Agreed about myth perpetuation, Nick. The latest Mazda offering almost looks competetive in fuel efficiency terms - it gets ~24mpg (pretty lousy considering i'm used to 55mpg
). Like all design problems, there is a tradeoff in deciding whether the engine is aimed at more power or efficiency. In the rotary this comes in the form of rotary piston width.
Ideally the rotary piston should have the same width as tip-to-tip distance. This then approaches an "oversquare" (bore/stroke>1) piston engine for surface_area/volume. The problem then becomes breathing (ie power), since a gasoline rotary needs the ports in the sidewall to avoid the "valve" overlap that causes unburnt fuel emissions. Direct inject allows cylinder wall ports, but only diesels actually inject the fuel near or during combustion. The maximum compression ratio is too low for diesel combustion though.
Put in simple terms every engine design is a compromise for the given requirements and constraints. The rotary engines biggest area of compromise is the fluid dynamics during combustion. For this reason it is a concept which will always be there, but will never really be competative. For my money a good piston layout (eg radial for aircraft) beats rotary every time, and turbos are the way to boost power.
Mart