Okay, so you know about Depth of Field (DOF) right? The amount of an image from foreground to background that is in focus? And you know that it varies dependant upon the aperture (f-number) and focal length of the lens. But did you know it also varies between film and digital, and from sensor to sensor?[Technical bit - ignore if you like] It is dependant upon something called the Circle of confusion - all light can only be focussed by a lens on a single plane, in front of or behind which light rays form minute discs rather than points of light, known as circles of confusion. Within certain tolerances, film grain or digital sensors which have a finite resolution, these circles of confusion are small enough to still appear as points, and so there seems to be a whole range of distances from the lens at which the subject is in focus. Within this range, the circles of confusion appear to be dots. Further from this range, or nearer than it, the circles widen and become diffuse.
[Edited by Moderator on 7/5/2011 9:02:38 AM]
[Back to interesting stuff]Now, DOF is a powerful tool to the photographer - you can blur distracting backgrounds and make your subject stand out with a shallow DOF, or give your images depth with a greater DOF. Landscapes, particularly, benefit from maximising DOF, and this is where hyperfocal distance comes into play.If you focus your lens on infinity, the DOF will extend from some point in front of the lens, the hyperfocal distance, all the way through to infinity, as shown here:
If you refocus your lens on the hyperfocal distance, the DOF will extend from half the hyperfocal distance all the way through to infinity, as here:
[Edited by User on 7/5/2011 9:40:40 AM]
That's all very well, I hear you say, but how do I know where to focus? Easy. A simple calculation will provide you with the answers:H = ((l x l)/(a x c)) + lwhere: H = hyperfocal distance l = focal length (actual, not 35mm equivalent) a = aperture c = diameter of least circle of confusionThe circle of confusion diameter changes from film to digital, and from sensor to sensor. For film (and Canon EOS 1DS, 1DSmkII) it is 0.03; for Canon EOS 300D, 350D, 10D, 20D it is 0.019; for Canon EOS 1D and 1DmkII it is 0.023; for Nikon (D100, D70, D1H, D2H, D2Hs, D1X, D2X) it is 0.022A simple spreadsheet provides a nice printable table that you can carry around in your camera bag. Heres one I made for my 20D:
The chart shows varying focal lengths that are available to me, and the greyed out cells are just apertures that are not available to me at that focal length.
That's it. It is as simple as that, but it does mean that you can get images that are in focus from front to back, from the nearest blade of grass or poppy head, through the trees, to the furthest clouds:
Craigr1978 said:
I found this reference table with Google: linky Allows you to look up the Circle of Confusion (in mm) for your camera.
Fishcat said:
I think there is a slight typo in the formula. It should read:H = ((l x l)/(a x c)) / l,000.I used a slash for the last division because it kinda looked like a "plus" sign in Dragon's formula. The last character was the number "one" (which looks exactly like a lower case "ell"), but should have been "one thousand" to convert from millimeters to meters. The answer should be "3.803" meters. I received Dragon60's spreadsheet and am checking why it reads "3.82" as that is what it calculates. Hmmmmm....
Dragon60 said:
you nearly had it... the first part is ok up to 3,802.63, the ADD focal length again, +17 = 3,819.63That gives the result in mm, so divide by 1000 to get 3.81963 metres :)
Sorry Dragon60, should have let you respond. I did a quick look on Google for the formula and found it both ways; without Focal Length added (the one I used) and with Focal Length added. Not sure which is right, but it really does not make much difference because you are not going to be that precise in the field anyway. Oh, and if anybody is interested, divide by 304.8 instead of 1,000 to get the distance in feet.
[Edited by User on 7/5/2011 9:36:55 AM]
theLimeyBrit said:
Me and my point-and-shoot will be sulking over there in the corner if anyone needs us./me looks longingly at the DSLR-shaped puppy in the window
Dallas_TX said:
I hate to ruin a great sulk, but you really can do hyperfocus with a point and shoot. Every single bit of it. When one presses the shutter release down halfway the focus is locked as you know. Just focus (about) one-third into the image lock your focus, recompose and let rip.
Tonebarge said:
Alan, this is so cool. But I'm doing something wrong and can't get the same result as you do on your chart. Couldja check me math, mate?Take the square of the focal length. I'll use 17mm. That would be 289. Check. Divide that with the result of aperture times least circle of confusion. I will use 4 for f/4. So 4 times .019 (for the 20D). That would be .076. That would make it 289 divided by .076. That would be 3,802.63. Check. Now divide by focal length and I used 17mm. That would make the result 223.68. Your chart shows 3.82.I'm sure my mistake is obvious but I'm just not seeing it. What am I doing wrong?
LimeyBrit summed it up nicely by pointing out that the focal length must be actual, not 35mm equivalent. The sensor size factor has no bearing on this whatsoever. It seemed counter intuitive to me too, but when you consider that the sensor size issue is exactly the same as having a much larger sensor and then cropping, that begins to help understanding.
Hangerhead said:
Dragon60 - re-created your chart and I think you've got a copy paste error on your 35mm (the 28mm values have come across):Column should read:3535.8532.2723.0618.4616.1514.3611.5510.279.128.095.904.062.972.262.05