point no one can tell, not even Rob himself. He may believe it to be true, but he has no way to conclusively show this to the world. Hence, the “gentle” prodding to go the extra mile to do more than just post some conjectures about indoors and outdoors and some perceived variances between sport types. All I’m trying to do is get him to up his game prior to his next upcoming report.

    fyi, the D3s is an amazing camera! As is the MKIV. This is not a platform war, although I know that may disappoint some of you. The D3s is the clear winner on the high ISO front, as we recently reported here. I personally own and shoot both platforms.

    Drew

  36. Guy Collier says:

    ‘Not sure where people got the idea that AI Servo mode on a Canon requires a focus lock to fire. It does not. One shot mode does. Perhaps that is part of the confusion here.
    All of these tests were done on AI Servo for Canon and Continuous for Nikon.’

    You’re mixing up an AF drive mode with AF settings here. Both AF-S AF-C can be used for either release or focus priority with custom settings a1 and a2. (which is the heart of this argument). The Canon does the same so you’re looking at the defaults for C.Fn III (specifically -3). It’s set to 0 which is AF priority/Tracking Priority (ie. equal to focus priority on the Nikon).

    The apples to apples argument is utterly disingenuous – you’re saying that what the camera maker sets as an ‘out of the box’ set-up is the one you’ll use, regardless of whether they match. That’s deeply, deeply flawed.

  37. admin says:

    @Guy and inTempus

    The MKIV AI Servo mode actually has 4 c.fn possibilities- they are as follows:

    0: AF priority/Tracking priority
    1: AF priority/Drive speed priority
    2: Release/Drive speed priority
    3: Release/Tracking priority

    The D3s Continuous shooting mode has the following 3 choices:

    0: Release
    1: Release+Focus
    2:Focus

    Which do you think makes an apples to apples comparison?

    Canon 0 vs. Nikon 2- I take it. Even though it will most likely slow the D3s frames per second down to a crawl? This is not likely to improve the overall %, but I’ll be happy to do the test, if there is enough interest. My doubt stems from the simple fact that with a lot fewer frames shot even 1 out of focus frame is going to hold a lot more weight than it does now.

    I will be out of pocket at trade shows and traveling until the end of March. Perhaps it could be done sometime in early April?

    What do you all think. Is this a more fair comparison, and if so, is it worthwhile doing?

    Thanks,
    Drew

  38. le_eiji says:

    this isn’t all that surprising to me. there are significantly more 1DMark IV users in the latest Olympic games and that tell you everything.

  39. Jerry says:

    admin says:
    March 4, 2010 at 5:09 am
    @inTempus and Guy Collier

    However, I don’t think this is much of a reason to dismiss the findings.
    ——————————–

    Actually, it is.
    inTempus is spot on. Activate the FOCUS PRIORITY on the D3s (which is turned OFF by default) and leave the Canon settings as they are (the Canon equivalent turned ON by default), and re do the test. I am sure the results will be drastically different.

    You cannot expect fair results when one camera is dialed in to give you maximum FPS irrespective of focus and the other camera dialed in to give you maximum focus irrespective of FPS.

    Set up a fair test and try again, you will be very disappointed, unfortunately, so you either will not post the results or you will not redo the test.

    Either way, this test is mildly entertaining, but not much more than that. In terms of real world accuracy, it has sadly not delivered the goods.

  40. admin says:

    Hi Jerry,

    I’ll take that as a vote to do another test. Anyone else want to see Canon 0 vs. Nikon 2?

    Drew

  41. Guy Collier says:

    Yes – exactly that. That’s apples to apples. Default settings to default settings never will be.

    As a point of reference – I took my D3 out yesterday and shot cars coming directly at me at 30mph. There was no slowdown whatsoever between release and focus priority. There were, however, more frames in focus with focus priority (as I’d expect). That tells me that the AF module in the D3 is fast enough to track a 30mph car AND fire at 9fps. So I don’t expect yours to ‘slow to a crawl’.

  42. inTempus says:

    Drew,

    Another vote for 0, 2. I think that would be ideal.

    Also, if you want to clip the nay-sayers in the bud, do as RG did and make your RAW files available for download so folks can check them out. You’re catching flak on the forums discussing this topic as the Nikonians are saying your tests aren’t as good as RG’s because, you guessed it, you didn’t share your RAWs.

    If you care to expand your next round of tests, shoot a car running in a fairly tight circle so that you have it running parallel to your camera, diagonally towards it, directly towards it and then away from it. I’ve seen a couple of professional motorsports photogs saying they’ve had trouble with the Mark IV in this scenario. I would like a definitive test in this area as it also puts to bed two other points RG made. He claimed the Mark IV stumbles in bright light (do it on a sunny day) and that it gets confused when the subject moves around.

    But, just redoing the original test would be a bonus. I am really interested in finding out how much more the D3S slows down.

  43. Jim says:

    I think in the end either the Mark IV or D3s will provide you the right tools for the job. Nothing is going to be 100 percent, but I do think the Nikon is much closer to the Mark IV than what your test shows. I have used both in a variety of sporting events and both worked well. Was every frame in focus , no. But i’ll settle for a 85-90 percent in focus rate. In fact the 7D is very close to the Mark IV in real world AF results. I would still like to see a Mark2N, in your test since that camera had the best AF of any camera I had used.

  44. malla says:

    Mmmmmmmmmmmmmmm the 7d smokes the D3 lol.

  45. Mark says:

    The bottom line is they are both fantastic camera’s, I personally shoot with the 1d4 but I’m sure I could enjoy the iso bettering D3s as well, am I going to sell $15,000 worth of lenses because most Nikon owners say there camera is better NO. I am happy with the 4, nothing is perfect, I feel its pretty close though. Get over it, its an old boring argument, just be happy both companies have winners.

  46. Frank says:

    First, I have to state that I am a Mechanical Engineer and have run many tests of various equipment. Although Rob G. gets an A for effort, I would not make a purchasing decision based on his tests. I do find the tests interesting but can only conclude that better testing methods should be done.

    I know I will catch flack but that flack is from those that do not understand real scientific method; there are too many variables not nailed down in his tests.

    For example, lighting conditions, person holding the camera, for sports the target clothing, and human-fatigue factors during the test. You do not think these can alter a test; you would very surprised! I have seen too many unexpected factors show up in test results causing too many do-overs.

    So his test shows that both manufactures cameras, and only those particular cameras, being used by those two people were inconsistent at some things and consistent at others. At most this very unscientific test might show that more testing is required.

    I think it would be absolutely silly to change brands based on these ‘trials’. In the case of Rob’s testing of the 1D MK III, his testing led Canon to do real scientific testing and it turned out he was correct.

    Unless you are a pro and your livelihood depends on getting the money shot more than 89.4% of the time, both brands sound good enough for now.

  47. Ricelow says:

    any chance of including Pentax K-5 and Nikon D7000?

    Thanks

Trackbacks

  1. [...] the camera “nailed it“ in regards to the primary issue he had with the MKIII. I would say so, our most recent testing shows the same thing. The Canon 1D MKIV outclasses all other cameras in this autofocus test getting [...]

  2. Anonymous says:

    [...] con respecto al realizado por Galbraith. Dice que incluso la 1D MkIII supera a la D3 en aciertos Autofocus Torture Test Updated (Canon 1D MKIV & Nikon D3S added) | Pro Photo Home Saludos __________________ Blog *1Ds Mark III*16-35 2.8 II*24-105 4 IS*70-200 4 IS*300 2.8 [...]

  3. [...] I can't really say without knowing all the variables involved, but generally the performance of the AF system tails off as the amount of available light drops off. The AF system needs a contrasty subject to be able to lock focus properly (try autofocusing against a bright white wall), so I'm guessing his subject matter isn't illuminated that well. Additionally, some lenses hunt for focus worse than others. My Canon 15-85mm EF-S (one of Canon's latest) is much, much better achieving focus in low light than my Canon EF 50mm (an old lens design), so the question is what lenses is he using with the camera? Is he using Canon lenses? The AF system in the 1D series is (as you would expect) the best as they use cross-type AF sensors across the whole AF sensor array where as the 5D Mk2 only has single axis (less accurate) sensors on the peripheral AF sensors (however,the centre focus point is a cross-type point). It's possible that the AF may be less reliable in low light whilst using an outer AF point as they are less sensitive. FWIW, the 7D's AF system is more akin to the 1D in that it employs cross-type AF points for all of the AF points. HTH? Also, take a look at this AF torture test: Autofocus Torture Test Updated (Canon 1D MKIV & Nikon D3S added) | Pro Photo Home [...]

  4. [...] Gracias de nuevo Es que todas fallan, en mayor o menor medida. Mira esta prueba: Autofocus Torture Test Updated (Canon 1D MKIV & Nikon D3S added) | Pro Photo Home Pues bien, por mi poca experiencia, te puedo decir que con mucha menos luz (pabellones), con [...]