Maybe the apparent lack of colour variations in the bricks in Paul's images is actually
due to the fact that this is down at the pixel level. AFAIK, with image chips in 
digital
SLRs and compacts the colour is produced by the software from arrays of four pixels 
with
colour filters over them, so the colour information has a lower resolution than the 
actual
number of pixels. Film of course has separate colour-sensitive layers with the grains
spread across at random. The microstructure of a film image shows cyan, magenta and 
yellow
grains that combine to give the correct colour. The actual grains, I think, are smaller
than the clumping that we perceive as grain. So with film we see a grainy image of the
right colour, while with a digital image at the microstructure level we just see a 
rather
blurry image that has the correct average colour.

It is presumably for the same reason that a digital file of, say, a blue sky has much 
less
colour variation than a film image, so the apparent graininess is lower, allowing us to
get away with larger prints. 

If this is the case, I'd expect a closeup of the same bricks (or for that matter a test
chart) to be reproduced faithfully in terms of colour by either system. The colour 
would
be spread out over many pixels in the case of the digital chip, and should be 
reproduced
accurately.

I find that my astro shots of stars show the same effect. On film, the star colours
reproduce nicely (where the film is not saturated), but faint star images on the 10D 
tend
to reproduce as white. 

On the other hand, if I am talking rubbish, I'm sure someone will tell me so!

Robin Scagell
Galaxy Picture Library
www.galaxypix.com
===============================================================
GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE

Reply via email to