On Thu, Jan 15, 2009 at 04:17:51AM -0500, Robert Cummings wrote:

> On Wed, 2009-01-14 at 21:17 -0500, Paul M Foster wrote:
> > On Wed, Jan 14, 2009 at 01:39:02PM -0800, Daevid Vincent wrote:
> >
<snip>
> > >
> > >    The pages are significantly slower than straight PHP by orders of
> > >    magnitude: [1]http://paul-m-jones.com/?p=315
> >
> > What a great link! I've never seen this kind of comparison before. HTML
> > is 70% faster than straight PHP, and the frameworks (even codeigniter)
> > deliver less than 20% of the performance of straight PHP.
> 
> It's not a fair comparison. It's like saying here's a bucket of water. I
> want you to take it across the road using one of the following methods:

I wouldn't consider it a truly scientific comparison. The testing method
seems a little odd to me. Nonetheless, the point is makes is clear: PHP
is 70% (more or less) efficient in rendering pages than straight HTML,
and the "best" frameworks are only about 20% as efficient as straight
PHP. We can argue about the exact numbers, but the results make clear
that for speed HTML > PHP > frameworks. (And really, can you logically
argue that point?) From this, you don't draw the conclusion to not use
frameworks or PHP. From this, you now know one of the trade-offs in
using PHP and frameworks. And you get some idea of the magnitude of its
impact.

(These guys didn't even bother to test HTML with a bunch of Javascript
or complex CSS in it. Might PHP have been faster?)

Is *coding* faster and more efficient with frameworks? Sure. Does the
code execute as fast? No. If execution speed is your priority, then you
either scrap the framework, resort to a caching solution (which some of
the frameworks already have in place, but which the testers didn't
test), or figure something else out (like C?). If execution speed isn't
your priority, then you might look instead at a framework.

Anyway, the survey is just a tool which lets you know about one of the
trade-offs in web design. I doubt any other method of testing would skew
the results all that much.

Paul
-- 
Paul M. Foster

-- 
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php

Reply via email to