Re: 16 bit per sample support

Oliver Rauch (oliver.rauch@Wolfsburg.DE)
Tue, 06 Apr 1999 17:36:47 +0200

Jonathan A. Buzzard wrote:

> oliver.rauch@Wolfsburg.DE said:
> > If the 12 bits are streched to a 16 bit range, the lower 4 bits are
> > simply not used. If you want to add two images, you have to dived the
> > values by two before you add the values.
>
> Er, if I stretch to use the *full* dynamic range this is not the case.

If you do a linear tranformation this is simply a multiplication and so
the lower 4 bits are not used.

And that this is done as a linear tranfromation should be clear!

So: IT IS THE CASE DESCRIBED ABOVE!

>
>
> > It is not the sense of SANE to deal with such special features. SANE
> > is for scanning, we don`t want to make a new pixelgraphic program like
> > gimp
>
> SANE is for scanning, and what about those people that will be throwing
> their hands in the air if the scanning software is mugging their precious
> pixel values. For some applications this stretching of the values is
> simply not acceptable, why is SANE seeking to rule these applications
> out?
>

It is not right that the software is "mugging their precious pixel values"

Bye
Oliver

--
EMAIL: Oliver.Rauch@Wolfsburg.DE
WWW: http://www.wolfsburg.de/~rauch

--
Source code, list archive, and docs: http://www.mostang.com/sane/
To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com