I want to know, how it was possible to create the cubes, with their shinny, dark and cubed parts? I know it could be doable with Photoshop, but how? Is there any software better suited to make it than that?
To expand on Marc's answer I dug into the source code of the page.
div tag in the html like such:
<div id="runner"> <img class="parallax-layer" src="http://mortenstrid.no/wp-content/themes/mortenstrid/images/runner_layer_bottom.png" alt="" style="width:1347px; height:642px;"/> <img class="parallax-layer" src="http://mortenstrid.no/wp-content/themes/mortenstrid/images/runner_layer_middle.png" alt="" style="width:1307px; height:623px;"/> <img class="parallax-layer" src="http://mortenstrid.no/wp-content/themes/mortenstrid/images/runner_layer_top.png" alt="" style="width:1266px; height:603px;"/> </div>`
That means there's a script, or numerous scripts that interact to move the images around the page. Off the top of my head I'd say, as I mentioned in my comment, that the images rely on the cursor's X and Y coordinates and adjust themselves according to the coordinates.
At the top of the source code we find a number of different scripts. One of them is called "jparallax" which traces back to this website, where you can find numerous demos.
From the web site's about page:
I'm first and foremost an Interactive Visual Designer, but I also possess an high skill level in Programming, Motion Graphics, Video and 3D Animation.
Web: HTML/CSS, Flash/AS3, Jquery/AJAX, PHP/MySQL, Wordpress, Tumblr, Dreamweaver.
Design and Motion Graphics: Photoshop, Illustrator, After Effects, Premiere, Mocha.
3D: 3DS Max, V-Ray, SynthEyes.
Given the way the image looks, my guess is that it was modelled in 3D Studio Max, rendered using V-Ray, possibly with some post production in Photoshop (it looks like the parallax effect was created by using 3 separate images, so a little tweaking in Photoshop may have been needed).
I think this must have been created with Cinema 4D.