Techempower Benchmarks 2/18

I noticed the new benchmarks (round 15) were put out 2/14/2018 and the tests have Phalcon Micro screaming fast and Phalcon (non Micro) only slightly faster than Laravel.

https://www.techempower.com/benchmarks/#section=data-r15&hw=ph&test=query&l=hr9zpb

Can someone explain what is happening besides "this is a test that isn't real world so don't worry about", is there a better way for them to implement this?

What they did with Micro: https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/frameworks/PHP/phalcon-micro/public/index.php

What they did with Non Micro: https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/frameworks/PHP/phalcon/app/controllers/BenchController.php

probably we can improve that code to get better results

It's interesting why in some frameworks they use plain pdo, query builder or even orm...

For example in PhalconMicro they used plain PDO, while in the PhalconNonMicro they are using ORM... How can both even be compared?

Not a Zend or Laravel expert, but in the Zend example they are using some sort of simple query builder or whatever, while in Laravel they use ORM for the queries... Of course pdo/simple queries are way faster than orm's..

Well, techempower did a really poor job writing this benchmark... :D I wouldn't base any decision on it.

We know and love Phalcon, but many developers do not know and check these statistics to choose the new framework to work

Well, techempower did a really poor job writing this benchmark... :D I wouldn't base any decision on it.

edited Mar '18

Its not only about Phalcon, most frameworks are in bad position, probably because the person writting the code is not familiar with all the frameworks. Its like comparing apples to eggplants O_o

As Lajos said, poor job :)

We know and love Phalcon, but many developers do not know and check these statistics to choose the new framework to work

Well, techempower did a really poor job writing this benchmark... :D I wouldn't base any decision on it.

Well this is the site that got Laravel and Symfony in a pissing match, I tweeted to them to add Phalcon but it is not included (yet). Phalcon's number 1 feature - speed - should be highlighted properly . so yes, Benchmarks are not always great analysis, however Phalcon sells itself on performance.

http://www.phpbenchmarks.com/en/

Just a note form another post I created , they include the following line in their response in the non micro. This line also slows down my local benchmarks. Could be culprit. Switched to generic php header() command

$response->setStatusCode(200,  "OK");