From xemacs-m  Mon Dec 16 11:25:19 1996
Received: from palrel1.hp.com (palrel1.hp.com [15.253.72.10]) by xemacs.cs.uiuc.edu (8.8.3/8.8.3) with ESMTP id LAA24463 for <xemacs-beta@xemacs.org>; Mon, 16 Dec 1996 11:25:17 -0600 (CST)
Received: from mordor.rsn.hp.com (mordor.rsn.hp.com [15.99.150.126]) by palrel1.hp.com with ESMTP (8.7.5/8.7.3) id JAA21653; Mon, 16 Dec 1996 09:24:46 -0800 (PST)
Received: (from holder@localhost) by mordor.rsn.hp.com (8.7.1/8.7.1) id LAA00817; Mon, 16 Dec 1996 11:30:51 -0600 (CST)
To: turner@lanl.gov
Cc: xemacs-beta@xemacs.org
Subject: Re: 19.15-b4 bench results (Hanoi)
References: <199612160002.RAA28973@branagh.lanl.gov> 	<m2engrfgsj.fsf@altair.xemacs.org> 	<199612160501.WAA02208@branagh.lanl.gov> 	<199612160646.XAA02372@branagh.lanl.gov> 	<199612160657.XAA02385@branagh.lanl.gov> <199612161619.JAA03083@branagh.lanl.gov>
From: Shane Holder <holder@mordor.rsn.hp.com>
Date: 16 Dec 1996 11:30:50 -0600
In-Reply-To: John Turner's message of Mon, 16 Dec 1996 09:19:42 -0700
Message-ID: <fawn2vemm5x.fsf@mordor.rsn.hp.com>
Lines: 37
X-Mailer: Red Gnus v0.66/XEmacs 19.15

>>>>> "John" == John Turner <turner@xdiv.lanl.gov> writes:

  John> Following up to myself again:

  >> When I use a frame of the size I normally use (the full height of my
  >> screen, I believe 58 lines or so), it's slow.  Times like I showed
  >> before, over 10.  When I use a tiny little frame, just big enough to
  >> display the towers, it rocks.  Under 4 seconds.

  John> I was thinking that the size of the frame should affect other
  John> benchmarks as well, like the scrolling one, so I tested that.

I was wondering the same things my self, I just hadn't gotten around
to testing it yet, thanks for verifying what I thought.

  John> It does indeed matter, but the advantage is in the opposite
  John> direction for scrolling.  For a big frame it takes 75 seconds,
  John> for a small (10 lines or so) frame, it's 87.  Makes sense.

Yeah, that's to be expected if you don't set the scroll-step
variable.  When the cursor hits bottom, the window will scroll half
the screen, so the larger the window, the fewer redraw's of the
screen.  I think I'll probably set scroll-step to 1 for the bench
marks.

  John> I'm wondering if bench.el should pop up a frame of a
  John> predetermined size in which to run benchmarks for which the
  John> frame size would matter...

Yeah, that's another thing I've been meaning to put in.  I think that
will affect the frame display bench marks as well.

-- 
Shane Holder                                 e-mail: holder@rsn.hp.com
Hewlett Packard                               phone:     (214)497-4182
3000 Waterview                       Never underestimate the bandwidth
Richardson, TX 75083                 of a truck moving at 70 MPH.

