Dec 122017
Peformance of PKARC v3.6, ARC v6.00, PKZIP v0.90, PAK v.16, ZOO v2.01.
File PKZBENC2.ZIP from The Programmer’s Corner in
Category Tutorials + Patches
Peformance of PKARC v3.6, ARC v6.00, PKZIP v0.90, PAK v.16, ZOO v2.01.
File Name File Size Zip Size Zip Type
$COMPRES.BAT 214 140 deflated
$CONTENT.BAT 220 148 deflated
$EXTRACT.BAT 244 164 deflated
PKZBENCH.BAT 5390 1447 deflated
PKZBENCH.TXT 6062 2669 deflated

Download File PKZBENC2.ZIP Here

Contents of the PKZBENCH.TXT file

-=[ PKZBENCH.TXT: Performance of PKZIP, PKARC, PAK, ZOO, and ARC ]=- Page 1

Written by Eric Burleyson, sysop February 21, 1989
That Programming Place
(301) 474-1702
College Park, Maryland
PCPursuitable as Washington DC area code (202)
300/1200/2400 bps, 24 hours a day
Specializing in C/ASM/Pascal source code and utilities for MS-DOS

Now that the SEA versus PKware lawsuit is behind us, we have only to
contend with the turmoil of its aftermath! With this in mind, I
devised a test to satisfy my curiosity of the relative performance of
several file compression packages (and, my but there seem to be quite
a few of them since the SEA vs PKware legal wrangling muddied the
waters!). My goal was to do a real-world comparison, not a scientific,
verifiable, controlled experiment. I did, however, take reasonable
care to produce fair and accurate results (e.g., all runs were
performed in the same clean environment: no multi-tasking or obtrusive
TSR's were active, and the disk had been de-fragmented recently.) For
the insanely curious I've included the BAT files that I used for the
test -- be forewarned that you'll need the STOPWATCH or similar
program to perform the same test that I did.

[Note1: I use the word 'archive' to encompass all of the various
forms of compressed files (ARC, PAK, ZOO, ZIP, etc.). 'Archive'
means anybody, 'ARC' means SEA's flavor. I absolutely refuse to
use the unpronounceable '???chive' that has been popping up lately!]

All archives consisted of the same 7 files (the 2 .EXE and 5 ascii
files from the shareware package TheDraw v3.10) totaling 332,504
bytes. I chose these arbitrarily as a representative mix of files. The
test was run on an XT clone (7.44 MHz 8088-2, 640K, 65ms-access 20MB
hard disk.) Some of the packages maniplulate more than one style of
archive file -- I ran these once for each type of supported file.

[Note2: The archive packages I used were the latest versions I could
find, with one exception: I used PKARC/PKPAK v3.6 in this test, not
the identical but renamed PKPAK/PKUNPAK v3.61. Neither package is
available form PKware anymore, but I understand that SEA has the
rights to the source code as of January 31, 1989. So don't be too
surprised if the next version of ARC comes out faster, with smaller
ARC files! All perfectly legal, of course, as agreed to in the
settlement of SEA VS PKWARE last fall.]

My results are summarized on the next page. It seems clear to me that
PKware products are the hands-down performance leaders, and so I shall
be switching from PKARC to PKZIP as my primary archiver on my BBS.

[Note3: I obtained the new version of ARC (v6.00) just after I wrote
this, so I've included it in the table of results. SEA claims a
major rewrite with this version; it is indeed noticably faster, but
the ARC file is LARGER than with version 5.30! My conclusions are
not altered in any way by the new version of ARC.]

But, like it or not, we all, users and sysops alike, must now contend
with more than one brand of archive program; some boards will use
PKZIP, others won't, but rest assured that if you download from more
than one BBS, you will have to have more than one unarchive program.
Face it now and save yourself some grief. Same goes for my fellow
sysops. You are free to make any policy you want, but people upload
what they have, not what you want. Get used to converting PAK's to
ZIP's or ZIP's to ARC's, or whatever. It is inevitable.

-=[ PKZBENCH.TXT: Performance of PKZIP, PKARC, PAK, ZOO, and ARC ]=- Page 2

Compress Extract
Software Style Size Ratio Time Rate Time Rate
------------- --------- ------------ ------------ ------------

PKZIP/PKUNZIP ZIP 165818 50% 48 3473 42 3969
v0.90 ZIP(1) 164552 49% 96 1750 38 4420
ZIP(2) 160427 48% 103 1671 38 4528
ZIP(3) 156120 47% 117 1508 38 4642
ZIP(4) 151959 46% 136 1328 38 4751

PKARC/PKXARC ARC(sq) 173293 52% 47 3387 42 3791
v3.6 ARC 178152 54% 53 2912 43 3590

PAK v1.6 PAK 156223 47% 154 1145 121 1457
ARC(sq) 173840 52% 104 1526 94 1688
ARC 178232 54% 110 1402 98 1574

ZOO v2.01 ZOO 174186 52% 93 1702 99 1599

ARC v6.00 ARC 183285 55% 100 1492 65 2296
ARC v5.30 ARC 180856 54% 271 560 195 778
ARCE v3.1c ... ...... ... ... ... 57 2660

Uncompressed file size was 332504 (7 files).

'Style' values: ARC is the SEA-standard format
ARC(sq) is the PKware variant (includes 'squashing')
ZIP(n) are the 4 user-selectable compression factors

'Size' is the number of bytes in the archive.

'Ratio' is size of the archive vs the size of the uncompressed files.

'Time' is number of seconds of wallclock time the operation took.

'Rate' is the number of bytes saved or restored per second (the bytes
saved/restored is the uncompressed size minus the archive
size). It is, I think, an interesting measure of relative
efficiency, but feel free to ignore it if it strikes you as
just another meaningless statistic!


 December 12, 2017  Add comments

Leave a Reply