Deep Color increases the number of bits available for transmission for each channel. This means that there are more shades available for a TV to mix together. So, for example, a TV that accepts the new standard in 12-bit form can mix together any one of 4,096 shades (levels of brightness) of each primary color for 68.7 billion possible colors (4,096 red x 4,096 green x 4,096 blue = 68,719,476,736 colors). HDMI 1.2 could only transfer 8 bits per channel. So, there were only 256 shades of each color to choose from and fewer colors overall (256 x 256 x 256 = 16.7 million). These different shades help decrease artifacts (like color banding) and increase color fidelity. The visible picture-quality increase from 8 bits per channel to 10 or even 16 (in its highest 1.3 form) has been and is still being debated, but having the ability to transmit xvYCC and Deep Color sure can't hurt. Together, they mean that there will be more and better colors for future displays.
In order to make for a wider color gamut and a higher bit depth for even more realistic-looking displays (capable of creating a wider range of colors), every step in the chain needs to do that exact thing, as well. If the camera can only do Rec.709, it won't matter that your TV can do more than that, because that extra color isn't in the source (which is, uh, the situation we have now). If the camera can do xvYCC but the medium (say, HDTV broadcasts) can't, again, it won't matter that your TV can do it. In other words, for you to see the new colors, material will have to be shot, transferred, encoded, and mastered in xvYCC and Deep Color.
