Newsgroups: comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!portc01.blue.aol.com!news-peer.gsl.net!news.gsl.net!hammer.uoregon.edu!arclight.uoregon.edu!news.sprintlink.net!news-peer.sprintlink.net!howland.erols.net!EU.net!usenet2.news.uk.psi.net!uknet!usenet1.news.uk.psi.net!uknet!dispatch.news.demon.net!demon!netcom.net.uk!ix.netcom.com!netcomsv!uu3news.netcom.com!netcomsv!uu4news.netcom.com!telesoft!east.alsys.com!news
From: Bob Manjoney <rjm@snet.net>
Subject: Re: Predicting Footprint Size
Content-Type: text/plain; charset=us-ascii
Message-ID: <3288B4E3.4F04@snet.net>
Sender: news@thomsoft.com
Reply-To: rjm@snet.net
Content-Transfer-Encoding: 7bit
Organization: Thomson Software Products, Burlington, MA, USA
References: <00001aa7+0000350c@msn.com>
Mime-Version: 1.0
Date: Tue, 12 Nov 1996 17:33:23 GMT
X-Mailer: Mozilla 3.0 (Win95; I)
Lines: 25

Mark Allard wrote:
> 
> Our team is curently using VisualAge 3.0.  We're trying to get our
> hands on metrics for determining final image size based on number of
> classes.  Has anyone ever seen such metrics?

I'm confused by your question.

The topic is "Predicting Footprint Size".  In my mind, footprint is
applied to the runtime-size of your application, i.e., its memory
footprint.

"Final image size", on the other hand, refers to the size of the
resulting image file.

I'm not sure how useful # of classes would be in predicting the former,
based on the fact that your application's dynamic behavior is probably
the primary contributor to it's memory footprint.

The latter may be more predictable based on # of classes, however, this
too is not guaranteed, due in part to the fact that your image consists
of far more than simply "classes", i.e., cached instances, compiled
methods, pool dictionaries etc..

Bob/TSP
