STZip v2.2 vs. LHA v2.01 Comparison
                    -----------------------------------
                              by Steve Johnson

This is a performance comparison between Vincent Pompey's STZip v2.2
ZIP archiver and Roger Burrows' LHA v2.01 LZH/LHA archiver.  This 
comparison does not, nor does it intend to compare the aesthetics of
each archiver and is only intended to compare the overall raw 
performance of the two.

To make as fair a comparison as possible, I chose several different
types of files to archive, from picture files to ASCII text files to 
executable and data files, and also combinations of these file types.

STZip v2.2 was set to use the 'best' deflate compression and LHA v2.01
was used to produce -lh5- format LZH/LHA archives.  All tests were done
on the same system (a 1040STE w/ TOS 1.62 and 4MB RAM) under the same
circumstances (e.g. no RAMdisk, same source/destination file paths, etc.
for BOTH).


Following are the results of each archive tested (times listed are in
min:sec format):


File: PCSPICS2 - a collection of Photochrome picture files

 (uncompressed)               - 1,121,488 bytes (100.0%)
 (STZip v2.2 - deflate(best)) -   849,405 bytes ( 75.7%)
 (LHA v2.01 - lh5 format)     -   826,477 bytes ( 73.7%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time      6:00           9:25
 Extract time      3:00           2:42


File: GOGOST50 - GoGo ST v5.0

 (uncompressed)               - 161,661 bytes (100.0%)
 (STZip v2.2 - deflate(best)) -  52,965 bytes ( 32.8%)
 (LHA v2.01 - lh5 format)     -  52,873 bytes ( 32.7%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time      1:30           1:14
 Extract time       :31            :20


File: ZNET9312 - Z*Net Online Magazine #9312 (ASCII text file)

 (uncompressed)               - 118,093 bytes (100.0%)
 (STZip v2.2 - deflate(best)) -  46,113 bytes ( 39.0%)
 (LHA v2.01 - lh5 format)     -  48,232 bytes ( 40.8%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time       :41           1:00
 Extract time       :17            :13


File: PGS22 - my entire PageStream v2.2b folder w/ subdirectories

 (uncompressed)               - 3,792,526 bytes (100.0%)
 (STZip v2.2 - deflate(best)) - 1,865,944 bytes ( 49.2%)
 (LHA v2.01 - lh5 format)     - 1,886,210 bytes ( 49.7%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time     33:08          32:57
 Extract time     12:36          10:18


File: SAMPLES - a collection of digital sample files

 (uncompressed)               - 128,199 bytes (100.0%)
 (STZip v2.2 - deflate(best)) -  92,892 bytes ( 72.5%)
 (LHA v2.01 - lh5 format)     -  91,799 bytes ( 71.6%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time       :47           1:07
 Extract time       :26            :21


File: MIDISTUF - a collection of MIDI song files

 (uncompressed)               - 622,546 bytes (100.0%)
 (STZip v2.2 - deflate(best)) - 105,320 bytes ( 16.9%)
 (LHA v2.01 - lh5 format)     - 100,579 bytes ( 16.2%)

                STZip v2.2     LHA v2.01
                ----------     ---------
 Archive time      6:02           5:23
 Extract time      2:08           1:09


As you can see from the above results, sometimes STZip v2.2 is faster
than LHA v2.01 and vice-versa, although when STZip IS faster, it's 
usually a more noticeable difference.  On the other hand, in EVERY case,
LHA v2.01 was faster at extracting the archive, and quite noticeably so
on larger archives.

In most cases, LHA v2.01 also produced a smaller archive.  STZip v2.2
seems to only really yield better compression with text files 
(especially pure ASCII ones).  Both formats are usually within a few
percentage points in amount of compression, though for non-'text-heavy'
files, LHA v2.01 seems to yield better overall compression than STZip,
however slight.

As previously noted, this comparison does not take the aesthetics or 
operation of the archivers into consideration.  That is left up to the
individual to decide for himself or herself.



Steve Johnson
GEnie: STEVE-J


           *** PERMISSION TO REPRINT IS GRANTED BY AUTHOR ***