home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!ames!olivea!sgigate!odin!glass.esd.sgi.com!donl
- From: donl@glass.esd.sgi.com (donl mathis)
- Newsgroups: rec.photo
- Subject: Re: Contrast mask woes
- Message-ID: <1992Nov23.212157.14555@odin.corp.sgi.com>
- Date: 23 Nov 92 21:21:57 GMT
- References: <2254@sousa.tay.dec.com>
- Sender: news@odin.corp.sgi.com (Net News)
- Reply-To: donl@glass.esd.sgi.com (donl mathis)
- Organization: Silicon Graphics, Inc., Mountain View, CA, USA
- Lines: 31
- Nntp-Posting-Host: glass.esd.sgi.com
-
-
- In article <2254@sousa.tay.dec.com>, madler@vssad.enet.dec.com (Michael
- C. Adler) writes:
-
- |> I made a mask that seemed about the minimum density to make the image
- |> printable without dodging. I wouldn't say the cloud region was especially
- |> dark. The mask had relatively low but perceptible detail (developed about
- |> 1/3 of normal) and was unsharp. When I laid the mask over the slide
- a miracle
- |> occurred: the subtle detail in the clouds disappeared completely. The mask
- |> compensated perfectly for the original and generated a very nice gray.
-
- It sounds like there is too much detail in the mask. They are
- opposites, and opposite subtle details will cancel each other out as
- you have described, because a negative plus a positive is zero (or
- nearly zero, in our case.) You want an *unsharp* mask -- you want to
- make it fuzzy, so you get an overall blending in the masked areas; just
- a wash of gray. No detail. The mathematical analogy for this is a
- positive plus a constant; everything moves a little, but retains its
- basic shape. The relationships between the lighter and darker areas
- within the highlights are maintained.
-
- Some people do this with a layer or two of acetate or some foggy
- material; some do it by methods like putting a piece of glass between
- the original and film, and then spinning the whole thing around
- off-axis under the exposing lamp, so the shadows come in from all
- sides. (Sounds a little tricky, and somewhat difficult to repeat, but
- I've never done it.)
-
- - donl mathis at Silicon Graphics Computer Systems, Mountain View, CA
- donl@sgi.com
-