I've been thinking a bit more about how digital sound...

I've been thinking a bit more about how digital sound reconstruction (DSR) makes AKP's speakers louder than conventional ones for the same size. A few months ago there was a discussion about this and I figured a separate thread might be good for those interested. Thanks to everyone who contributed at the time, I think I understand the basics but if the following doesn't sound right then post away. So why are they louder?

One key reason is that each pixel generates an extremely quick pulse of compressed air (or decompressed air when the pixel is instructed to move in the opposite direction). Note that a pulse from a pixel isn't sound we hear. It's the subsequent, sequential movement of other pixels (de)compressing more air in combination (i.e. significant SPL) that constructs the sound wave we hear. This differs to a conventional speaker driver which must always be moving at the speed, excursion and direction that matches the desired sound frequency and volume.

The other key reason is because the "conventional formula for computing pressure produced by a vibrating piston in an infinite baffle" doesn't appear to be accurate for such small pixels. According to measurements taken from AKP's experiments the pressure is actually 10x greater than what is calculated using the formula, and hence the surprising ability for AKP's speaker to produce the SPL it does for its size.

With respect to sound quality, the quickness at which each pixel can pulse and the fact there are so many of them is one of the main factors enabling the constructed wave to be so acoustically accurate. As each microsecond passes, the ASIC is controlling exactly which pixels should pulse (both up and down) to construct the exact sound frequency required at precisely that time.

Final note, adding more pixels in speaker = even wider dynamic range, frequency range, and less distortion.