Henning Wulff wrote:
>> Well, why?  Even when cropping a D1s image to D60 size (1/1.6 factor), 
>> it will still have slightly more pixels than a full D60 frame
>>
>> Per
> 
> 
> Not quite correct :-)
> 
> The 1.6 factor is linear, but the pixel count describes an area. So you 
> have to square 1.6, which gives you a factor of 2.56. This, into 11.1Mp, 
> gives you 4.35Mp, while the D60 has 6.3 or so.


Of course you�re right! Stupid of me...



> 
> The other thing that hasn't been mentioned is that film has a much finer 
> basic unit: the silver halide grain than digital, which has an active 
> electronic element. For the foreseeable future, we're not likely able to 
> make a transistor with wires smaller than a silver halide crystal. Then 
> comes the fact that to make a 'unit' of colour information we need 4 
> active electronic elements, with conductors (one red, one blue, and two 
> green in the usual bayer patter). An 11.1Mp sensor gives use 2.775Mp of 
> true information which with some elegant interpolation algorithms gives 
> us 11.1Mp of output again. We're not in film territory by a long shot.
> 


Well, here is a complication: a CCD (CMOS) cell has a more or less 
continuous scale of excitation, which the A/D converter represents by, 
say, a 12 bit binary number (in RAW) for the intensity there.  A single 
silver halide grain (not to be confused with silver grains after 
development) is excited or it is not, giving one bit of intensiy 
information.  So, paradoxically, film is a digital device, while sensor 
cells are analog...  The continuous tone scale in film is the result of 
many grains cooperating to a greater or smaller degree, in proportion 
with the photon density received.  So any comparison between film and 
sensors is far more complicated than it would seem.  Agreed, we�re not 
in film territory yet (look at grass or foliage in a MF shot beside a 
digital one), but it is not entirely clear how far off we are...

Anyway, we seem to have digressed seriously from the initial question :-)

Per

*
****
*******
***********************************************************
*  For list instructions, including unsubscribe, see:
*    http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************

Reply via email to