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: Digital Darkroom Computer Builders?



Hi Paul,

Yes, passing the 8 bit data through an 8 bit LUT would cause gaps/combining
in anything but a linear/monotonic LUT (1:1)...it simply has to, which is
the same reason to do tonal manipulations in a larger space.

> The video card includes a 256-entry lookup table (for each color)
> which gets
> loaded with a gamma correction curve (e.g., by Adobe Gamma
> Loader). Assuming
> that table doesn't just have a straight line in it, some values will be
> squeezed together, creating duplicates, and other values will be spread
> apart, doubling their distance.

It may far more than double the "distance".  Any time you pass N bit data
through an N bit LUT, you will have gaps/combining, unless the LUT is
1:1...then what would be the purpose of the LUT?

> If the lookup table and DAC had two more bits of fractional
> resolution,
> those low values wouldn't be duplicated, and the high values wouldn't have
> such large steps.

Em, well...you are saying, say, a 1024 bit output LUT with a 256 input?
That entirely depends on the data in the LUT on whether it does as you
suggest (eliminate gaps/combining).  The same curve data set in a 8 in 8 out
vs an 8 in 10 out LUT will give you the exact same 8 MSBs, so Adobe would
have to be aware of this.  I assume you are talking about the curve being
created to provide 10 bits out?  Still, that can, depending on the curve,
cause gaps/combining...and if you take the 8 MSBs of the 10 bits, they
should be the same as the 8 bits out of an 8:8 LUT (which was my point
initially), or the "base" curve would be different.  So, I don't see how
that helps, unless the lower two bits were visibly perceptible...and as you
say, it may be (probably is) visibly imperceptible.

Regards,

Austin


----------------------------------------------------------------------------------------
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.