ðòïåëôù 


  áòèé÷ 


Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

  óôáôøé 


  ðåòóïîáìøîïå 


  ðòïçòáííù 



ðéûéôå
ðéóøíá












     áòèé÷ :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] RE: 8 bit versus 16



> From: Robert Logan
>
> The tools of tomorrow, be they better hardware,
> or better software, may allow me to manipulate
> the 16 bit data (14 bit in my case), better
> to produce a better looking image.
>
> If I have 256(8bit) greens in my file, and in the
> other I have 257(16bit), then I have more to work
> with to achieve an end.

I haven't read this entire thread, but in what I've read I haven't heard
mention of noise. What limits the number of useful bits is the noise level.
A good scanner may well have better than an eight-bit S/N ratio, but I doubt
anything out there does better than twelve, except perhaps a high-end drum
scanner. Small CCD cameras often provide raw files with more than eight
bits, but the extra bits are complete rubbish. The larger CCD cameras have
nine bits of useful data, maybe ten under optimum conditions.

If you scan a slide (or negative) with some clear blue sky, that's a good
test of the noise level of the scanner, since sky is completely textureless,
and virtually noise-free. If you blow up the image and examine the pixels,
and it looks like blue confetti on your eight-bit display, then you really
don't need any more than eight bits, because the noise will take any finer
gradations and dither them up into the top eight bits. This is pretty easy
to prove in practice, by taking a 16-bit scan, copying it and truncating the
copy to eight bits, and then seeing if a strong contrast enhancement curve
gives you noticeable banding in the latter but not the former. With enough
noise (whether from the CCD or the film grain), the extra bits won't matter.

--

Ciao,               Paul D. DeRocco
Paul                mailto:pderocco@ix.netcom.com

----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body



 




Copyright © Lexa Software, 1996-2009.