Canon EOS DSLR Autofocus Explained
by John Reilly (neuroanatomist)
With my outset SLR, I placed my bailiwick in the middle of the frame and turned the focus band until a feature in the split prism lined upward and the microprism collar lost it sparkly look. Today, things are a lot more complex and technological, and the number and diverseness of unlike terms used to characterize AF systems is a little bewildering. Hopefully, this article will clear up some of the confusion.
Phase observe AF
For this word, I’ll be talking only near phase-detect AF – the rapid method used in dSLRs involving a separate AF sensor. The other method, contrast-discover AF, operates in Alive View and is also found on P&S cameras. Although dissimilarity-detect AF has some advantages, it is essentially slower than phase detect AF, which is the main reason the latter is the primary AF method for dSLRs. For phase detect AF, the primary reflex mirror, which directs most light up to the viewfinder, passes some light through a semi-transparent window to a secondary mirror which directs the light down to the AF sensor at the bottom of the mirror box:
Earlier striking the AF sensor, the light passes through an array of microlenses which separate incoming low-cal to produce a pair of images that fall on each point in the AF sensor. A simple line sensor then measures the distance between those two images to decide if the image is front- or back-focused, and by how much. In essence, each AF bespeak is operating as a simple rangefinder.
That’s the simple instance, only modern systems are much more circuitous – and there’s a lot of jargon that make comparisons hard. What practise things like ‘high precision’ and ‘dual-line zigzag system’ really mean?
Characteristics of AF systems
The main characteristics of AF systems tin can exist broadly classified as:
• Number of points
• Geometry of points – basic and complex
• Coverage area
Number of points is cocky-explanatory. Geometry of points adds a layer of complexity, but both number and basic geometry are ordinarily part of the top-line spec for an AF system (e.g. the T3i/600D has 9 AF points with ane cross-type, and the 7D has 19 cross-type AF points). Complex geometry includes things like dual-cross points and dual-line zig zag arrangements. More on that later. Globally, accuracy is how ‘close to true’ the AF system gets, and Catechism doesn’t offering any specifications for this characteristic. Note that global accuracy is affected (and hopefully, corrected) by AF microadjustment. Locally, geometry affects accuracy, because a cross-blazon indicate has a amend take a chance of achieving proper focus than a single-orientation bespeak, since information technology samples features with different orientations.
Precision is repeatability – if yous accept several shots of the aforementioned thing, how close will the focus of any one shot be to all the other shots? For EOS AF points, in that location are two levels of precision – ‘normal’ which is within 1 depth of focus for the attached lens at max aperture, and ‘loftier precision’ which is inside 1/iii of the depth of focus for the fastened lens at max aperture. Annotation that depth of focus is related to, but not the same as, depth of field, although the same factors influence both.
Overall sensitivity is how well the AF system performs in low light. The AF sensor is composed of multiple 48-bit line sensors and associated amplifier circuitry – the more amplification (within the limits of point to dissonance), the less light needed to focus. AF sensitivity is specified as an EV range, and the lower the first number, the meliorate. For instance, the T3i/600D through 5DII can AF down to -0.v EV, the 1D IV/1DsIII down to -1 EV, and the new 1D 10 down to -2 EV. EV units are ‘stops’ and so the 1D X tin achieve AF in half as much light as the previous ane-series bodies. Sensitivity is used in some other context also, associated with lens aperture (east.grand. an f/5.half dozen-sensitive AF point) – in that case, it does not refer to the amount of low-cal (sensitivity is perhaps not the ideal give-and-take).
Speed is another characteristic for which Canon offers no metrics, probably considering there are too many variables. Merely better AF sensors and better processors result in faster AF performance.
Coverage expanse is a very of import gene – the broader the area of the frame where there are AF points, the more than likely that an AF signal will fall on your subject. More than on this later.
Of the above characteristics, number, sensitivity, speed, and coverage area are fairly straightforward. But geometry, accuracy, and precision are more than complex, peculiarly because the lens mounted to the camera will change the way in which the AF points operate in terms of geometry, accurateness, and precision.
Lines, crosses and double-crosses
A basic, single line AF betoken can observe contrast only in one dimension – the dimension ‘opposite’ to the orientation of the line. Remember that split prism in the manual focus SLR? The ‘split’ was horizontal, so you had to await for a vertical characteristic to focus on for the prism to be effective. So, a horizontally-oriented line sensor volition notice vertical lines (like a flagpole or the side of a door frame), while a vertically-oriented line sensor will discover horizontal lines (like a horizon or a boat on the h2o). There’s some confusing terminology here – a “vertical line sensor” is the same as a “horizontally-sensitive line sensor” and vice versa. A ‘dual-line zig zag’ arrangement (plant on a few points in some xxD bodies and the 7D, and on all the points of the 1D 10) means a line sensor that’s actually ii parallel lines instead of merely a single line, and the pixels in those two lines are offset by half a pixel, meaning the point of maximum stage alignment tin can be more accurately determined because it volition fall on a pixel in one line or the other, whereas with a single line it could autumn betwixt two pixels).
Generally, an discontinuity value is associated with an AF line sensor. The terminology usually used is “f/number-sensitive”, eastward.thousand, you may have an f/5.vi-sensitive line sensor, or an f/2.8-sensitive line sensor. The f/number refers to the maximum aperture of the lens, considering AF is performed with the lens wide open (i.e. the aperture you choose for the shot does non matter, only the max discontinuity of the lens). The use of ‘sensitivity’ in this context implies that light levels matter, because that’southward what we think of when nosotros normally utilise f/numbers. In this case, though, a wider discontinuity merely means a wider baseline for the rangefinder system is required for that line to function. Personally, I remember amend terminology might be to utilise threshold instead of sensitivity, so an f/2.eight-threshold line would require an f/ii.8 lens to part, and if you lot mounted an f/4 lens, that sensor line would not operate. An f/five.6-threshold sensor would work with any lens having a max discontinuity of f/5.half-dozen or wider.
Note that these thresholds are not absolute – a lens with a narrower aperture than the threshold might still piece of work, only at reduced effectiveness, accuracy, and speed. Thus, Canon limits the functionality to the rated aperture for a given AF sensor. However, some tertiary party lenses (e.thousand. Tamron and Sigma zooms with a max discontinuity of f/6.three at the long finish) effectively trick the AF system into thinking there’due south an f/five.6 lens attached. Likewise, although not condoned past Canon, it is possible to use record to block some of the contacts on a Canon 1.4x extender used with an f/5.6 lens, resulting in the camera attempting to autofocus with an f/viii lens on bodies which are express to f/5.half dozen. Sometimes, it even works…
All EOS bodies accept f/5.6-sensitive sensors, and thus will work with any Canon EF or EF-Due south lens. Some ane-serial bodies have an f/8-sensitive sensor at the eye AF signal, enabling them to autofocus (properly, and with official support) with an f/5.6 lens plus ane.4x extender or an f/4 lens plus 2x extender – a significant benefit for users of supertelephoto lenses. Notably, that feature is not included in the specification for the 1D X, which is express to f/5.half dozen lenses for autofocus.
An f/2.8 sensor line is more than accurate than an f/5.6 sensor line – the wider the aperture threshold, the wider the rangefinder baseline for triangulation and thus, the more accurate the measurement of focus. Still, the wider the aperture, the fewer lenses that piece of work with that aperture (and the more expensive those lenses are), and also, the detection range of f/2.8 sensors is narrower, meaning it may take longer for an f/two.viii line to achieve a focus lock when a subject area is well out of focus. As a result, AF systems will normally focus in two steps when possible – ‘fibroid’ focus with an f/5.vi line, then ‘fine’ focus with an f/2.8 line.
A cross-type sensor is a horizontal line sensor and a vertical line sensor at the same AF point, meaning that point is able to notice lines in both orientations – that makes it more than likely that the AF indicate will be sampling a feature that has the correct orientation to actuate the sensor. Some cantankerous-type points have the same threshold for both sensor lines, e.grand. the 9 points on a 50D/60D and all 19 points on a 7D are cross-type points with an f/5.6 threshold for both orientations. That’s skilful, in that all the points operate equally cross-type with whatsoever Canon lens, but the trade off is lower accuracy than f/2.eight lines. Some cross-type points only function every bit cross-blazon points when used with lenses that have a sufficiently wide maximum aperture, and if used with a lens with a narrower max aperture, they function as single line points but (usually vertical line sensors). This is a partial workaround to the trade-off mentioned above. A cross-blazon sensor with an f/two.8-sensitive horizontal line and an f/v.6-sensitive vertical line will function as a cantankerous-type sensor with an f/2.8 or faster lens, but if you use a slower lens, you’ll still take a functional AF point (just with only single-line orientation).
A dual-cross point is an even more complex AF sensor element. The cross-type sensors described in a higher place are a vertical line and a horizontal line, i.e.in the shape of a ‘+’. Dual cross-type points add together some other cross-type point in a diagonal orientation, i.due east. an ‘x’ that is superimposed onto that ‘+’. All of the implementations of dual-cross points to date combine an f/2.8-sensitive diagonal ‘x’ with an f/5.six-sensitive orthogonal ‘+’. So, with a lens slower than f/two.8 you get a standard f/five.vi cross signal, and with an f/two.8 or faster lens, yous get the increased accuracy of an f/ii.8 baseline, and with the ability to detect lines in multiple orientations.
Precision, invisibility, and other intangibles
As mentioned above with the precision give-and-take, in that location’s a modified type of AF point chosen a ‘loftier-precision’ bespeak, which focuses inside 1/three of the depth of focus of the lens at max aperture, vs. the normal precision spec of within ane depth of focus. Usually, the high precision point is the middle bespeak, and it operates in high-precision mode with an f/ii.viii lens on most bodies, or an f/four lens on 1-series bodies. The 1D X is an exception in two ways – it has five loftier-precision points in a central vertical column, instead of merely one, and they crave f/2.viii unlike previous 1-series bodies.
On some 1-serial EOS bodies, certain AF points are ‘assist’ points, pregnant that they cannot be selected when manually choosing a focus signal. Even so, they volition be automatically selected by the AF arrangement, either when next to a selected AF betoken with an AF expansion mode enabled in Ane Shot mode, or when needed for tracking a moving subject in AI Servo AF mode. The 9-bespeak AF organization found in the 5D and 5DII also has vi assist points, but unlike the 1-series assist points, these are ‘invisible.’ They are used in AI Servo AF mode to help with subject tracking, but are not available at all in One Shot way.
Another feature worth mentioning is extreme defocus detection. When nothing is even close to correct focus, it’s a challenge to the AF system, which mostly acts by refining an image that’s somewhat out of focus. Older AF systems just racked the focus back and forth until some feature(s) came into shut enough focus to activate the sensor lines. Newer sensors employ ‘farthermost defocus’ sensors to point the AF organisation in the correct direction with a preliminary gauge of the direction and magnitude of the alter needed to get close to correct focus. These are actually the same sensors every bit the ‘dual-line zig zag’ sensors described above – that configuration increases accuracy, but also allows improve defocus detection by ‘summing’ the kickoff lines into a single readout. Importantly, that extreme defocus detection uses combined data from the dual-line zig zag sensors. On xxD bodies and the 7D, three vertical points have the dual line vertical sensors (center summit, centre, and centre bottom), and data from those 3 inputs drives the extreme defocus detection. On the 1D X, all of the vertical lines (i.due east., all 61 AF points) have the dual line zig zag arrangement, and thus the whole AF sensor acts as one big extreme defocus detector.
Separate from the number and characteristics of individual AF points is the ability of those points to work together to rails a moving subject in AI Servo AF fashion. Because f/five.6-sensitive points generally acquire focus faster than f/two.8-sensitive points, the AF system will often rely on them for tracking data. However, overall tracking performance depends on many factors outside of the AF sensor itself, most notably the Digic processor and the algorithms driving the movement prediction, as well as supporting data integrated from other sources (e.g. the metering sensor on some EOS bodies).
Putting it all together
That’s a lot of features, and many of them are combined in various ways to etch an AF sensor. Here’s an example how it all comes together, in the AF sensor layout of the 7D:
Note that this is the entire AF sensor, fifty-fifty though at first glance it appears to be but one big cluster with 2 smaller clusters at the sides, such that you might be tempted to think information technology’s the center AF point and the two on either side. But in fact, you’re looking at 20 AF points (19 ‘+’ f/5.6 points and one ‘ten’ f/2.8 indicate superimposed on the center ‘+’ point). The way to interpret this diagram is that each adjacent pair of lines in a horizontal orientation, e.grand. — —, represents the horizontal sensor lines of ane to three AF points (longer lines contribute to more AF points), each adjacent pair of lines in a vertical orientation represents the vertical lines of one to v AF points (longer lines contribute to more than AF points), and the pairs of unmarried diagonal lines represent the f/2.8 ‘ten’ cross-blazon point.
Here is the actual AF sensor for the 1D X. The set up of five f/ii.viii diagonal crosses stands out pretty conspicuously.
When looking at the diagonal f/2.8 sensors, it’s credible that they are much more widely spaced than the f/5.6 sensors – nearly to the edges of the chip. This accounts for the longer baseline that results in greater accurateness than the f/five.half-dozen sensors.
AF Point Coverage
For many people, this is a big issue in comparing cameras. While it would exist wonderful to have AF points available over the entire extent of the frame, There are technical limitations on the spread of the AF points – at best, they tin can only occupy the center surface area of the frame, considering of simple geometry and eyes. In a nutshell, there are iv reasons for this limitation:
- Size of the secondary mirror. Light for AF passes through the semi-transparent part of the main mirror (most is reflected up to the viewfinder), then is reflected off the secondary mirror down to the AF sensor. At that place is limited space behind the main mirror, based on the necessary geometry (i.e. the principal mirror has to be at a 45° bending to the incoming light, and the secondary mirror has to be behind the chief mirror and at an bending of ninety° to the main mirror, so it’southward length is limited by the distance between the main mirror and the paradigm sensor).
- Distortion. With many lenses, the edges of the frame are subject field to distortion (barrel/pincushion), and that reduces the accurateness of stage detect AF.
- Vingetting. The AF system needs a certain corporeality of light to work. Almost all lenses vignette to some degree, meaning at that place might not exist enough low-cal at the edges of the frame. For example, the EF 17-40mm f/4L has >2 stops of vignetting wide open at the wide end – that ways at the edges of the frame, you’re below f/5.half dozen and AF sensors would not have enough light to operate (i.eastward. in dim light yous’d exist below the EV sensitivity of the sensor).
- Temperature. Canon has stated that larger AF sensors are more susceptible to changes in temperature with the consequence that they alter size, getting either larger or smaller as the temperature rises and falls. That reduces the accuracy of the AF organization overall.
It’southward worth noting that none of these limitations apply to dissimilarity observe AF, and so using LiveView yous can autofocus correct out to the edge of the frame.
Then, AF sensors are express to the key portion of the frame, but some bodies offer more extensive coverage than others. A mutual question is, “What does the relative spread of AF points wait like?” This prototype comparison below shows the AF indicate coverage of recent xD bodies (7D, 5DII, 1DsIII, 1D 4, and 1D Ten):
The 1D IV and 7D are about tied for the broadest spread in both dimensions (relative to frame size). The 1D X has the same lateral spread as the 1D 4 and 7D, and slightly less spread in the vertical dimension (about half a row shorter). The 5DII is in bluish, and you can see the narrow spread relative to the other xD bodies (the horizontal extent of the 5DII’southward AF points is the aforementioned as the 1DsIII, but the vertical extent is much less).
In addition to the area covered, density of points is also important. A more densely packed assortment of AF points will gives more options for selecting a focus point. More importantly, a more densely packed array delivers substantial improvements in AI Servo AF tracking, because subjects pass from i bespeak to the adjacent more quickly – less lag and more data to bulldoze the predictive algorithms.
Hopefully, the above information is helpful in evaluating the characteristics and relative strengths of Canon EOS autofocus systems. Although the AF system is only i feature of many to consider when comparing camera bodies, for many people information technology’southward an important one. Historically, Canon has differentiated the levels inside the EOS lineup partly by using different AF systems, with AF performance and particularly AI Servo AF tracking getting progressively better as you move from the xxxxD bodies upwards to the 1-series (with the 7D as an outlier, with the best AF system outside of a one-series torso). Keep in listen that just like image quality is almost more megapixels, AF functioning is about more than the number of AF points.