At the End of 2012 I had my first contact with a C-based PHP frameworks, namely YAF. Coincidently, some day afterwards Bruno from phpmaster.com pointed me towards Phalcon - a more modern interpretation of the same idea. So I was hooked.

In his "good idea" category he notes that it's faster because it's already loaded in on the request (no long list of includes) and the memory footprint is less than a PHP equivalent. The "bad" side of things mentions some pretty major hurdles though, including the small communities vs larger ones on PHP-based frameworks and the issues that could come with debugging/upgrading.

In recent months, a number of new frameworks have cropped up for PHP as PECL extensions (Including YAF and PhalconPHP). They promise to provide huge performance gains and lower resource usage for PHP applications. On the surface, they appear to be incredible tools for improving development. But they aren't all they are cracked up to be. In fact, I would argue that they are actually not necessary at all.

He breaks the arguments down into two sections - what you gain by having the framework based in an extension (like performance) and some of the things you give up (like readability, portability and maintainability).

My argument here would be that if you have a site where you can measure meaningful money savings by putting the framework into C (with taking the additional maintenance costs into account), you likely shouldn't be using a framework anyway.

Why care about another PHP framework? especially one that's actually called Yet Another Framework? Because Yaf isn't just a framework. It's a PHP extension (listed on PECL), so it's written in C -- so it's very, very fast. How fast? Here's a benchmark (requests/sec).

The only framework (of the list he compared) that came in faster was MicroMVC and that's not as full-featured as the Yaf functionality is. Disregarding that, it has a huge requests/second difference between even the next on the list, CodeIgniter.