From xemacs-m  Sun Feb 16 23:39:19 1997
Received: from beavis.bayserve.net (jmiller@port93.bayserve.net [206.148.244.184])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id XAA13353
	for <xemacs-beta@xemacs.org>; Sun, 16 Feb 1997 23:39:16 -0600 (CST)
Received: (from jmiller@localhost) by beavis.bayserve.net (8.7.5/8.7.3) id AAA09938; Mon, 17 Feb 1997 00:57:21 -0500
Date: Mon, 17 Feb 1997 00:57:21 -0500
Message-Id: <199702170557.AAA09938@beavis.bayserve.net>
From: Jeff Miller <jmiller@bayserve.net>
To: xemacs-beta@xemacs.org
Subject: efs
Reply-to: jmiller@bayserve.net
Mime-Version: 1.0 (generated by tm-edit 7.105)
Content-Type: text/plain; charset=US-ASCII


How "done" is efs?

I know it was talked about some on the beta list, but I didn't follow much
of it.  19.15b94 is my first exposure to it.

I am seeing some *serious* performance degradation with efs loaded.  I'm
certain that efs triggers it.  I normally load sample.emacs when starting
up.  I did ran some tests using identical start ups, the only difference
being that (require efs) was commented out of sample.emacs.

the test I ran was "xemacs -f w3"

elapsed time from starting the command to when I see the w3 home page is

	w/efs		wo/efs		19.15b93&ange-ftp
	7.15		3.10		2.45
	9.15		2.40
	8.45		2.45
	10.15		2.45
	9.45		2.10
	9.0		2.30
	8.15

Elapsed times are minutes.seconds.   This is on a 40M 486.  I'm not sure
why there is so much variation in the times w/efs.  

It seems to me something is very wrong here.

