Recently, a new consortium of consumer electronics companies calling themselves WirelessHD have announced plans to stream uncompressed HD video using 60GHz RF. A new startup company called SiBeam (which happens to live next door to Tzero) is launching this effort. SiBeam is founded by UC Berkeley Prof. Bob Brodersen (who is Chairman of the Board) and some other Berkeley guys. The potential benefits of transmitting at 60 GHz are the wide available bandwidth and the practically unlimited transmit power. However, in practice, neither helps.
First, the baseband processing will be just as limited for 60 GHz radios as for any other technology making the usable modulation bandwidth more or less the same as what UWB in the 3 – 5 GHz spectrum currently achieves (for CE price points). So, to achieve multi-gigabit data rates they will probably have to use higher order modulation schemes. This means they will need better SNR (signal-to-noise ratio).
So how to achieve better SNR? Well, first, you have to recognize that the radio links at 60 GHz suffer about 24 dB more free-space path loss than radios operating in the 3 – 5 GHz spectrum. So the first thing they have to do is make up for that extra loss. Prof. Brodersen’s group at Berkeley hasworked on building 60 GHz electronically steered phased array antennas onto CMOS chips. It’s natural to suspect that this is some of SiBeam’s secret sauce. Suppose you could get +12dBi gain doing this (questionable). Then having this gain on both the transmitter and receiver would achieve a total link budget gain of 24dB which makes up for the 60 GHz free-space path loss penalty. Even if this is possible the consequence is highly directed antenna beams which will not tolerate any kind of blockage.
One of the great benefits of a 3 – 5GHz UWB system, which is readily seen in Tzero demos, is that we take advantage of scattering. So you can block our antennas and the system keeps working just fine. Do that with a 60 GHz link with highly directional antenna beams and the link is irrevocably lost. For the sake of furthering the discussion, assume that the additional free-space path loss is compensated for with high gain phased array antennas and that the direct line-of-sight link between the transmitter and receiver is clear.
How do they now get the extra transmit power to achieve the higher SNR needed for the higher-order modulation? In theory, the transmit power is not really restricted (I think you can transmit 40dBm EiRP at 60 GHz). However, in practice, amplifying signals at 60 GHz is extremely inefficient. State-of-the-art 60GHz amplifiers have terrible power efficiency, like 1%! Furthermore, these devices are made out of Gallium Arsenide PHEMT technology (or possibly SiGe but with very low output power), not CMOS technology. So, in all likelihood, rather than being able to transmit more power at 60 GHz than UWB transmitters are able to under FCC restrictions, they will end up transmitting considerably less. I’m sure these issues have all been thought of so it will be interesting to see what technical resolution of these difficulties SiBeam and others will come up with.
Techdigs is a great new site focused -- according to its banner -- on "demystifying your entertainment tech." They recently published an overview of UWB that's very well done, providing insights into the current state of the market as well as the consumer offerings that will soon be available. Visit Techdigs to see the full article. 
