From xemacs-m  Sun Feb  9 10:46:27 1997
Received: from martigny.ai.mit.edu (martigny.ai.mit.edu [18.43.0.152])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id KAA04128
	for <xemacs-beta@xemacs.org>; Sun, 9 Feb 1997 10:46:26 -0600 (CST)
Received: from berne.ai.mit.edu by martigny.ai.mit.edu with SMTP
	(1.40.112.8/16.2) id AA297756779; Sun, 9 Feb 1997 11:46:19 -0500
From: Bill Dubuque <wgd@martigny.ai.mit.edu>
Message-Id: <199702091646.AA297756779@martigny.ai.mit.edu>
Received: by berne.ai.mit.edu
	(1.40.112.8/16.2) id AA019016778; Sun, 9 Feb 1997 11:46:18 -0500
Date: Sun, 9 Feb 1997 11:46:18 -0500
To: larsi@ifi.uio.no
Cc: xemacs-beta@xemacs.org
In-Reply-To: <m2d8uaduc0.fsf@proletcult.slip.ifi.uio.no> (message from Lars
	Magne Ingebrigtsen on 09 Feb 1997 13:42:23 +0100)
Subject: Re: wot i need

: From: Lars Magne Ingebrigtsen <larsi@ifi.uio.no>
: Date: 09 Feb 1997 13:42:23 +0100
: 
: Bill Dubuque <wgd@martigny.ai.mit.edu> writes:
: 
: > It would be useful to get more datapoints. If anyone else has
: > XEmacs and FSF compiled on the same machine, please contrast
: > their times on Kyle's original example, which he found twice 
: > as slow in XEmacs:
: > 
: > (let ((i 0)) (while (< i 300000) (setq i (1+ i))))
: 
: I have a 486 running Linux, and I've compared Emacs 19.34 with
: XEmacs 20.0 with Mule using this test: ...
: 
: And the results are:
: 
: => 29.94636797904968 (XEmacs 20.0 w/Mule)
: => 24.478706955909729 (Emacs 19.34)

On my 486/66 Linux, the above takes only 13s in FSF 19.28, and
14s for XE 19.11, but 28s for XE 19.12b25.

Probably this means either 19.34 has inherited the bug or else your 19.34
is not built optimally. The latter would seem more likely since Kyle has 
reported a factor of 2 slowdown of XE 19.20 vs. 19.34.

The same factor of two slowdown seems to occur even after byte-compilation,
so its not likely specific to the interpreter proper. I'm beginning to
believe Kyle's claim that 

  **** ALL e-lisp code is now a factor of TWO slower. ***

The only other likely culprit I can imagine (besides QUIT) is possibly the 
code for variable ref/set, which has changed slightly in 19.12.

Kyle or anyone on Linux: can you do the above in 13s or less with
FSF 19.xx or 19.11 or less?

-Bill




