The trouble with your example is that a car is not a dyson sphere. A car is a few square meters, which means that with a 30% efficient pv panel (double the eff

The trouble with your example is that a car is not a dyson sphere. A car is a few square meters, which means that with a 30% efficient pv panel (double the efficiency of the current standard), you'd only be getting around 15hp at noon. That's enough to toddle around a closed track at 30mph in a carbon fiber frame built by a university, but not enough to outdo a bicycle in real-world applications. Fixed solar power plants are more promising, but they're not perfect--Sure, a square 100 miles on a side in Arizona could supply a lot of power, but not to New York or Seattle, even over the best HVDC lines. And 1,000 square miles is a helluva lot of monocrystalline silicon. Solar thermal looks more promising, especially for keeping the power going at night, but it's far from a panacea, and there's lots of companies working on it. I've invested in a few of them.