From xemacs-m  Sun Feb  9 05:35:33 1997
Received: from altair.xemacs.org (steve@xemacs.miranova.com [206.190.83.19])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id FAA00062
	for <xemacs-beta@xemacs.org>; Sun, 9 Feb 1997 05:35:32 -0600 (CST)
Received: (from steve@localhost)
	by altair.xemacs.org (8.8.5/8.8.5) id DAA31698;
	Sun, 9 Feb 1997 03:46:32 -0800
To: xemacs-beta@xemacs.org
Subject: Re: wot i need
References: <199702090130.AA201311825@martigny.ai.mit.edu>
X-Url: http://www.miranova.com/%7Esteve/
Mail-Copies-To: never
X-Face: #!T9!#9s-3o8)*uHlX{Ug[xW7E7Wr!*L46-OxqMu\xz23v|R9q}lH?cRS{rCNe^'[`^sr5"
 f8*@r4ipO6Jl!:Ccq<xoV[Qz2u8<8-+Vwf2gzJ44lf_/y9OaQ`@#Q65{U4/TC)i2`~/M&QI$X>p:9I
 OSS'2{-)-4wBnVeg0S\O4Al@)uC[pD|+
X-Attribution: sb
From: Steven L Baur <steve@miranova.com>
In-Reply-To: Bill Dubuque's message of Sat, 8 Feb 1997 20:30:25 -0500
Mime-Version: 1.0 (generated by tm-edit 7.103)
Content-Type: text/plain; charset=US-ASCII
Date: 09 Feb 1997 03:46:30 -0800
Message-ID: <m2raiqmcbt.fsf@altair.xemacs.org>
Lines: 59
X-Mailer: Gnus v5.4.11/XEmacs 20.1

Bill Dubuque writes:

> Steve: are you saying that you also see no difference in times
> between both your test AND Kyle's simpler test?

No, Kyle's test is twice as fast, with far more variance:

Function Name    Call Count  Elapsed Time  Average Time
===============  ==========  ============  ============
bench-mark-loop  5           74.289485000  14.857897000 19.34
bench-mark-loop  5           82.535311000  16.507062200 19.13
bench-mark-loop  5           83.701252999  16.740250599 20.0/MULE
bench-mark-loop  5           90.795639000  18.159127800 19.14
bench-mark-loop  5           102.16979300  20.433958600 19.12
bench-mark-loop  5           110.96389700  22.192779400 20.0/Latin-1
bench-mark-loop  5           115.79728500  23.159457000 19.15

Why are the Latin-1 20.0 and 19.15 so much slower?  Why is 20.0/MULE
one of the fastest?  Why are Lisp function calls so slow?

> I tried Kyle's test in 19.11 and 19.12b25 and 19.11 is about the
> same speed as FSF 19.28, but 19.12b25 is about twice as slow.

Is there any possible way I can trick 19.11 into compiling on Linux at
least to a temacs?

> But, I don't remember if the latter is compiled with debugging
> or not.

I don't know.  The latter numbers are right in the ball park for what
Kyle was reporting.

;;; loop-test.el -- basic test of Emacs lisp evaluation speed
;; execute by starting emacs -q, loading this file and doing M-x eval-buffer

(setq bench-mark-loop-count 1000000)
(defun bench-mark-loop ()
  "How long does it take to run through a loop."
  (let ((count bench-mark-loop-count))
    (let ((i 0))
      (while (< i count)
	(setq i (1+ i))))))

(require 'elp)
(elp-instrument-function 'bench-mark-loop)

(garbage-collect)
(bench-mark-loop)
(bench-mark-loop)
(bench-mark-loop)
(bench-mark-loop)
(bench-mark-loop)
(elp-results)

;;; loop-test.el ends here

-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be billed at $250/message.

