I'd say it has more to do with the signal-to-noise ratio and the

long exposure associated with the small aperture (high f-stop).
The dust spot on the sensor surface will produce a constant signal (which is noise in this case) on the CCD read out regardless the f-stop. On the other hand, the real signal strength is propotional to the light that falls on the sensor.
Given the same exposure, the longer exposure image will have lower signal-to-noise ration and thus the dust spot is more obvious with high f-stop exposure.
Let's consider two equivalent exposure cases:
A. f/2.8, 1/125 sec
B. f/16, 1 sec
Let's also assume that the dust spot signal strength is 1 unit and the evenly lit subject is 100 units in case A. This 1/125 sec exposure has signal-to-noise ration of 100. In case B, the dust spot exposure is increased 125 times (1/125 sec vs 1 sec) but the subject exposure remains the same (time increased but light reduced). Now the signal-to-noise ratio is 100/125 in case B. That should explain why you see the dust spot in long exposure case.

所有跟帖： 

• make a lot of sense to me....THANKS!!! -小夏- ♂ (0 bytes) () 08/13/2007 postreply 18:22:22

• What if the dust signal is *~0* or dependent of the image bright -小夏- ♂ (0 bytes) () 08/13/2007 postreply 18:40:23

• There is always thermal noise. That is also -cYq- ♂ (96 bytes) () 08/13/2007 postreply 19:37:09