I have not attempted to think through those details before. But...

I have not attempted to think through those details before. But I think this is more complex than a simple issue of wall (90deg) vs roof (0deg). Sun angle varies during the day. So the optimal angle I'd imagine would be like a weighted-average based on (a) sun-angle vs time graph for the geographical location in question, and (b) the solar system efficiently vs sun-angle graph.

I found an online calculator which for instance provides the following Optimal Angles for Sydney:




Certainly roofs (in particular tilted ones) are of advantage. But it's not like walls are totally disadvantaged.

The second issue is the system itself which as I understand works by refracting solar rays to the edges of the glass. This is totally different from regular solar panels. So I guess the optimal angle can also be totally different (point b above). And without that information it is perhaps too difficult to get a conclusion. Perhaps we can ask CPV to provide some info on the optimal angle for the operation of the glass (or something like an power output vs angle graph which I'd imagine they've definitely examined).

When it's comes to the economy of the system, there is also the issue of available space. Certainly a lot more surface area on the walls which can offset angle disadvantages (if any). My view is that roof-top panels and CPV solar glass are two different products intended for different applications. E.g. CPV product comes handy for tall buildings where roof area is limited, whereas roof-top panels better suite ordinary 1-2 story  residential buildings. So it's perhaps not one or the other. The two systems seem to compliment each other.