Go to main content

Textpattern CMS support forum

You are not logged in. Register | Login | Help

#1 2009-06-24 13:03:42

janvi
Member
From: Norway
Registered: 2009-05-28
Posts: 41

Benchmarks?

Hi .

Are there any new benchmarks for TextPattern compared to other cms?
Thinking about page load and server load.


Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge.

The Future is Open – GNU/Linux

Offline

#2 2009-06-24 13:52:45

Gocom
Developer Emeritus
From: Helsinki, Finland
Registered: 2006-07-14
Posts: 4,533
Website

Re: Benchmarks?

Not atleast with current releases. Also, because Textpattern is completely empty canvas, it’s really hard to compare it to anything, plus not that comparing PHP management script is always hard.

But where it comes to the performance, Textpattern is one of the lightweightest CMSes out there, because of the empty canvas thingy. As good is your code, as fast it is. Textpattern doesn’t leak to the memory, uses very little of it, db sizes are usually small, Textile is saved as parsed and causes minimal queries by default. And doesn’t force extra output of anykind.

Lets think about it. How could we compare PHP scripts to each other? Joomla? WP? Drupal? EE? TXP? No way. All are pretty awesomely written, are fast and the load variates from the end users input. All are just barebones, used to insert data to db. Output is up to the front end code, made by the user.

You could ofcourse make Textpattern the slowest CMS out there by using 500 plugins, doing useless queries and forcefully making it to leak. Also note that some of the plugins are very heavy, for example most of the onload parsers, external services, taggers, pagers, multi-freaking-professional-query thingys and so on.

Last edited by Gocom (2009-06-24 13:58:44)

Offline

#3 2009-06-24 14:03:21

hcgtv
Plugin Author
From: Key Largo, Florida
Registered: 2005-11-29
Posts: 2,722
Website

Re: Benchmarks?

Sencer did a Wordpress vs. Textpattern benchmark.

Also, do a search on this forum, we’ve had benchmark discussions in the past.

Offline

#4 2009-06-24 14:17:12

janvi
Member
From: Norway
Registered: 2009-05-28
Posts: 41

Re: Benchmarks?

Thanx, Quite interesting this btw.


Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge.

The Future is Open – GNU/Linux

Offline

#5 2009-06-24 14:19:12

juanjonavarro
Plugin Author
From: Valencia, Spain
Registered: 2005-05-16
Posts: 485
Website

Re: Benchmarks?

(bigger is better)

This is with default installations, of course it depends on the site complexity.

Sources (in Spanish):

Offline

#6 2009-06-24 15:37:04

hcgtv
Plugin Author
From: Key Largo, Florida
Registered: 2005-11-29
Posts: 2,722
Website

Re: Benchmarks?

Juan,

I came to Textpattern from NucleusCMS, interesting to see it kick butt in your first graph. When I initially decided on using NucleusCMS for my websites, I was running a lowly Pentium Pro for my web server platform. Back then, early 2004, there weren’t many CMS/Blog platforms like there are today. But of those I tested, I figured if they ran well on low-end equipment, then my sites would be fine on a better hosting platform when the time came to move them up.

When I decided to move away from NucleusCMS, Textpattern came out on top in my evaluations. I was still running my sites locally, so I could run all kinds of benchmarks to determine the leanest/fastest system out there. There was one system that beat out Textpattern, that was sNews, but you could hardly call it a CMS back then.

But as Gocom says, you can make any system run slowly or use up precious server resources if you’re not careful. Plugins are great, but if you need 30 plugins to make Textpattern be what you want it to be, then maybe you should be looking elsewhere for a system that has those features in the core.

Offline

#7 2009-06-24 17:12:42

ruud
Developer Emeritus
From: a galaxy far far away
Registered: 2006-06-04
Posts: 5,068
Website

Re: Benchmarks?

I suspect the benchmark outcome also depends a lot on the way the server is setup:
  • mod_php / cgi + PHP / fast-cgi + PHP
  • is an PHP opcode cache like APC or Xcache used.

The fact that a CMS A is faster than CMS B on mod_php, doesn’t necessarily mean it’s also faster on a setup that uses an opcode cache. If you use a lot of plugins, I suspect the eval-method used by TXP for loading plugins is relatively slow when using an opcode cache.

Also note that all the benchmarks shown above were done in 2005 and 2007.
Things have changed since then.

Offline

Board footer

Powered by FluxBB