Gecko [ v99win.today ]

Name	Size	Permission
images	[ DIR ]	drwxr-xr-x
wand	[ DIR ]	drwxr-xr-x
ChangeLog-2001.html	23.56 KB	-rw-r--r--
ChangeLog-2002.html	95.11 KB	-rw-r--r--
ChangeLog-2003.html	257.78 KB	-rw-r--r--
ChangeLog-2004.html	87.64 KB	-rw-r--r--
ChangeLog-2005.html	58.03 KB	-rw-r--r--
ChangeLog-2006.html	14.43 KB	-rw-r--r--
ChangeLog-2007.html	57.83 KB	-rw-r--r--
ChangeLog-2008.html	127.75 KB	-rw-r--r--
ChangeLog-2009.html	106.29 KB	-rw-r--r--
ChangeLog-2010.html	44.81 KB	-rw-r--r--
ChangeLog-2011.html	40.59 KB	-rw-r--r--
ChangeLog-2012.html	58.38 KB	-rw-r--r--
ChangeLog-2013.html	35.93 KB	-rw-r--r--
ChangeLog-2014.html	57.66 KB	-rw-r--r--
ChangeLog-2015.html	111.57 KB	-rw-r--r--
ChangeLog-2016.html	59.17 KB	-rw-r--r--
ChangeLog-2017.html	72.3 KB	-rw-r--r--
ChangeLog-2018.html	170.85 KB	-rw-r--r--
ChangeLog-2019.html	84.87 KB	-rw-r--r--
ChangeLog-2020.html	74.61 KB	-rw-r--r--
ChangeLog-2021.html	50.82 KB	-rw-r--r--
Changes.html	4.78 KB	-rw-r--r--
Copyright.html	15.27 KB	-rw-r--r--
FAQ.html	47.41 KB	-rw-r--r--
GraphicsMagick.html	376.79 KB	-rw-r--r--
Hg.html	15.76 KB	-rw-r--r--
INSTALL-unix.html	60.78 KB	-rw-r--r--
INSTALL-windows.html	40.33 KB	-rw-r--r--
ImageMagickObject.html	7.17 KB	-rw-r--r--
NEWS.html	211.05 KB	-rw-r--r--
OpenMP.html	12.77 KB	-rw-r--r--
README.html	22.68 KB	-rw-r--r--
animate.html	41.85 KB	-rw-r--r--
authors.html	8.6 KB	-rw-r--r--
batch.html	10.6 KB	-rw-r--r--
benchmark.html	11.78 KB	-rw-r--r--
benchmarks.html	7.1 KB	-rw-r--r--
bugs.html	2.89 KB	-rw-r--r--
color.html	30.19 KB	-rw-r--r--
compare.html	19.27 KB	-rw-r--r--
composite.html	39.9 KB	-rw-r--r--
configure-target-setup.png	7.21 KB	-rw-r--r--
conjure.html	15.4 KB	-rw-r--r--
contribute.html	6.85 KB	-rw-r--r--
convert.html	90.16 KB	-rw-r--r--
display.html	140.58 KB	-rw-r--r--
docutils-api.css	8.56 KB	-rw-r--r--
docutils-articles.css	9.95 KB	-rw-r--r--
download.html	7.25 KB	-rw-r--r--
favicon.ico	922 B	-rw-r--r--
formats.html	41.29 KB	-rw-r--r--
gm.html	891.41 KB	-rw-r--r--
identify.html	13.4 KB	-rw-r--r--
import.html	30.34 KB	-rw-r--r--
index.html	9.75 KB	-rw-r--r--
links.html	14.44 KB	-rw-r--r--
magick.css	2.36 KB	-rw-r--r--
miff.html	17.44 KB	-rw-r--r--
mission.html	3.32 KB	-rw-r--r--
mogrify.html	81.71 KB	-rw-r--r--
montage.html	54.05 KB	-rw-r--r--
motion-picture.html	32.02 KB	-rw-r--r--
perl.html	54.33 KB	-rw-r--r--
process.html	5.97 KB	-rw-r--r--
programming.html	5.74 KB	-rw-r--r--
project.html	3.82 KB	-rw-r--r--
quantize.html	11.81 KB	-rw-r--r--
reference.html	2.42 KB	-rw-r--r--
security.html	19.6 KB	-rw-r--r--
smile.c.gz	982 B	-rw-r--r--
thanks.html	6.07 KB	-rw-r--r--
time.html	4.77 KB	-rw-r--r--
tools.html	11.6 KB	-rw-r--r--
utilities.html	4.99 KB	-rw-r--r--
version.html	5.98 KB	-rw-r--r--

Code Editor : benchmarks.html

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="Docutils 0.16: http://docutils.sourceforge.net/" />
<title>GraphicsMagick vs ImageMagick Benchmarks</title>
<link rel="stylesheet" href="docutils-articles.css" type="text/css" />
</head>
<body>

<div class="banner">
<img src="images/gm-107x76.png" alt="GraphicMagick logo" width="107" height="76" />
<span class="title">GraphicsMagick</span>
<form action="http://www.google.com/search">
  <input type="hidden" name="domains" value="www.graphicsmagick.org" />
  <input type="hidden" name="sitesearch" value="www.graphicsmagick.org" />
<span class="nowrap"><input type="text" name="q" size="25" maxlength="255" />&nbsp;<input type="submit" name="sa" value="Search" /></span>
</form>
</div>

<div class="navmenu">
<ul>
  <li><a href="index.html">Home</a></li>
  <li><a href="project.html">Project</a></li>
  <li><a href="download.html">Download</a></li>
  <li><a href="README.html">Install</a></li>
  <li><a href="Hg.html">Source</a></li>
  <li><a href="NEWS.html">News</a> </li>
  <li><a href="utilities.html">Utilities</a></li>
  <li><a href="programming.html">Programming</a></li>
  <li><a href="reference.html">Reference</a></li>
</ul>
</div>
<div class="document" id="graphicsmagick-vs-imagemagick-benchmarks">
<h1 class="title">GraphicsMagick vs ImageMagick Benchmarks</h1>

<div class="contents local topic" id="contents">
<ul class="simple">
<li><a class="reference internal" href="#foreword" id="id1">Foreword</a></li>
<li><a class="reference internal" href="#strategy" id="id2">Strategy</a></li>
<li><a class="reference internal" href="#the-benchmark" id="id3">The Benchmark</a></li>
</ul>
</div>
<div class="section" id="foreword">
<h1><a class="toc-backref" href="#id1">Foreword</a></h1>
<p>This page used to document benchmark results which compared
GraphicsMagick 1.3.8 performance to ImageMagick 6.5.8-10 performance.
We found that GraphicsMagick was usually considerably faster at
executing image processing operations from the command line than
ImageMagick 6.5.8-10 was. One ImageMagick algorithm ran as much as 770
times slower.  GraphicsMagick clearly ran much more efficiently under
Microsoft Windows.</p>
<p>We now find that a simple head to head performance comparison between
GraphicsMagick and recent ImageMagick is no longer possible because
ImageMagick has changed the meanings of the existing arguments.  In
particular, we discovered that ImageMagick filter arguments producing
convolution matrices are now producing lower-order convolution
matrices representing perhaps an order of magnitude less work given
the same arguments.  The resulting images are visually substantially
less filtered.  Using the same command-line filter arguments causes
GraphicsMagick to appear slower when it is actually doing far more
work than ImageMagick.</p>
<p>Due to it not being feasable to do a head to head performance
measurement between GraphicsMagick and ImageMagick, we have decided to
not post results at the moment.  However, the strategy and a simple
benchmark driver script are still provided for those who want to do
their own performance comparisons between GraphicsMagick and
ImageMagick.</p>
</div>
<div class="section" id="strategy">
<h1><a class="toc-backref" href="#id2">Strategy</a></h1>
<p>The benchmark focuses on the ability to process many medium sized
HD-resolution (1920x1080 pixels) images. In order to prevent disk I/O
from being a factor, we use a small input image and tile it to create
a larger input image via the &quot;tile:&quot; coder. The processed image is
sent to the &quot;null:&quot; coder so that file writes to a slow disk are also
not a factor.  Static executables are used and executed via full paths
in order to minimize variability from the execution environment. In
order to obtain accurate and useful timing, we use the bash shell to
execute the command 40 times and see how long it takes. This is a very
simple benchmark approach which is quite representative of the
performance that the typical user observes.</p>
<p>This new benchmark reveals the following performance criteria:</p>
<blockquote>
<ul class="simple">
<li>Basic execution overhead of the software.</li>
<li>Image processing algorithmic efficiency.</li>
<li>Efficacy of OpenMP enhancements.</li>
</ul>
</blockquote>
</div>
<div class="section" id="the-benchmark">
<h1><a class="toc-backref" href="#id3">The Benchmark</a></h1>
<p>The benchmark is quite simple. It reads a list of commands to execute
from a file named &quot;commands.txt&quot; and times how long it takes to
execute each command 40 times using GraphicsMagick and ImageMagick.</p>
<p>Here is the simple benchmark script:</p>
<pre class="literal-block">
#!/usr/bin/env bash
#
# Measure the performance between two 'convert' commands by
# executing a subcommand through many iterations and seeing
# the total time that it takes.
#
# Written by Bob Friesenhahn, October 2008
#

# GraphicsMagick
convert1='/usr/local/bin/gm convert'
#convert1='/c/Program\ Files/GraphicsMagick-1.3.8-Q16/gm.exe convert'

# ImageMagick
convert2='/usr/local/bin/convert'
#convert2='/c/Program\ Files/ImageMagick-6.5.9-Q16/convert.exe'

# Input file specification
input_image='-size 1920x1080 tile:model.pnm'

# Ouput file specification
output_image=&quot;null:&quot;

# Should not need to change any of the rest
typeset -i iterations=40
echo &quot;Convert-1:   ${convert1}&quot;
echo &quot;Version:     `eval &quot;${convert1}&quot; -version | head -1`&quot;
echo &quot;Convert-2:   ${convert2}&quot;
echo &quot;Version:     `eval &quot;${convert2}&quot; -version | head -1`&quot;
echo &quot;Date:        `date`&quot;
echo &quot;Host:        `uname -n`&quot;
echo &quot;OS:          `uname -s`&quot;
echo &quot;Release:     `uname -r`&quot;
echo &quot;Arch:        `uname -p`&quot;
echo &quot;Input File:  ${input_image}&quot;
echo &quot;Output File: ${output_image}&quot;
echo &quot;Threads:     ${OMP_NUM_THREADS:-1}&quot;
echo &quot;Iterations:  ${iterations}&quot;
echo &quot;========================================================================================&quot;
echo
typeset -i count=0 i=0
cat commands.txt | while read subcommand
do
  echo ${subcommand}

command1=&quot;${convert1} ${input_image} ${subcommand} ${output_image}&quot;
  i=0
  count=$iterations
  time while ((i &lt; count))
  do
    eval &quot;${command1}&quot;
    let i=i+1
  done
  sleep 1

command2=&quot;${convert2} ${input_image} ${subcommand} ${output_image}&quot;
  i=0
  count=$iterations
  time while ((i &lt; count))
  do
    eval &quot;${command2}&quot;
    let i=i+1
  done

echo
  sleep 1
done 2&gt;&amp;1
</pre>
</div>
</div>

${this.title}

Code Editor : benchmarks.html