Hi Julian,
>How did you arrive at the CoC of 0.032?
A long time ago I wrote a spreadsheet that calculates DOF and allows you
to specify which CoC you want to use. I just took the info from the
table and (using my spreadsheet) back calculated what CoC they were
using.
>Is this the same figure used for the DOF scales on the EF
>lenses like the EF 50mm f/1.8 ?
My understanding is that DOF is a mathematical calculation by formula
for
a specific focal length at a specific aperture at a specific focused
distance for a specific measure of sharpness (Circle of Confusion), not
a physical measuring of a specific lens. If you were to measure the DOF
of a specific lens it might vary due to an incorrectly stated focal
length and also for varying focal lengths over its range of focus (my
50mm
f/1.4 lens is really 51mm at infinity). Also the measured aperture of a
lens will have some slight error that will affect DOF.
>Is there a CoC for using DEP mode?
To calculate this would be very difficult. The focus scale on most
lenses is not accurate enough or graduated enough. I tried to rough it
out and the best I could do was a range of CoC from .02 to .04.
If you want to try, here is the formula:
Depth of Field Formula
DOF in front of a subject = (C x F x D²) / (L² + C x F x D)
DOF behind a subject = (C x F x D²) / (L² - C x F x D)
where:
F = Aperture (f/ )
D = Focused distance (mm) (don't forget to convert this to meters after
calc)
L = Focal length (mm)
C = Acceptable circle of confusion (mm)
Convert your 'DOF in front', 'DOF behind' and 'Focused Distance' back to
meters by dividing by 1000. Then subtract 'DOF in front' from the
'Focused Distance'. Then add 'DOF behind' to 'Focused Distance'.
Solve the formulas above for C and plug in the numbers from any table
and you'll have what CoC they are using.
Art
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
