Re: Near and Far Focusing
Gerry Winterbourne wrote:
Bright Angel wrote:
I'm sure you know this first bit but it's here for completeness. A lens can focus perfectly at only one distance; everything nearer of further is out of focus to some extent. The amount of out-of-focus blur depends on how remote things are from the focus plane (among other factors - see later).
Our eyes can't resolve extremely fine detail. As long as the amount of out-of-focus blur is no worse than what our eyes can resolve things look OK - "acceptably sharp" is the usual phrase - they are within the depth of field.
Note that the app you use is wrong - it describes the DOF as "in-focus", which it isn't except at the focus plane.
But while our eyes can't resolve the fine detail when looking at a picture at typical size and distance, the lens/sensor can often resolve finer detail. The result is that while in ordinary viewing conditions things at the limits of DOF look as sharp as things at the focus plane, the resolution is likely to be noticeably better at the focus plane when the image is greatly enlarged. When I said "see later" for other factors this is an important one of them.
Below is a 1:1 screengrab of 2 unprocessed shots (part of a exposure and focus bracket set).
The left image was focused just before the lectern, on the top step (see small image top left for the overview) at a distance of 1.56m (5.11ft) (according to the software). The right image was focused on the carpeted steps, at a distance of 6.16m (20.2ft).
Both were taken at 24mm (full frame camera) f11 at 10secs.
Why should the image focused at a further distance, be more in focus far to near (steps to book), than the image that's near to far?
Both shots have a range of resolution from near to far; starting relatively low, increasing to maximum at the focus plane and then falling again. For the shot focused at ~6m you see what you'd expect - the text in the book is within the DOF but not perfectly sharp at the size you show. The shot at ~1.6m ought to be at least as sharp on the text but it isn't; this suggests that the camera moved very slightly during the 10s exposure: this isn't anything to do with DOF; but the far steps being blurred is, of course.
I always thought you got 1/3 in front of the point of focus and 2/3 behind it.
No. That's a gross simplification that is far more often wrong than right. Look at the hyperfocal distance shown in your app. One significance of the hyperfocal distance is that if you focus at it the near limit of DOF is half the hyperfocal distance (about 0.9m here) and the far limit is at infinity. In other words, rather than 1/3 in front you get 1/infinity = (as a proportion) zero.
Info from a Hyperfocal App. Which I didn't use at the time of shooting.
Here are a couple of graphical illustrations of what Gerry is saying. As Gerry noted, only things that are at the plane of focus are "in focus", or as sharp as the lens can make them. At that distance, any given point on the object would appear as a point in the image (or as close to it as the lens can make it). The image of objects nearer and further from the plane of focus will be made up of progressively larger "blur circles". If these blur circles are smaller than what we can perceive in the final print (at a certain size and viewing distance), then we still see them as "acceptably sharp" and they are considered within the depth of field. But at a certain point the blur circles become large enough to be seen and distinguished from a point by average human vision. The distances at which this occurs define the near and far depth of field limits.
The charts below show the size of the blur circle as a function of the distance from the plane of focus. For reasons I won't go into here the limit of "acceptably sharp" is set as 0.03 mm (for a full frame camera). As long as the size of the blur circle stays below this limit (represented by the red line), the object is considered to be within the depth of field. (The green line represents the effects of diffraction which we can ignore for now.)
The first chart is for your first case; 24 mm, f/11, focus distance of 1.56 m:
The blur size crosses the red line at about 0.8 m and at about 13 m (different DoF calculators can give slightly different results). You can see that the 1/3 in front and 2/3 behind rule doesn't work very well even in this case. Your focus distance is pretty close to the hyperfocal distance of about 1.8 m, so the far DoF limit is rapidly extending toward infinity.
This chart represents your second case; 24 mm, f/11, focus distance of 6.16 m:
Although the near DoF limit is actually at about 1.4 m, this chart does not extend that far, so you can see that the size of the blur circle never approaches 0.03. On the far side, the DoF now extends to infinity. The total depth of field is clearly much greater the focus distance, as you saw in your church example.
I have found these charts useful for visualizing depth of field and how it is delimited by the size of the blur created by distance from the plane of focus. I hope you find it useful too. (You can play around with the calculators that produces the charts at http://www.tawbaware.com/maxlyons/calc.htm.)
Dave
