Of course "World's Best" is rightly a matter of opinion rather than an objective fact, but in all my time in Headphonedom I've never seen a can so universally acknowledged as a world-class performer. When I visited Sennheiser four years ago, Axel Grell (their chief headphone engineer) was already hard at work developing these headphones. The iteration I saw then was just a 3D model on a computer and a whole bunch of data from their laser interferometer, but the work was imaginative, thorough, and compelling even early in the development process.
There is a lot to say about these cans, but what I'd really like to focus on for the moment is the driver in these headphones. A regular driver diaphragm has a dome in the middle; the voice coil about halfway to the edge; and a surround or suspension outside the voice coil which has a fixed attachment to the driver housing around its outside edge. The problem with this type of driver is that at very high frequencies it may stop acting like a pure piston moving only straight in and out, and may take on additional vibration modes so that the surface is wobbling or twisting as it moves in and out. This is called "cone breakup" in speakers. The traditional approach for solving this problem is to make the speaker cone, or driver diaphragm in this case, stiffer so it is less likely to begin vibrating on its own. This is where you find folks making aluminum speaker cones, or cones with ridges and various features to make it more rigid. Unfortunately this often leads to a heavier cone that it more difficult to accelerate, and therefore lowers the slew rate (and high frequency response) of the cone.
Recently another approach has begun to appear commercially called a "ring radiator." Both Scan-Speak and Vifa produce a version of this type of tweeter. In this case, the driver is not a simple circular driver, but a ring-shaped driver which is attached both at its outer circumference and at its inner edge. The diaphragm is driven by the voice coil which is attached behind the "V"-shaped groove halfway between its inner and outer edge.
This configuration dramatically increases the amount of structural support for the surface area of the driver and reduces to amount of surface area that can begin to take on its own vibrations. Voila! You have very much reduced the ability of the surface area that can suffer cone break up. Headphone drivers have a LOT more work to do than a tweeter. They may not need to be as loud, but they have to reproduce sound over the entire audio spectrum. As a result, there are surely significant differences between the tweeters shown above and the driver in the HD800, but the principles remain the same. I never got a chance to see it, but I'll bet you Axel Grell had a big grin on his face when he first stuck one of the new drivers into his laser interferometer and viewed the lack of cone break-up. So the new HD800 driver is clearer, cleaner, and faster sounding ... but wait, that's not all.
Because the new driver could become larger without breaking up, we also get two other sonic benefits: tighter bass, and better imaging. A larger driver will allow the diaphragm do displace more air for any given unit distance moved compared with a smaller driver. Moving more air means the driver can achieve better bass extension before it runs out of voice coil travel. It's pretty obvious when listening to these cans, they really have an astonishing sense of ease and control in the lows. Not overly big and bloated, just controlled right down to the point you feel your Adam's apple wiggling. The second advantage the new ring radiator has in the HD800 is improved imaging and sense of coherence. When sound approaches your head from a speaker or natural sound source it travels a substantial distance to reach your head, and therefore the wavefront of the sounds is fairly planar when it hits your ears.
Normally, with headphones, the source of the sound is very close to your ear and being radiated from a virtual point source (the dome of the driver). This produces a strongly curved (spherical section) wavefront. We are used to hearing flat wavefront sounds bounce against us and our outer ear before entering the ear canal to be heard. The nature of this chaos of reflections is well categorized by your head and is used strongly when you localize sound. When the curved wavefront from headphones bounces against your outer ear, the geometry is substantially different as is the resulting sound that enters your ear canal. These differences disrupt the psycho-acoustic cues your brain listens to in order to properly localize sound. The more planar wave of the HD800 allows more of the normal reflections at the outer ear to occur and allows the brain's normal ability to localize sound to be used. The result is better imaging. Note: I have to say that I am somewhat skeptical of this. Because the driver in headphones is so close to the ear I assume it is better to think of the physics of couplers and waveguides rather than free-space propagation of acoustic energy when considering headphone acoustics. I'm not at all convinced that there's a "wavefront propagating" in the volume of space between the driver in the normal, free-space sense when the half wavelength of the sound is much longer than the dimension of the enclosed space of a headphone for the most part. I am looking forward to the next time I get to talk with Axel and am able to ask him some questions about this. For more information on these great cans visit the Sennheiser HD800 page