Date:         Mon, 24 Sep 2001 13:02:53 -0700
Reply-To:     Cassell.David@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <Cassell.David@EPAMAIL.EPA.GOV>
Subject:      Re: Compress=Binary and WinZip
Content-type: text/plain; charset=us-ascii

Roger DeAngelis wrote [in part]:
>   Sometimes the compress=binary option produces a file that WinZip cannot
> compress into a smaller file.

Excellent!  I'm really glad to hear that, because my experiences with
compress=binary have been *lousy* - due to the nature of the data set.
I have even gotten a rousing 2% INCREASE when using compress=binary on
a dataset where there are a lot of numerics but not a lot of pattern within
a record.

> It is interesting thet the Zipped version of the compress=binary
> SAS Table is larger then source SAS Table(Compress=binary).

Yes, that's a common problem when trying to compress a file which has, in
essence, been compressed as far as is reasonable without losing
information.
For fun, do a maximal Zip of a file, then try to compress that zipped file
even further.  You'll usually do a little worse.

Thanks for the info,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician