home *** CD-ROM | disk | FTP | other *** search
- Installation instructions for HCOMPRESS on Unix machines
- R. L. White, 20 April 1992
-
- This directory contains a makefile for building HCOMPRESS on a Unix
- machine and some C-shell scripts to make using the programs a little
- easier. This makefile has only been tested on Suns and DECStations, so
- some small modifications may be required for other machines.
-
- Steps for installation:
-
- (1) Copy the contents the contents of this directory into the
- source directory.
- (2) Do "make all" (or just "make") to compile and link the programs.
- (3) Edit makefile: change the INSTALLDIR variable to point to the
- directory where you want the executables to reside. By default
- they will go into ${HOME}/bin, i.e. your bin directory. You should
- chose a directory that is included in your path.
- (4) Do "make install" to install the programs in your chosen directory.
- (5) Do "rehash" to update your hash tables.
- (6) There is a man page in the file hcompress.1. It can be installed
- in your man directory. The doc directory includes both a plain text
- version of the man page (hcompress.txt) and a postscript version
- (hcompress.ps).
-
- Now the programs should be ready to run. You may want to customize
- the shell scripts for your applications. Some possible changes to
- consider:
-
- (1) There are shell scripts to compress multiple files in FITS format
- and in hhh/hhd format. If you usually use some other formats you
- could include in the script a conversion from your format to
- one of the accepted formats.
- (2) The amount of compression is determined by the scale factor, set by
- the -s option in the programs. For astronomical images the compression
- is generally limited by the amount of noise in the images and the
- scale factor should be about 2 times the RMS noise in the image.
- For noiseless images any non-zero scale factor will give up some
- information, so the choice of scale factor will be determined by how
- much you are willing to lose. A default scale factor of 666 is used
- in the scripts -- this should probably be changed to a value appropriate
- for your typical images. If you like, you could set this default to 0
- (lossless compression.)
-