TransWikia.com

What does "number of AF points" mean in scope of Canon's Dual Pixel AF

Photography Asked by Ed.ward on January 13, 2021

According to the Canon’s Dual Pixel AF system paper,

"EVERY pixel on the sensor can provide AF information [skipped].
Figuratively speaking, instead of hundreds of potential "focus
points," DPAF is a system with millions of focus detection points"

but in tech specs of cameras there are still "number of AF points" is presented. For example EOS RP has 4779 AF points, whereas EOS M50 has only 143 (according to dpreview site). Why so, if every pixel on the sensor is the "dual pixel".
One would assume, that it is because CPU of M50 can’t handle more points, but both these models have exact the same DIGIC8 CPU.

Could anyone provide me with a reading, what they mean by "number of AF points" in scope of DPAF? Didn’t find good explanation myself :/

One Answer

The term AF "point" has always been a bit of a misnomer. Rather than a single discrete point with no dimension, all AF systems have areas or zones of sensitivity that they call "points".

  • This is the case with dedicated PDAF sensor arrays, which use microlenses to direct light from certain parts of a lens' image circle to pairs of lines on the AF sensor. Each line is either one relatively large "pixel" or several smaller "pixels" wide and many more "pixels" long.
  • This is also the case with imaging sensor based CDAF and hybrid PDAF/CDAF systems. Each AF "point" uses multiple photosites (a/k/a sensels or pixel wells) to measure contrast within the area of sensitivity.

In both cases the camera is usually programmed to bring into focus the area of strongest contrast anywhere within the entire selected area of sensitivity. For many cameras and systems from all the major camera manufacturers, the areas of sensitivity for each "AF Point" are far larger than the little squares, rectangles, or dots that you see in the viewfinder. Often these "points" share different, overlapping sections of longer lines on a dedicated PDAF sensor array or share overlapping areas on the main imaging sensor.

With Canon sensors that have Dual-Pixel AF each "pixel" on the sensor is actually two photosites.

enter image description here

For a single DPAF pair to be able to measure contrast, the line between the darker and lighter part of the scene must be exactly over the line between each side of the Dual Pixel pair. If the line between the black line and the white line on a test chart is to the left or right of a specific Dual Pixel pair, then both sides will be either in the white or black area and both sides will measure the same amount of luminance, thus that one pair will not be able to detect any contrast. So even Dual Pixel AF uses groups of pairs in an area or zone to detect contrast anywhere within that area or zone.

The first reference cited in the question points out that each "pixel" on a Canon DPAF sensor is actually two photosites capable of detecting contrast by comparing the luminance measured by each. Each pair is capable of detecting contrast, but only if the edge of the contrast passes over that single pair in the right orientation.

The camera specifications mentioned in the question approach it from a more practical angle: It tells how many areas or zones of multiple photosites can be selected by the user as areas of sensitivity. The end user does not have (nor need) the capability to designate a specific DPAF pair. The area of sensitivity for a single pair would be much too small to be practical and would only lead to frustration that the camera would rarely, if ever focus on anything unless it could be pointed at a target with absolute precision measured in small fractions of seconds of arc. For various cameras the engineers and designers at Canon (or Sony, or Nikon, or Fuji, etc...) decide how the sensor is divided into groups that are then called "AF points" selectable by the user.

One thing that contributes to decisions about how many "AF points" to divide the sensor into is the processing capacity of the camera within which the sensor is placed. A camera with lower processing capacity will typically be divided into fewer zones that each cover a larger percentage of the total sensor area, while a higher end camera with more processing capacity will typically have more zones that each cover a lower percentage of the total sensor area. When balancing processing capacity between things such as AF, metering, and image processing other factors can also come into play. Sensor resolution is one such factor. A sensor with fewer total photosites requires less processing capacity than a sensor with more total photosites to convert the raw data to a viewable raster image such as a JPEG or HEIF file. Battery capacity is another. Smaller cameras typically use smaller batteries with lower total capacity than larger cameras that typically use larger batteries with more total power capacity.

Beyond that, I'm not convinced that when Canon specifies 4779 AF points for the EOS RP and DP Review lists 143 "AF Points" for the EOS M50 they're both talking about the same thing.

  • DP Review is talking about the number of user selectable areas available in the AF menu with the M50.
  • The same spec for the EOS RP is also 143 areas.
  • To use the 4779 "AF Points" with the EOS RP one must use the cross keys to navigate a pointer around the EVF or rear LCD screen.

In all likelihood, each of the 4779 "points" have total areas of sensitivity many "pixels" wide that overlap with the "pixels" used by their neighboring "points". The 4779 number is derived from the total possible number of places where the area of sensitivity can be centered. As one scrolls across the viewfinder the center of a multiple number of DPAF pairs is moved and pairs on the edge of one side of the cursor are dropped off as pairs on the edge of the other side of the cursor are added.

Think about how the arrow for your mouse cursor moves across the screen of your computer. The arrow is much larger than one set of RGB pixels wide and tall. If your hand is precise enough, you can move the mouse up, down, left, or right by a single pixel. The new position uses a different set of pixels, but only the ones on the edges of the arrow change. The same number that are dropped off on one side are added on the other side.

Correct answer by Michael C on January 13, 2021

Add your own answers!

Ask a Question

Get help from others!

© 2024 TransWikia.com. All rights reserved. Sites we Love: PCI Database, UKBizDB, Menu Kuliner, Sharing RPP