CONFOCALMICROSCOPY Archives

November 1994

CONFOCALMICROSCOPY@LISTS.UMN.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Danny Thomas <[log in to unmask]>
Reply To:
Confocal Microscopy List <[log in to unmask]>
Date:
Tue, 15 Nov 1994 08:26:34 +1000
Content-Type:
text/plain
Parts/Attachments:
text/plain (46 lines)
>When doing this, you will get the best results by setting the "Mesh size"
>and "Image dimensions" on the print set menu so that you have one mesh
>point for each pixel in the data.  i.e. for 768 wide data and 150/inch mesh
>size, choose 768/150 = 5.12 inches wide.  Ditto  for the vertical size.
 
>The Laserwriter Pro 630 should give you 32 grey levels with Photgrey turned
>on and a 150/inch mesh size.  This is more than enough for most
>fluorescence confocal images.
it's actually called PhotoGrade, with the usual computer industry
propensity for BiCapitalization. See the entry in the paper or electronic
version of "The New Hacker's Dictionary".
 
>It is also important to set the Mesh Orientation to be
>"up-down/side-to-side" rather than at 45 deg. (which is the usual default
>setting) because the former matches the way that the data is collected.
>These two precautions prevent the program from having to re-calculate your
>data set, a process that reduces resolution and may introduce aliasing.
 
by "mesh" I assume you're referring to what the industry calls "lines per
inch" halftoning cells. Yes at 600 dpi with either a 0 or 90 degree screen
you'll end up with sqr(600/150)+1 = 17 shades of gray. Running at 150lpi is
not really a good idea at low 600dpi resolutions - the small number of gray
levels will lead to posterization and may contribute to the moires you've
observed. It is true that if you scale the image to _exactly_ (like <0.1%
mismatch) fit a halftone cell printing will go faster - something I did
with the image I print on our BW letterhead logo - but that really isn't
something to aim for with grayscale images. If you get it slightly out that
may explain the moire, though I've never seen them myself. The examples in
the book mentioned below clearly show why 0 and 90 degree screens are NOT a
good idea, even for grayscale printing (CMYK presses represent rather more
difficult compromises).
    The default halftone screen for the LW630 is 85lpi @ 45 degrees which
gives approx 50 gray levels. PhotoGrade is only applicable at 300dpi on the
LW630, but may produce somewhat better results than 600dpi by itself.
Postscript halftoning and printing technology is well covered in
"PostScript Screening - Adobe Accurate Screens" by Peter Fink. I found it
surprisingly useful and quite easy to read. I'm not sure whether there'll
be a later edition to cover stochastic screening, aka CrystalScreens or FM
screening, which obviates the problems of moire in colour printing, albeit
requiring tighter control of printing presses (particularly of dot-gain in
the crucial mid-tone region, at the small pixels used).
 
 
cheers,
Danny Thomas  ([log in to unmask])

ATOM RSS1 RSS2