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]

Re: filmscanners: Grain Aliasing at 2700pppi



Thanks, guys. I intend to spend the better part of a day (or two, if I have
to) working on those two pictures (Tiger and Graduation). I've been putting
this off for too long. :-)

Best regards--LRA


>What about the same thing - except using smart blur?  I have had some good
success with smart blur (which of course tries to preserve the edges).  I
generally have to use the low end of the settings, but it can be quite
surprisingly nifty on some images if you take care with settings.

Julian

At 10:48 06/04/01, Maris wrote:
>I have been changing to LAB and splitting the channels, then applying
either
>a Gaussian blur or Dust and Scratches, depending on the size of the grain,
>in the A and B channels only.  Most of the sharpness remains in the L
>channel when you recombine.
>
>See Dan Margulis's chapter from Professional Photoshop at
><http://www.ledet.com/margulis/LABCorrection.pdf> where he suggests this
>
>Maris
>
>----- Original Message -----
>From: "Lynn Allen" <lalle@email.com>
>To: <Filmscanners@halftone.co.uk>
>Sent: Thursday, April 05, 2001 6:15 PM
>Subject: filmscanners: Grain Aliasing at 2700pppi
>
>
>| Grain aliasing and noise has been a regular topic on this list. It should
>| be--Mark, Rob, I and others have been talking at it hard enough. Without
>any
>| spectacular results, I could add. :->|
>| It's a pity that TIFFs can't be sent reasonably on the Net, because I
just
>| ran up against one that makes the "Tiger" I wrote about into a
"pussycat."
>| This new TIFF, done in Vuescan with 6 passes because Miraphoto couldn't
>| handle it, has grain aliasing in every square milimeter! True, it was
>| under-exposed in existing artificial light, hand-held at probably 1/15th
>or
>| 1/30th tops, with a Pentax 1.8 lens. So what?
>|
>| "There probably isn't enough 'picture there' to make a picture, there,"
>you
>| might say. You've heard it before, said it before, and so have I, more
>than
>| once. But the thing is, there *is* quite a bit of picture there, and the
>| Scanwit "sees" it. Getting it *out of there* and making it presentable is
>| the difficult part.
>|
>| Most people I know would say, "Give it up, man." Well fine, but I don't
>| think my daughter will be graduating from highschool any time again soon.
>| It's been 22 years since her last go. :-)
>|
>| Every discussion we've had on this list about G-A begs the question "How
>to
>| deal with it?" We know (or do we?) what causes grain aliasing and/or
>noise,
>| what films to use in future, what scanners to buy in future, et cetera.
>But
>| how does one get those hundreds of blue-green pixels out of the dark
areas
>| and the red-brown pixels out of the flesh-tones today, this afternoon?
>|
>| That's my question, and I'm stickin' to it. :-)
>|
>| Best regards--LRA
>|
>| PS--BTW, have you noticed that using a soft brush and Cloning smoothes
out
>| those offending pixels? Not a lot of help unless one wants to "repaint"
>the
>| whole picture, but it might be a start. Or not.
>|
>|
>| -----------------------------------------------
>| FREE! The World's Best Email Address @email.com
>| Reserve your name now at http://www.email.com
>|
>|
>|


Julian Robinson
in usually sunny, smog free Canberra, Australia


-----------------------------------------------
FREE! The World's Best Email Address @email.com
Reserve your name now at http://www.email.com





 




Copyright © Lexa Software, 1996-2009.