LensRentals.com

Autofocus Reality Part 3B: Canon Cameras

Posted by

Some days it's good to be a geeky gear-head. This is one of them because a) I finished testing autofocus on all of the Canon camera bodies we had, b) I actually found out some interesting stuff, and c) I got worked up about camera marketers while doing it, so I have my next post in the works already.

Because I've been writing this series on the fly (telling you what we found as we found it), I'll keep going in that fashion and keep the conclusions for the end of the article.

What We've Found So Far

In the first Autofocus Reality article, we demonstrated two things:

1. Phase-detection autofocus (even using still targets and center-point only) wasn't nearly as accurate as contrast detection.

2. The contrast-detection autofocus was about as accurate as the most careful manual focusing.

Part two of the series showed that a few newer lenses did focus as accurately as contrast detection on 5D Mark III cameras but not on 5D Mark II cameras. The third article (part 3A) showed that the newest Canon lenses (40mm f/2.8, 24mm f/2.8 IS, 28mm f/2.8 IS, 70-300mm L IS and 300mm f/2.8 IS II) focus more accurately when mounted to 5D Mark III camera but not on 5D Mark II cameras.

The Canon 70-200 f2.8 IS II, oddly enough, seemed not quite as good as the above lenses, but more accurate than the older ones. Whether this was oddness in my measurements or a real finding, I wasn't sure.

Since we knew that the newest lenses autofocused accurately on 5DIII but not on 5DII cameras, the next step was obviously to compare an accurate AF lens on different camera bodies to see with which ones it was capable of accuracy.

We expected the 1Dx (which has the same AF system as the 5DIII) would be accurate. We weren't sure about the others.

Today's Contestants on The Focus is Right!

We know the Canon 28mm f/2.8 IS II had accurate autofocus on the Canon 5DIII so we chose one copy to be our test lens.

We AF microadjusted each camera to the lens prior to shooting. If the camera did not offer AF microadjustment, we checked the body with the test lens and exchanged it for another copy if the lens back or frontfocused at the test distance.

(Microfocus adjustment makes absolutely no difference in the shot-to-shot variation in AF--it only improves the average value of the group. But I got tired of explaining that to people in the previous articles. It was easier to just do it than to answer another 50 emails.)

We then tested it in our Imatest lab using one copy of each Canon camera we carry. To save you from running amok on the internet, finding out which cameras are how old and have what kind of autofocus, I've listed that information in the table below.

 


Camera Year released Year firmware AF description
1DsIII Dec-07 12/16/09 45 point, 19 cross, 26 assist, f/4 at center, dedicated AF processor
5D II Dec-08 2/28/12 9 point, f/2.8 cross center
50D Dec-08 2/28/12 9 point cross, dual diagonal center cross
7D Aug-09 4/25/11 19 point, all cross, center dual diagonal cross (advanced algorithm), first "zone AF" and "spot AF"
1DIV Dec-09 3/29/12 45 point, 39 cross (f/2.8-f/5.6, f4 at center), first AF expansion camera
60D Aug-10 6/19/12 9 point, f/2.8 cross center
T3i Jan-11 1/30/12 9 point, f/2.8 cross center
1Dx Mar-12 2012 61 point reticular, 41 cross type, 5 dual diagonal cross, including center
5D III Mar-12 2012 61 point reticular, 41 cross type, 5 dual diagonal cross, including center
T4i Apr-12 2012 9 point, f/2.8 dual cross center, hybrid CMOS AF Live View

 

I suspected that the autofocus improvement we've seen had more to do with hardware than firmware, but I listed both year of release and year of latest firmware upgrade for completeness. I also listed the basics of the camera's AF system, as well as any marketing comments made about it at release like the 7D having "the most advanced AF algorithms."

My thought going in was the difference would be in hardware not firmware. I don't think any amount of firmware is going to make the AF sensor on the lower left behave like the one on the lower right, even in center-point, single-shot mode.

 

The Results

In the lens variation article, we used the standard deviation (SD) to measure how accurately the camera focused: Softer focus results in lower Imatest values. As an example, we've put up a graph of Imatest values for the 5D Mk II versus the 5D Mk III below.

 As you can see, the 5D Mk III shots (red square) are all very similar. The 5D Mk II shots (blue diamonds) are more spread out: The variation in shot-to-shot focus is greater.

In this example, the SD of the 5D Mk III samples was 17 lp/ih, while the SD of the 5D Mk II was 38.5. Those are similar to the numbers we've seen over and over---accurate focusing combinations have SDs in the teens while less accurate ones have SDs in the 30s.

Rather than clog up the post with a lot more graphs, I'll list the SDs of the various cameras with the 28mm f/2.8 IS lens in the table below.


Camera SD
1DsIII 29
5D II 38.5
50D 34
7D 41
1DIV 22
60D 34
T3i 41
1Dx 17
5D III 17
T4i 29

 

It's a little confusing. There's a range of variation, of course. But clearly the 5DIII and 1Dx do better than the other cameras, while the 1D Mk IV seems to be a bit between those two and the rest of the pack.

It seems a little clearer to me if we graph the standard deviations and separate the cameras by type (more expensive at the top, less expensive at the bottom.) The oldest cameras in both groups are on the left, while the newest are on the right.

 

A couple of points are worth making.

The graph of the more expensive cameras seems to show a pretty logical progression. The 1Ds III is by far the oldest, having been around since 2007. But it included every possible AF technology of the day, including a separate AF processing chip. The 5D Mk II, even on its release, was known to have "consumer-grade" autofocus.

Despite my well-recognized modesty, I will also point out that when the 5D Mk III was first released, and Canon fanboys were dropping off cliffs right and left, I said "the 5D III is no minor-upgrade camera; it’s an entirely new camera using the old camera’s name". Its autofocus system is certainly not a minor upgrade--it's moved over to the big-boy camera side.

I had hoped the T4i might be more accurate than it was, at least with new lenses. It does seem more accurate than the other consumer / prosumer cameras in phase detection, but it's not nearly as good as the 1Dx or 5D III.

I assume that it's new hybrid LiveView system does not carry over to create phase-detection AF. I will say, in it's defense, that when focusing in LiveView it is obviously faster than and just as accurate as any of the other cameras, including the 5DIII and 1Dx.

So Why Could This Be?

All this autofocus stuff 1) gave me a headache and 2) made me rather curious and uncertain.

I started doing a simple demonstration of what I already knew: Phase-detection AF isn't as accurate as contrast-detection AF. But then I got results that indicated sometimes it is just as accurate as LiveView. But you have to have a certain camera and a certain lens or it doesn't happen.

This didn't make much sense to me.

I would have understood if each generation of newer cameras and lenses got a little better. Or if a new camera or lens was dramatically better. But why a rather sudden change, and why did you need both a new camera and a new lens?

I spent a fair amount of time emailing with Dave Etchells of The Imaging Resource  and SLRGear.com who was kind enough to bounce ideas back and forth with me. His thoughts sent me on the right track for figuring this out.

I started off by trying to find out how long it took for an autofocus improvement to go from idea to released-to-the-public. I found one answer in a Canon patent from 2003 describing combined phase / contrast-detection AF in the camera mated to a lens using a stepper motor.

Which we saw  . . .  oh, yeah. Now. With the Canon T4i / EOS-M sensors and new STM motored lenses.

One thing of interest in this patent application, tucked away in the background section, is the following statement:

. . . though the prior art indicates a method for realizing high speed and high precision of autofocusing at the same time, it does not accompany a lens drive control for realizing this and thus does not adequately realize [. . .] high precision of autofocusing.

In other words, back in 2003, Canon recognized that a high-accuracy AF system in the camera required a more accurate lens drive to yield precise autofocus. You can't have one without the other.

Suddenly the need for both a new camera and a new lens to get accurate autofocus began to make sense. But wait, the patent was talking about using a stepper motor to achieve accurate AF. We did find the 40mm pancake (stepper motor) was more accurate. But we also found that the new 24mm, 28mm f/2.8 IS and 300mm f/2.8 IS II lenses were more accurate. Yet they don't have stepper motors.

Then I read on a bit further in the patent:

In order to achieve this objective, this invention provides a camera system comprising: a first focus detection unit, a second focus detection unit, a stepping motor that drives a focusing lens, . . . or a rotation detector, which detects the rotation . . . of the motor. . . The control circuit performs closed-loop control, based on the output of the rotation detector to control the motor.

The wording of this patent, back in 2003, suggests that closed-loop was not how AF worked at that time. It was largely open loop. The camera took a measurement and told the lens where it should go. Done.

This is backed up by a lot of other information, including statements in Canon's "EF Lens Work III" and quotes made by a lot of people who would have known.

It's clear that later AF systems were closed loop (the camera double checked where the lens had gone), although exactly when and where that change occurred I don't know. But this probably explains the lengthy, and now largely meaningless, debates about whether systems were open or closed-loop: They used to be open. Now some (probably most) are closed.

More to the point, though, is the comment that a rotation detector would be needed in lenses driven by ultrasonic motors to increase accuracy of the focusing movement.

If this is the case, then the newer Canon lenses should definitely have a rotation detector built into them. We know there are rotation detectors in many lenses released after 2000, but if they are  in older lenses we can't identify them, so this fits too. (As an aside, I am particularly skilled in finding them because usually if you touch them with your fingers the lens won't focus anymore and the unit has to be replaced.)

 

From Left: A Canon and Nikon magnetic rotation detector.

 

Just to be certain, I asked Aaron to find an excuse to take apart one of the new lenses. Or failing that, I told Aaron to take one apart and not tell Tyler or Drew, who never really believe we can put them back together correctly.

The result: Aaron checked and the newer lenses did have rotation detectors.

There was one other big hint hiding in plain sight in Canon's resource article on precision cross-type AF sensors. The article states that the farther apart each pair of AF sensors are (each pair compares phase from opposite sides of the lens) the more accurate the sensor. It also states only the 5D Mk III and 1Dx have them. I italicized some of the quote for emphasis:

[...]simply by reading which pixels on each sensor line are being struck by light, the AF system can tell instantly what direction to move the lens in for proper focus, and by how much to move it. . . Canon EOS SLRs with high-precision AF sensors simply move the pairs of sensors much farther apart, and accordingly, the AF information can be more finely broken-down and reacted to.

The 5DIII and 1DX have f/5.6 cross-type sensors (most previous cross-type required f/2.8 or faster lenses). The central AF sensors in these cameras also have diagonal cross sensors. The article continues:

For pros who prefer to use the center AF point, simply manually selecting it [. . . ] gives the user the highest precision AF possible with these cameras (assuming an f/2.8 or faster lens is being used). And, with their unique diagonal cross-type layout, the AF points are much more likely to latch-on to typical horizontal or vertical subject details.

In Summary

As it turns out, my little tests just confirm things that were already published (if obscurely.)

The two newest Canon cameras have more accurate phase-detection sensors than their previous cameras. The newest lenses have more accurate focus movement (or provide more accurate focus movement feedback, or both) that takes advantage of those sensors.

Older cameras don't have accurate enough AF sensors to take advantage of the new lenses' capabilities.  Older lenses can't move their focusing elements with enough accuracy to take advantage of the new cameras' accurate sensors.

It's rather sad (given the amount of other work that's piled up during this little series) that I didn't find this information until after I'd done all this testing, but it's rather a testament to our times. There's so much marketing drivel and useless verbiage thrown about that the marketing noise drowns out the actual useful information the camera makers offer us.

But that's the topic of another blog post in which I will take personal offense at the marketing crap that's been shoveled our way at an ever-increasing rate.

I don't believe we photographers are nearly as stupid as the marketers seem to think we are. In the meantime, while I'm on this rant, I highly recommend Thom Hogan's very funny look at camera icons "A Different Kind of Focus." Thom writes better stuff sitting on the tarmac waiting for takeoff than I do at my desk.

And I guess, given all the marketing fluff, I wouldn't have really believed "better autofocus" if I hadn't seen it with my own tests.

I've heard it before and it wasn't so. Not to mention I doubt seriously we'd have seen the marketers say, "better autofocus...but not with the lenses you already have." So maybe this is worthwhile after all.

Otherwise I might have put my trusty old 85mm f/1.8 on a 5D III and thought "not better at all, they're lying to me again."

 

Roger Cicala (with thanks to Aaron Closz and Dave Etchells)

Lensrentals.com

August 2012

169 Responses to “Autofocus Reality Part 3B: Canon Cameras”

Peter said:

So, should I now feel a little less bereaved that I sold the following to help pay for my mk III?

17-40 f4
135 f2

Woody said:

Very interesting. Since it is such an outlier (both in expectation and in contradictory results) I would almost want to try some other tests on the EOS 7D to see just what it is actually good at (or which lenses it is good with), since it obviously isn't center point with the lens you were using.

Markus said:

It'l be interesting to see if the coming major FW ugrade for the 7D will make a difference.
Roger, do you have any idea on the accuracy of the 7D's AF sensors (at least, it also reads "center dual diagonal cross")?

btw: Great article - it's really sooooo difficult to separate marketing hypes from real advances. And most of us simply don't have the possibilities for these checks.

Ivan said:

Hello Roger, thank you very much for your article. I find it very useful and your findings are very close to my experience with Canon 7D.

Personally, I do not like AF of 7D. Yes, I got used to it, I use 7D professionally and make money with it, but too many odd OOF images for my taste. As you found out, the SD of 7D is great (not that good).

But the biggest disappointment for me was a 7D + Canon 100mm L macro IS combo. I thought with a newly built and quality L lens with a f2.8 aperture AF would be dead on every time. I was wrong. My 7D and 100mm can miss some obvious and "easy" situations with good contrast. The focus variation is large - sometimes it's dead on, sometimes it's close, sometimes it's off by a mile. And microadjustment doesn't help.
Don't get me wrong, I do get usable and extremely sharp images with this combo but I expected more precise AF, especially on f2.8.

The best results I get when I press AF-ON button several times before taking the image with very shallow DOF. It's slightly more precise than using af-servo. And you can clearly see how camera shifts the focus, even though both times the camera gave green light for AF.

So, I cannot wait buy MK3. Thanks again for your article!

Woody said:

Good point, Markus. The 7D was a test bed for so many focus technologies, perhaps now what Canon has learned can be applied retroactively.

Tobias said:

Very nice and informative articles, thanks!

Hello Tern,

you asked:

"It’d be interesting to see if the theoretical choice of ‘accuracy (if you have f2.8)’ (Canon) vs ‘availability (cross point with any lens)’ (Nikon/Pentax) actually matches to reality."

Matt and Jamieson, two "international award winning Toronto Wedding Photographers", made a little test.
It wasn't such an extensive testing, like with SD outputs and selection of the better performing cameras and lenses of a stack (just two cameras), but at least for me it yielded surprising and kind of diverse results, like the one on this site.

http://xerodigital.ca/canon-1dx-nikon-d4/

Greetings from Germany

Graham said:

Excellent articles, thank you. I've noticed a considerable improvement with my 5DIII and 70-300L combination. My walkaround lens is the 24-105, and it seems to behave better on the 5DIII than it did on the 5DII, but the improvement isn't as dramatic. As the 24-105 is almost seven years old now, perhaps it's time for an update with better AF?

Todd said:

I'd love to see you include a 1Ds MkII in the mix. I suspect the result would be surprising.

Peter said:

This has been a brilliant series of tests. Thank you so much guys. Can I blame you guys when my wife asks what on earth inspired the lens renewal program I will now have to start to get the best out of my 1D X?

LensRentals Employee

Roger Cicala said:

Richard that's a good suggestion. I'll see about working that in next time.

Ed said:

Brilliant work. I'm grateful that there are people like you who not only understand how to do proper testing, but are then able to carry it through and share the insights. I find proper testing procedures to be akin to crawling naked over broken glass, which is why I know better than to try to focus tune my own gear - those with patience will outperform me. Can do, have done, hate it.

I did do a little bit of comparison between three Nikon bodies - D800e, D700, and D300, triggered by wanting to know if my new D800e had the left sensor focus issues that some do. I did several dozen shots each with differences in focus rotation direction, target distances, zoom settings, and then clustering results, two different lenses. What I saw suggests Nikon has also made a breakthrough in the D800e, sample quality problems notwithstanding. The D800e's data clusters were much more compact, (and my body doesn't have the problem, but that news didn't make the process any less hateful for me.) My real world shooting seems to confirm that.

Bravo, Roger, for bringing some real data based insight to a topic usually smothered in emotional blather.

Martin said:

Yes, I second Todd's suggestion, add some version of 1D Mk2 camera to the mix, please. Personal function 16 "Camera shoots when in focus" would be also great to test enabled/disabled.

Raoul J said:

Hello Roger !

Thanks for your answer!

But if I come back to my
"3. If I understand you correctly, your results mean that the focus will actually be better on 1DX and 5DIII with lenses without feedback if you double click." : How would you explain this is not the case ?

If the old lens (the 50mm 1.4) is lacking some feedback to position correctly from the one order received from the new camera -one camera which AF measurement is accurate enough to perform a very accurate positioning -that's your hypothesis- then why does the accuracy NOT increase if you do it several time ?
If the camera measurement is accurate, it should get the lens closer and closer from the "correct" focus, no ?
Any idea ?

Gary said:

Thanks Roger. I love your down-home style.

Luckily I switched to Canon late (mid-2009) and
have moved up to the 5D3. The 135 L works
exceedingly well with the 5D3 in dark environments like casinos (candid photos of patrons). Yet it is an
"old in the tooth" lens, correct? The 135L did not
work anywhere near as well with the 5D2.

Andre said:

Hi Roger. You mentioned that you chose one copy of each camera to test. Don't you think you need to test several copies of each model in order alleviate copy to copy variation? Or, have you already done copy to copy testing to find how close the SD is for a group of cameras of the same model?

Eric said:

Great job! Glad you checked cropped bodies too. I wonder how much better the T4i is than my old XSI.

Alan said:

Based on what I read it was a bit of a disappointment to find out that everything I purchased (7D, 5DII, and 70-200L f2.8 IS II) in the last year are all outdated technology. Actually, after a trip back to Canon for my 7D (adjustments were performed) and my 100-400L f4-5.6 (adjustments were performed) the combo works fairly well. The 7D from the beginning did not perform AF well with that lens in particular and I was very frustrated as it worked great on my prior 50D. I also upgraded from a 5D to a 5DII to stay relatively current with technology although I'm not a trail blazer in camera technology as you can see. I use the 7D for action shots (mostly birds) and the 5DII for landscapes and the pair suit my purposes very well (after Canon tune up). I have never had issues with the 5DII and think it's a nice incremental update. I would not been able to justify the extra cost of the 5DIII regardless of how good it is. Based on your findings I was surprised to see the 7D produce less than stellar results even with its more advanced AF system. I did not see any reference to the 2.0 firmware update making any AF changes but only stuff to make it a better movie camera.

Jeremy Stephenson said:

Thanks Roger, This is a great series.

I would love to see a comparison of shooting one-shot vs. continuous AF on your target as well. Is that a possibility? Personally I prefer using continuous AF, shutter half-depressed, waiting to catch the moment as the subject moves around (for example candid expressions at a wedding as someone's involved in an animated discussion). However I wonder sometimes if that affects my keeper rate with static shots.

I have tried micro-adjustment with my Canon 7D's on various (f2.8 and L) zooms, but generally found the focus seemed be randomly situated, sometimes in-front, sometimes behind the mark on my lensalign calibrator. It didn't seem to matter if I used one-shot or continuous AF.

However, I would love to know if there is any decrease in accuracy using continuous AF on a stationary subject — even if it's just with the 5DMarkIII and the newer lenses. I wouldn't imagine it would increase the workload too much — just switch AF mode and fire off a few more rounds for each camera/lens combo?

K D Sandmann said:

Im thinking there are a lot of beaks opened wide eagerly awaiting Rogers regurgitation.

A funny or perhaps disturbing visual.
Chirp chirp.
:-)

James said:

What I don't understand is how the 7D has wores numbers than the 5DII. Can we challenge this call and check the instant replay? Also great articles I've enjoyed all of them even the technical ones.

Richard Hatch said:

I liken your work to the new age anatomists who actually dissected real bodies to determine how things work instead of relying on information that was quite old and not correct. Unfortunately, the masses over at DPreview are going crazy (because they are idiots)suggesting your data shows this that or the other and arguing this that and the other.... sheesh...

As a 7D and 5DmkII user and a handful of L lenses I'm grateful to have great equipment. Knowing how it works and what limitations it has is a bonus.

Richard

Ed said:

Wait - I think we're missing the big picture here... What I'd REALLY like to see tested now is what manual focus performance looks like for a relatively experienced shooter, without the magnification of live view, etc, and with a relatively short time window to focus in. Then you'd see whether the normal human being is going to be generally better off with AF or not. My own experience is that if I allow myself time for focus hunting - tweaking focus back and forth a few times and iterating in on the point of focus - I can beat the AF system. But not under normal real world shooting conditions with living, breathing, moving subjects. That would put the whole "better/worse bodies" into perspective, I think.

LensRentals Employee

Roger Cicala said:

Ed,
I'll need an experienced photographer who can manually focus accurately for that test. I'm not it! There's a reason I use Live View :-)

Roger

Tom said:

Great stuff - curious minds free from marketing chatter usually get the bottom of the pile. As a side note: in my line of business a closed loop system does something in addition to what i read your description to suggest, ie. here's your statement "It’s clear that later AF systems were closed loop (the camera double checked where the lens had gone)" - here's what it is in my industry "the camera double checked where the lens had gone, then double checked that it went to the place it was supposed to go - and if not, then repeat the whole process, if yes then take the darn picture". I'm not being critical, just think this is an important nuance (that maybe is obvious to you and thereore could have gone without mention).

Thanks for putting so much work into this. But makes me want to upgrade my 7D.

Stephen Feingold said:

Roger,
Do more recently manufactured older lenses have improved focus mechanisms?

Marcus said:

Very interesting.

Nikon NZ have lent me a D4 to test drive and I have to say the AF is quite astonishing even compared to the D3s. They increased the RGB sensor from 1005 pixels to over 900,000 and it seems to have zipped things up no end.

Oddly, I was test driving a D800E a week ago and - although the AF is supposed to be the same module - it was not as good as the D4. I've always thought there is a 'secret sauce' in Nikon's pro bodies and maybe that is the case here. It's even very fast with older Nikkor screw-drive AF lenses (such as the AFD 85mm f1.4), much less the AFS ones with ultrasonic motors etc.

Certainly I have almost concluded that the D4 is what I will be buying as I think it is a more useful tool to me than the D800, which for (mainly) subjective reasons was not good for me. I also think that 36Mp is just too much for my work and 16 is a great compromise between speed, high ISO ability and being able to crop a bit without loosing significant resolution.

Scott McLeod said:

As so many others have said, thanks so much for this info! The back-of-the-pack position of the 7D is interesting (and unfortunate) and probably goes a long way to explaining a lot of the AF "issues" that 7D owners complained about. It also probably explains why that AF sensor has not been seen in any other body since.
I know that my 5DIII will AF cleanly and accurately in light so low I can barely see what I'm shooting (which still amazes me but backs up the stated AF spec of going down to EV-2) while the 7D with the same lens will be flashing and sizzling away with little to show for it.
I am keeping my fingers X'ed that the v2 7D firmware will include some AF improvements in addition to the hugely increased buffer (which is my main interest).
Keep up the great work!

Ed Gill said:

Roger

Repeated thanks to you for your hard work and service to the photo community. Just more evidence to keep me turning to mirror-less systems with fast contrast detection. My old panasonic G1 is far more accurate and consistent then any of my Canons through the 50D. Disappointed but not surprised about the 7D - I'll pass. The 5D III, maybe, but what is not really addressed is wear and tear and tolerance creep as flopping mirrors, AF sensors, and lens motors vibrate out of tolerance. The inherent problem with Phase detection is that it uses "Dead Reckoning" to guess where in space the lens is putting the "too often curved" focus plane and where the very flat sensor surface is. Somewhat akin to the old need to shim focus screens to match light path distance to the film surface. Since contrast detection uses information from the sensor itself no "calibration and calculation" is needed - simply a more reliable system. I would guess, maybe one more generation of floppy mirror cameras (tight tolerances are expensive to produce) then the world will turn EVIL. Best Regards, Ed

Dan Tong said:

Here is something that might interest you:

Here is an abstract of research with defocus and how that can be used for focusing on natural images, and suggests that this method will soon be developed and used in cameras.

http://www.pnas.org/content/early/2011/09/16/1108491108.abstract

CarVac said:

It's interesting how the 50D and 60D performed the same, (they have the same AF sensor), but the T4i did better with the same exact AF sensor.

Raoul J said:

Hello
I again try to understand why the double click method doesn't increase accuracy with 1DX and 5DIII.
I own an old 1DII and 50mm 1.4, 85 1.8, 70-200 2.8L IS (and more)...
What I know by experience is that the 50mm is -by far- the worst of all when it's about focus accuracy. Where does it come from ? Possibly the type of focus motor -which is different than the others. Possibly an old AF chip inside the lens. Possibly both.
So I was wondering if the results you got was relevant of all USM lenses or just the 50mm 1.4.
Because, once again, if your hupothesis are right, you SHOULD see an improvement with old lenses and 1DX/5DIII when you double focus...
What do you think ?

mark lowden said:

Fantastic piece of work! Non subjective and analytical. I would love to see something like this in the camera press for all the new kit that arrives - we don't (thats why I have just about given up buying it) and rely on Roger and Thom for some clarity. Why?? Love to see similar on the Nikon line up I do a lot of action stuff (surf, aircraft) and auto focus is the only way to fly (for me anyway).

Kevin said:

Roger, while the AF on the 5DIII is clearly more precise and consistent than the 5DII, is there any way to translate your results into inches or millimeters of error? For example, if the 5DII is off (on average) by 2 millimeters while the 5DIII is off by only 1 millimeter, the difference may not be very significant from a practical perspective. However, if the difference is 10 and 20 millimeters, it could be very significant. In both cases, the difference is 100%, but its less significant if the numbers are very small to begin with.

Sérgio said:

Thanks for the wonderful work of putting into numbers something that some of us have been wondering for such a long time.
From my short experience I agree with you that AF accuracy is a function of lens actuator (and motion measurement system) and phase detection sensor (and control algorithm) but in addition I think the lens sharpness / contrast at the wider aperture might also change the AF overall performance.
Hopefully photography gear reviewers will start to measure and rate the AF side by side with resolution, dynamic range and such. The big question that is left open (and I hope you will be able to answer) is: how do other system compare accuracy wise with the one you tested?

Sérgio said:

Thanks for the wonderful work of putting into numbers something that some of us have been wondering for such a long time.
From my short experience I agree with you that AF accuracy is a function of lens actuator (and motion measurement system) and phase detection sensor (and control algorithm) but in addition I think the lens sharpness / contrast at the wider aperture might also change the AF overall performance.
Hopefully photography gear reviewers will start to measure and rate the AF side by side with resolution, dynamic range and such. The big question that is left open (and I hope you will be able to answer) is: how do other system compare accuracy wise with the one you tested?

fem2008 said:

Thanks, Roger. I always enjoy your reports and insights. Just a couple of comments regarding your conclusion about closed loop Vs. Open loop focusing. Just because the patent says “The control circuit performs closed-loop control, based on the output of the rotation detector to control the motor”, it does not mean that the Autofocusing system is closed loop as well. In fact, you could have two separate closed loop controls, one for the focusing, and one for the motor alone. As an engineer, I have done this with control systems. You can have a loop within a loop. So I don’t think that this can prove if the Canon system was open loop before and now is closed loop or anything different. This is not the point here. There is evidence from other manufacturers such as Minolta that it has been closed loop for long time, but I cannot speak for the Canon.

I think this new information relates more to how to achieve faster focusing with better accuracy at the same time. In order to drive a motor fast, you can use course steps or drive pulses. But course steps are not that accurate. You can drive a motor in half steps, and quarter steps, etc., to achieve ever increasing positional accuracy, but that slows it down. You can drive the motor at high speed using coarse pulses, then slow down and use finer pulses as you get closer to the point of focus, but how do you know when to switch between the two speeds (closed loop autofocus, open loop motor control situation)? What if your optimal point of focus is between two control pulses? You can’t send 10.5 pulses. This is where the encoder can improve things. The encoder can have a much higher resolution than the drive circuit pulses, and can inform the drive circuit to switch to finer drive control in mid pulse (if the auto-focus sensor senses ti needs to), in order to achieve optimal focusing. The encoder can also account for slip in the motor. With this system, it truly becomes closed loop focusing and closed loop motor drive.

By the way, just relying on the number of rotations of a motor or focusing element alone still does not constitute closed loop focusing, even if an encoder is used to verify the right amount of turns. Because, this assumes that the focusing helicoid is made with enough precision to properly position the element exactly where in needs. Even if that was true, with wear and tear, the lens focusing accuracy would drop over time. With a true closed loop focusing system, the helicoid accuracy does not matter (as long as you don’t have excessive slop or friction).

Jeff said:

Here's a blunt question.

I understand from the article that in order to take advantage of the 5D MKIII and its improved focusing system, then you need the newer revision to popular lenses (e.g. 24-70L 2.8 MKII). This leads me to understand that I won't see a difference with my (now) older 24-70L 2.8 MKI on a 5D MKIII body.

With that said, say I want to upgrade from my APS-C sensor body to a full frame sensor body. Would it be better to save the $1500 and get one of the remaining 5D MKII bodies over the 5D MKII? Unless there is an unfortunate accident, I don't plan on upgrading my 24-70 f2.8 for a while. Is there any other reason to get the 5D MKIII over the MKII?

LensRentals Employee

Roger Cicala said:

Jeff,

Right now there aren't (apparently) a lot of lenses available in the 'new' category. But this was just a lab test: from what I see and hear from others in the field the 5D III is more accurate in general. I certainly have found it so, with my current 70-200 IS II and several 'older' primes. I can't say if that's worth the money to you or not. The 5D II is still a very good camera and the price difference will be coming down if you wait a while.

Roger

LensRentals Employee

Roger Cicala said:

Thank you Farid. That makes a lot of sense and fills in some blanks very well.

One of the best things about writing articles like this is a lot of people take the time to broaden my knowledge in the comments, and I do appreciate it! As do all of the others who read them.

Roger

Alan said:

Sounds like some 7D owners are hoping the new firmware will address the AF issues described ad nauseum. Based of the Canon update info it seems like most of the changes are either for movie makers or add some useless on camera functionality that most shooters like myself don't care about. I take pictures with my camera and make pictures with my computer.

Arthur said:

very interesting - thanxs a lot for that great efford. Could you write or say something about using Sigma/Tamron for example a Sigma 120-300mm 2.8 OS on Canon Cameras 5d2 / 5d3? For sigma/tamron the whole reeingenering must be a nightmare - but i am seriously interested in that lense - there is hardly a alternative in the Canon Catalog. Your article makes me really thinking about it. ( i do not have $$$$$ to find out) also i am from switzerland, no returning in if you don't like - and very high prices compared to the US)

Tim said:

What is the possibility that mirror-flop and shutter vibration could influence these results? What shutter speed were these shot with? Obviously for the 1DX/5DMKII on a 300mm f2.8 it looks like mirror-flop and shutter speed had no effect...but what if that is a factor with other bodies?

LensRentals Employee

Roger Cicala said:

Tim,

It's always a possibility, but the only difference I could imagine would be in a group between full=frame and APS-C, I think. The target has standard lighting and all shots are at ISO 400 so shutter speeds vary a tiny bit by focal length, and a lot by aperture, but slowest are around 1/400 and fastest about 1/1000. In theory that should be out of the range of significant effect.

Roger

LensRentals Employee

Roger Cicala said:

ARthur, I don't think this little study would show anything about the Sigma 120-300. It's not the fastest autofocusing lens, that's for sure, but it does bring a lot of advantages to the table.

Roger

Edwin Herdman said:

Something odd I noticed after the "contestants" header in blue:

"We know the Canon 28mm f/2.8 IS II had accurate autofocus on the Canon 5DIII so we chose one copy to be our test lens."

28mm f/2.8 IS II?

LensRentals Employee

Roger Cicala said:

Well, I guess it was a "II" in my own mind. . .

Gozo Kiona said:

What's the point of putting graphs of everything with numbers, with no units at all? "Yes, the car was running at 1902",... feet per hour? Very non-technical for a dslr review...

LensRentals Employee

Roger Cicala said:

If you actually read the articles, you'll see the graphs are in Line Pairs / Image height.

Leave a Reply