Oliver> Hello David, hello Tristan, hello others I want so support
Oliver> the 30-bit mode of the UMAX Astra- and the UMAX High-End
Oliver> scanners.
Sounds good!
Oliver> That can be devided into two parts, from which only the
Oliver> first is important at the moment:
Oliver> 1) Gamma correction with 10 bits input and 8 bits output ( =
Oliver> 1024 bytes instead of 256 bytes )
I don't understand. Gamma is something defined by the backend. If
you want to define a vector of integers with a range larger than
0..255 then you just change your backend accordingly. Everything
will (should) fall out automatically.
Oliver> 2) a real 10 bits per pixel (greyscale) and 30 bits per
Oliver> pixel (rgb) support by sane => 2 bytes per pixel (greyscale)
Oliver> and 6 bytes per pixel (rgb) and 10 to 10 bits gamma data
Oliver> support
Yes, more than 8 bit/channel is certainly desirable and planned. I
think we defined that you'll have to declare it as 16 bits/channel and
scale the 10 bit values accordingly (I may be misremembering though,
so be sure to look at the spec).
Unfortunately, neither GIMP nor PPM support more than 24 bpp so the
real thing that's missing is frontend support. Incidentally, isn't 30
bpp mode normally used to achieve higher resolution through
interpolation? I imagine such a feature would be more useful for the
average user than 30bpp support---though I'm not claiming that 30bpp
is never useful.
--david
-- Source code, list archive, and docs: http://www.azstarnet.com/~axplinux/sane/ To unsubscribe: mail -s unsubscribe sane-devel-request@listserv.azstarnet.com