home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.graphics
- Path: sparky!uunet!usc!news.service.uci.edu!unogate!stgprao
- From: stgprao@st.unocal.COM (Richard Ottolini)
- Subject: Re: large sun raster files
- Message-ID: <1992Dec22.155607.14682@unocal.com>
- Sender: news@unocal.com (Unocal USENET News)
- Organization: Unocal Corporation
- References: <1992Dec18.152616.14324@neptune.inf.ethz.ch> <1h6k30INNe4r@werple.apana.org.au>
- Date: Tue, 22 Dec 1992 15:56:07 GMT
- Lines: 20
-
- >bahi@inf.ethz.ch (Hicham Bahi) writes:
- >
- >>I have the following problem:
- >
- >>I am working with huge sunraster files (27152x37937 pixels)
- >>which are run_length encoded. Since I don't have enough memory
- >>to load the whole picture, I have to cut it into smaller pieces.
- >
- >>Because of the run-length encoding, I have to read all the bytes
- >>of the file to find the starting position of the sub-image. For
- >>huge files, this means hours of computing.
- >
- >>Does anybody know a better way to cut run-length encoded rasterfiles?
-
- Change the data organization, I suggest to "tiling". Uncompress the
- image. Divide into tiles say 64 by 64. You may want to make the tile
- size equivalent ot a page of memory if you are storing the image in
- virtual memory, or a disk block if on disk. You may want to recompress
- each tile. As you pan through subimages, you only need to add tiles
- to the appropriate edge.
-