Re: Levels & Curves

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

 



In article <20011221152045.99878.qmail@web20708.mail.yahoo.com>, Michael 
Greer <mgreer316@yahoo.com> writes
>
>I've never seen or driven a scanner with a hardware exposure control.

I am pretty sure that all of the Nikon range do, they certainly all used 
to do this, but I can't be certain about the more recent units such as 
the LS40/4000/8000 etc.

>What does
>it do anyway? Does it vary the intensity of the light source?

No - it varies the exposure time.

>In the past, the scan process
>was:
>1) the scanner's hardware would capture the image in higher bits.
>2) the software driver would process that image in higher bits.
>3) the software driver would return the image to the host in 8 bits.
>
>In step 2, the computer is being used to process the image. Whether it's done
>in the scan driver software or the image editor is irrelevant. The computer
>still needs to horse power to handle it.
>
Eh???

Either:
a) you haven't explained what you mean particularly clearly,
b) you've got this backwards or,
c) scanner manufacturers have secretly developed a matter transporter 
and are temporarily beaming the CPU down those SCSI wires to the scanner 
to implement avail themselves of that "horse power" in the scanner.

My bet's on (a), but I would love to be convinced that its (c).  :-)

Either the data is processed in the scanner's firmware (and reduced to 
8-bits by that) or it is transferred to the computer in high bit level 
and processed by the CPU (requiring the appropriate "horse power" to do 
so) - you can't have it both ways!

If it is returned to the computer in high bit, then it is available to 
the application in high bit (unless the driver has been  deliberately 
crippled, in which case Ed Hamrick's application will probably overcome 
that restriction, ensuring that a competitive upgrade will be available 
from the vendor eventually.  If it is only returned in 8-bit then it 
MUST be processed in the scanner's firmware without any significant load 
on the host computer.

Scanners already contain an ALU for calibration correction anyway - 
reduction of the data to 8-bit range is simply a modification of the 
correction data fed to that ALU for each pixel in the CCD.  There is a 
marginal increase in load on the computer in calculating each of the 
pixel correction coefficients for the ALU - one entire pixel line's 
worth of calculations.
-- 
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers
-
Turn off HTML mail features. Keep quoted material short. Use accurate
subject lines. http://www.leben.com/lists for list instructions.

[Index of Archives]     [Photos]     [Yosemite]     [Steve's Art]     [The Gimp]     [PhotoForum]     [Epson Inkjet]
  Powered by Linux