Typ... hittade svaret efter mycket letande på dpreview.
Enklast (fast så enkelt var det inte) är att läsa här:
http://forums.dpreview.com/forums/read.asp?forum=1029&message=11495298
Klipper in texten ifrån Doug Kerr:
> doug, thanks for the input. so correct me if i am wrong:
>
> m' = m + L/f
>
> so if i use a 1:4 lens (so m is .25), and if i stack 20mm + 36mm
> ETs, which would be 56mm (L), and if i set the lens to 70mm focal
> length (24-70 f2.8 lens), this comes out to...
>
> m' = 1.05, so a tad over 1:1 mag ratio...is this accurate?
That's what I get. (I assume the maximum magification of the lens at 70 mm is 1:4. Note that in general, it will be different for different focal lengths of a zoom lens.)
> any
> idea how many stops of light i lose with these 2 ETs stacked?
You bet!
With the lens proper at a magnification of 0.25, the effective f/number (from an exposure standpoint, and assuming 1.0 transmission of the lens) is 1.25 times the indicated f/number. (See later for discussion of why.) So if the lens is set to f/4, the effective aperture is f/5. (The "bellows factor" which governs this is m+1.)
With the rig at a magnifcation of 1.05, the effective f/number is 2.05 times the indicated f/number. So if the lens is set to f/4, the effective aperture then is f/8.2.
The decline in effective aperture from the first case to the second is about 1.4 stops. (That will be true regardless of the transmission factor, which cancels out in this comparison.)
All of this results from the fact that the property of teh aperture that governs exposure (in any case, extension tubes or not) is not really the f/number, which is this ratio:
N = f/A
where f is the focal length and A is the diameter of the entrance pupil of the lens (the effective size of the actual aperture stop, as it would be seen from in front of the lens).
The actual quantity governiong expposure is:
N' = Q/A
whwre Q is the distance from the second principal point of the lens to the focal plane (film or sensor).
For focus distances that are many times the focal length, Q is for all practical purposes the same as f. Thus we can use the f/number (N), which we have handy, instead of the more actual governing quantity N'. This is the dirty little secret of the f/number!
But when the focal distance becomes smaller (with or without extension tubes), this approximation is no longer very accurate. (At 1:1, Q=2f, for example.) So we adjust the f/ number by multiplying by the bellows factor (m+1). But what we are really doing is correcting for the fact that f and Q are not the same (which we do without even learning about Q)!
Izzat cool or what!
Best regards,
Doug
