PHP and .NET > Speed?

[Jeremy W.]

... we will be testing PHP in a real world scenario as well. To use the suggested rephrasing:

We'll test php against .NET with php on LAMP and .NET on .NET server.

As an aside, we'll also see how they both perform on Win2K.

[necrominator]

Originally posted by Jeremy W.
... we will be testing PHP in a real world scenario as well. To use the suggested rephrasing:

We'll test php against .NET with php on LAMP and .NET on .NET server.

As an aside, we'll also see how they both perform on Win2K.

Why not test php on linux, its what everyone uses and its faster, better comparison. Also rembmer one thing PHP IS FREE

[Jeremy W.]

Sorry, my last response was rude. LAMP stands for Linux, Apache, MySQL, PHP. So it will be run on Linux.

Also, .NET is free. Free langauge, free framework, free database, free IDE. The only things that do cost are the OS (if you're using Windows) and the hosting (if you're hosting).

[Andrew-J2000]

so what are the scenario's then?
A good idea might be to build a small scale ecommerce site, from an educational and performance point of view from both languages. I think this would turn out to be an excellent article, which could help people make their mind up about what language to choose (dont argue about it). Inspire techniques in oop / uses of xml and web services. What do you think?

I find the problem with xml is where and when to use it, many people post and ask about implementing xml/xhtml with a cms, but so far M@rco is the only person I know to pull it off. (not for long tho)

[firepages]

building an application which may require different techniques in different languages to achieve the same result could further skew/abstract the `results`.

I think the ideal test would be a box that dual booted (ok hot-swapped/whatever) with .NET and *NIX else there are going to be questionmarks hanging over any results.

For a change I will agree with Jeremy on the point that 'PHP is supposed to be x-platform so whats wrong with using anyMSOS for both tests' (he did not say that exactly but that was the gist that I got (And I still consider PHP X-platform))

However if we are at any point running .NET on .NET server which is optimised for such then PHP should be run on an accellerated *NIX platform on the same machine , unless both of those occur then then its pointless as the .NET vs PHP on say 2K will always have those questionmarks (on both sides (IMO anyway))

[Jeremy W.]

PHP will be run in it's own accelerated environment as well, it's only fair.

Personally I'd love to do a thorough application test, I really would, but I'm not sure how reasonable it is. Dunno, maybe if there were 2 developers in each camp we could come up with something decent...

J

[mgkimsal]

Originally posted by Jeremy W.
PHP will be run in it's own accelerated environment as well, it's only fair.

Personally I'd love to do a thorough application test, I really would, but I'm not sure how reasonable it is. Dunno, maybe if there were 2 developers in each camp we could come up with something decent...

J

Rather than just 2, it might be better to make it a bit more open. Define a list of specs - like the pet store.

"Build a system to do X,Y and Z".

Open this up to various teams who submit code for systems - some people can choose to work together, others alone. At the end of a specified period of time (1 week? 2 weeks? 3 days?) require all code to be submitted, then those who want their code benchmarked can do so.

Again, I say that machines don't necessarily have to be exact hardware - relative conclusions can be drawn in some cases. Especially if this is intended to be more of an educational project than a speed competition, this would be wortwhile. Yes, it's a bigger undertaking than was originally thought of, but planned properly, it would probably be a first of its kind.

[csn]

If using MySQL make sure to use InnoDB tables.